JP3916395B2 - Test system with sample pretreatment function - Google Patents

Description

[0001]
1. Field of Invention
The present invention relates to a specimen or specimen inspection system, and more particularly, to a systematic method in which an operator manually examines a large number of individual specimens and classifies them into individual subsets such as “suspicious”, abnormal or abnormal specimens. . As used herein, the term “specimen” is not limited to medical or biological specimens, but can be extended to a whole group of individual specimens or parts. The present invention can be used, for example, for examination of histological specimens (ie based on tissue-incision as in anatomy or surgical pathology), cytological specimens (ie cytologists, cytopathologists, cytogenetics, hematologists). Cell samples (such as samples prepared from body fluids, excreted urine, saliva, and samples taken from gynecological organs) analyzed by neuroscientists, microbiologists, cell biologists, etc. It is particularly used in various situations, such as inspection of silicon wafers in the manufacturing process of integrated electronic circuits in the semiconductor industry, and inspection of other materials.
[0002]
2. Explanation of related technology
In a typical scenario, the examiner must manually examine and analyze a large number of samples every day to determine whether the sample is deviated from a predetermined normal range. Once abnormal specimens are identified, more detailed investigation and analysis is required. Subsequent more detailed investigations will require investigators with special expertise, such as pathologists in the cervical Papanicolaou trial. For example, in the normal case of a sample from a woman who has no subjective symptoms, most specimens are considered “normal” or “in normal range” (“WNL”), so additional scrutiny is required. There is no. However, due to the detail and breadth of this examination and analysis, the additional scrutiny is unfortunately very time consuming, laborious and error prone and can be an expensive method.
[0003]
Illustratively, the present invention will be described in the context of cytological specimen analysis, such as cervical smear (hereinafter referred to as cervical smear) or Papanicolaou test analysis. Cervical smears routinely collected from women facilitate the detection of precancerous changes and / or early stages of cancer, and the prognosis of the patient is worsening or progressing in the clinical stage of the cancer and related abnormal conditions Reduce opportunities for negative impact. A cervical smear is first prepared by collecting tissue from the vagina, cervix, and internal cervix from the patient. The sample is then fixed to the slide, for example by alcohol fixation, and appropriately stained (in a manner similar to the Papanicolaou (Pap) method) to allow microscopic visualization and analysis. FIG. 9 shows a cervical smear according to the conventional method.
[0004]
However, conventional cervical smear methods result in a large number of cells (typically 50,000 to 300,000 cells per cervical smear), thus making inflammatory or other difficult to make accurate and sensitive diagnosis difficult Sometimes it is obscured by the substance, and there is a limit. Since the cervical smear method deposits cells spatially inhomogeneously, it is difficult to analyze visually and consumes time as shown in FIG.
[0005]
Instead, several single layer sample or liquid-base formulation (LBP) methods have been developed. One method relies on centrifugation to separate cells prior to deposition on, for example, a glass microscope slide. Another method relies on physical filtration of the specimen through a filter paper having a controlled pore size distribution. An example of the latter exploration technique is the Cytyc ThinPrep® device and method.
[0006]
Cytyc's ThinPrep (R) sample preparation generally consists of filtering the sample with filter paper that is smaller than the average sample cell diameter, but with enough pores to pass through cellular debris and other debris. Is done. In one study-target embodiment, the LBP device has a pore size of 40 to 50 μm. Desirably, the filter paper does not have a rough edge or other aspect that destroys the sample cells. During filtration, the sample cells remain on the surface of the filter paper, but the debris passes through, resulting in the sample cells being enriched on the filter paper surface and the debris being drastically reduced. The filter paper surface is then pressed onto the microscope slide, and sample cells are transferred from the filter paper surface to the slide. This results in a slide with a more homogeneous distribution of cells (i.e. a single layer), and by using a filter paper with a smaller area than the slide, the concentration per unit area of the sample cells is increased, and the cellular events by the cytotechnologist There is virtually no cellular or other debris that interferes with detection and slide analysis. Cytyc's ThinPrep® sample preparation method thus provides a more spatially uniform cell distribution within the cellular disc (CD), for example as shown in FIG.
[0007]
In practice, once a sample is prepared from a specimen, the slide is then used by advanced technicians (“cytologists” for the purpose of identifying possible cell abnormalities in the specimen and determining the validity of the specimen. Selected by “). The cytotechnologist issues a record for each specimen considered to have cellular abnormalities. The cytotechnologist then submits the specimen slide along with his or her knowledge paper or electronic record to a skilled cytopathologist (i.e. a specialist physician) for further examination and final specimen diagnosis. To do.
[0008]
To sort out cervical smear specimens, cytotechnologists typically look at cervical smear slides containing cervical smears with a conventional optical microscope to detect the presence of potentially rare cancer cells or other abnormal conditions. To do. However, since cancer cells may appear only in the thousands of other normal-looking specimens, cytologists generally do not want to make a valid (i.e. accurate) decision on the entire slide. The area must be inspected. Missing a certain area may possibly lead to a false negative (FN) diagnosis. In addition, many parts of the specimen slide may contain no cells at all, but the cytotechnologist determines that these areas are also at least free of substances that are relevant to the heart (i.e., diagnostically meaningful). Must be inspected for. Of course, the process of completely selecting specimens for confirming the presence of cancerous or abnormal cells is difficult, tedious, prone to mistakes, and expensive. Nevertheless, cytologists examine more than 20,000 slides annually, classify specimens as being within normal range, identify abnormalities, and allow pathologists to diagnose cervical smear specimens. It is known. In many cases, the testing speed of this specimen is strongly influenced in part by financial concerns such as competition based on the number of specimens analyzed. The dilemma faced by laboratories is that the sample processing rate needs to be decreased to reduce the rate of false negative negligence while at the same time increasing the sample processing rate for economic reasons.
[0009]
Based on the first examination (i.e., screening) of the specimen by the cytotechnologist, the cytotechnician is in a state where the specimen is unsatisfactory, satisfactory but limited, or satisfactory for analysis To decide. If satisfactory, the cytotechnologist then determines whether the specimen contains suspicious material, such as precancer or cancer cells, or whether the specimen is apparently within the normal range. In general, statistically speaking, “suspicious” and “abnormal” specimens will account for about 5% to 10% of cervical smears in laboratories screening women without subjective symptoms in the United States. The remaining statistically 95% to 90% are apparently classified as normal by a cytotechnologist.
[0010]
If a specimen contains one out of ten or hundreds of thousands of well-conserved and well-stained cancer cells, the cytologist will confirm that the specimen is suspect, atypical, or abnormal. You should find something. Failure to correctly identify a specimen as anomalous in this sorting process is catastrophic because it can leave the cancer undetected and untreated, ultimately resulting in patient death. It is.
[0011]
Cytologists are required to pathology all "suspicious", "atypical", or "abnormal" specimens to "sign and submit", taking into account the detailed examination and final diagnosis and cytotechnologist's records and findings Turn to a scholar. One of the pathologist's ultimate goals is to analyze the sample in question and determine whether the sample contains cancer or precancerous cells based on advanced medical knowledge. In doing so, pathologists must strive to minimize false negative diagnoses and false positive diagnoses, which leave false negative diagnoses undetected for cancer, while false positive diagnoses are Because it may result in unnecessary or inappropriate, harmful and expensive surgery, chemotherapy or similar cancer treatments.
[0012]
Many of the specimens that cytotechnologists determine are “apparently normal” are classified as “in normal range” (WNL), and analysis of these specimens is complete. However, in order to minimize the possibility of false negative diagnoses in the sorting process and to identify cytotechnologists with poor sorting quality performance, at least some of these specimens are second sorted by the cytotechnologist. Or it is necessary to "re-sort". In the United States, at least 10% of these “apparently normal” specimens are randomly selected and re-submitted for quality assurance by another cytologist (according to a well-known, high-priority risk rate). It is obliged to be screened.
[0013]
In addition, in order to further minimize false negative diagnoses in the cervical smear sorting process, the cytotechnologist must spend sufficient time to sort each specimen slide. For this reason, in some states in the United States, legislation restricts individual cytologists from selecting more than 100 slides of cervical smears per day. Other states have more stringent limits, such as 80 slides per day. Assuming an average of 7 hours a day, this regulation would allow a representative cytologist to select and classify an average cervical smear 1 slide within 4.20 to 5.25 minutes.
[0014]
Despite these maximum limits, the average of cervical smear slides selected per day by cytotechnologists in the United States is 50-60, and a cytotechnologist to carefully determine whether abnormal cells are present or not. Corresponds to spending typically 7 to 8 minutes on each slide inspection. Of course, the cytotechnologist takes more time to sort each slide, so in theory it would be less likely to make a false negative error. At the same time, however, since the cytotechnologist takes more time to select each slide, fewer slides are selected per day, resulting in increased labor and cost for sample selection. This is a painful scenario for laboratories in the United States, for example, because the third-party health insurance reimbursement rate is generally very low, and many if not most clinical diagnoses. For the pathology laboratory, the Papanicolaou test has little or no benefit.
[0015]
Therefore, it minimizes the presence of false negative sample classification errors, while the analysis of the sample is accurate and useful for suspicious, atypical, or abnormal samples for the reference of diagnostic experts. There is a need for a more efficient specimen sorting system that reduces the time required to collect information.
[0016]
3. Integration by related patents
The following US patent applications are specifically incorporated into this patent as related patents:
[0017]
(i) US patent application Ser. No. 08 / 529,220, filing date September 15, 1995;
(ii) U.S. Patent Application No. 08 / 529,198, filed September 15, 1995; and
(iii) U.S. Patent Application No. 08 / 736,790 Filing Date Oct. 25, 1996
Summary of invention
The present invention provides a comprehensive system for improving efficiency in specimen testing and analysis. The invention described herein is for purposes of illustration and not limitation, and is a clinical diagnostic cytology system. In one aspect, the invention examines a plurality of samples each prepared from a cytological specimen, and the specimen is apparently within a normal range (“apparent-WNL”) or apparently not within a normal range (“ It helps to classify it as either apparently-non-WNL "). The present invention includes, for example, the following functions:
(1) Capture a set of digital images of the sample area into the machine;
(2) The machine analyzes the digital image of the area;
(3) identify cellular objects within a subset of the region;
(4) The machine evaluates whether the sample is in a satisfactory (eg sufficient) state to proceed with classification;
(5) the machine calculates the probability of cell atypia for the cellular objects identified in the sample;
(6) The machine orders and organizes the cellular objects in the sample according to the probabilities calculated for each cell variant, and the first subset of cellular objects in the sample has the highest probability of calculating the cell variant in the sample. Defining cellular objects having;
(7) the machine obtains a calculated probability of the analyte atypia of the corresponding specimen based at least in part on the calculated probability of the cell atypia of the cellular object in the sample;
(8) If the calculated probability of specimen atypia falls within a predetermined probability range, the machine will apparently interpret the corresponding specimen as -WNL;
(9) If the calculated probability of specimen atypia does not fall within the predetermined probability range, the machine will apparently interpret the corresponding specimen as non-WNL;
(10) When the machine interprets the corresponding specimen as apparently -non-WNL, (i) gives prior knowledge about a series of samples for the observer's follow-up-displays information, information giving prior knowledge is Including a subset of cellular objects with the highest probability of cell variants within the sample, subsequent inspection of the sample by the observer gives the observer prior knowledge-influenced by a review of information, (ii) subsequently A subset of the sample area is displayed for inspection by the observer.
[0018]
Other features and advantages of the present invention will become apparent to those of ordinary skill in the art by reading the following description, with reference where appropriate to the drawings.
[0019]
Detailed Description of the Preferred Embodiment
Referring to the drawings, FIG. 1 is a block diagram of the system flow in a preliminary embodiment of the present invention. In block 12, a sample is taken and a sample such as cervical smear or LBP is prepared from the sample, but for example a typical slide for 1 inch × 1 inch (2.54 cm × 2.54 cm) patient and slide recognition. A 1 inch x 2 inch (2.54 cm x 5.08 cm) cover adjacent to the label A 1 inch x 3 inch (2.54 cm x 7.62 cm) slide with an area covered by glass. In blocks 14 and 16, the sample is subjected to a cytological sorting system including pre-sorting by automated or interactive equipment and visual sorting by a cytotechnologist. Based on this screening process, the cytotechnologist determines whether the sample is suspect or appears to be within the normal range. All specimens that are considered suspicious are forwarded to the skilled person for block 18 examination or diagnosis. At least 10% of specimens that appear to be in the normal range are rescreened at block 16 for quality control, particularly to reduce the possibility of false negative decisions.
[0020]
FIG. 2 shows, by way of example, some functions performed in the pre-sorting and sorting steps 14, 16 of the present invention. These steps are combined and adapted for use in a clinical laboratory or similar facility, and preferably include an image capture device 100, a map maker 104, and an automated microscope-subject screening station 110. The image capturing device 100 is preferably composed of a camera and a frame capturing device. Preferably, the camera is a CCD (Charge Coupled Device) scientific grade camera with a 1K × 1K or higher element array and a third or higher class sensor. Such a camera is commercially available from Kodak, Rochester, New York, under the trade name ES-1, and is also available from Pulnix America, Sunnyvale, California. Such a camera is characterized by a sensor working area of 9 mm × 9 mm or more with a fine pixel spacing of 9 μm or more, and can capture or scan images at a rate of at least 30 frames / second, A digital output can be provided at a data rate of at least 30 MHz. The optical system provides an effective pixel resolution of about 2.4 μm in the sample. Such a resolution is sufficient for the preferred embodiment described herein, but can be varied for other applications. The specifications mentioned here are examples of a particularly preferred embodiment and can be substituted. As an example, a camera having an element array of 1K × 1K or more can reduce the number of images to be captured because each captured image includes a wider portion of the slide. As another example, pixel spacings finer than 9 μm will reach high spatial resolutions that are subject to optical physical limits.
[0021]
The camera provides a digital output to a frame grabber that operates to store the digital data received from the camera. The frame grabber preferably has a PCI type interface and is characterized by a data transmission rate of at least 50 MHz. In addition, it is desirable for the frame grabber to employ digital signal processing for optical shading correction and blob finding. A preferred frame grabber is the construction of a Data Raptor type frame grabber available from Bit Flow, Woburn, Massachusetts. In other embodiments, the frame grabber performs some kind of image processing and advanced functions using special hardware in order to increase speed even more than software software can perform such functions. There is also. For example, a frame grabber may be configured with special hardware, such as a digital signal processing circuit, to perform some functions to be performed by software described later.
[0022]
Image capture of the sample on the slide 102 is preferably performed by dividing the slide into a plurality of regions of the same size as indicated by the dotted line on the slide 102 and individually capturing the digital image of the sample one by one. Once the digital image of each region is captured, it is stored in memory and analyzed by the map builder 104. The slide area shown in FIG. 2 has been simplified for ease of explanation. In practice, the slide usually consists of much more area than shown in FIG. For example, a typical slide measuring approximately 75 mm × 25 mm roughly occupies an area of 50 mm × 25 mm by the sample. Such a slide includes about 200 non-overlapping areas of about 2.5 mm × 2.5 mm.
[0023]
In the preferred embodiment, the mapper 104 is incorporated as a software program stored in a semiconductor, magnetic, optical, or other storage device and runs on a general purpose digital computer. One such slide-map generator is the TRACCELL® device available from AccuMed International, Inc., Chicago, Illinois. The map builder 104 performs automatic image analysis of the captured digital image. For example, the mapper operates to automatically analyze the presence of cytological material for each region. If any cytological material is detected, the area is identified as a “selectable” area by the cartographer. In addition, the mapper identifies and excludes normal squamous cells and epithelial cells by subsequent analysis. As another example, the TRACCELL device can be configured to make preliminary measurements on the overall sample, such as whether the sample is in a satisfactory (ie sufficient) state for analysis. If unsatisfactory samples are then identified, they are returned without further analysis. Criteria for “unsatisfactory” specimens include samples with an insufficient number of cells and samples that are understained or overstained.
[0024]
Once all regions of the slide 102 have been captured and analyzed, as indicated by block 106, the cartographer 104 generates a plurality of tiles, as indicated by block 107. These tiles are shown as circles in slide 102 of sorting station 110 for ease of display. Each tile will correspond to the field of view chosen for inspection by the cytotechnologist using a microscope at the sorting station. Collectively, the tiles surround the cytological material that the cartographer has determined that requires inspection by a cytotechnologist. For this reason, as those skilled in the art will appreciate, other tile shapes, such as hexagons, may be employed instead to further improve sorting efficiency.
[0025]
Mapper 104 assigns spatial slide coordinates (including focal plane coordinates) to each tile or sample area of interest, and provides a routing function that limits the best path for microscopic display of a specified area of the sample. expand. The cartographer then communicates its coordinates to the sorting station 110.
[0026]
The sorting station includes a motorized microscope with microscope and a focus drive component, each operated by a computer control, an operator using an ergonomic input device, or by a combination of computer and manual control. The sorting station is coupled to the cartographer 104 via a data communication ring and, upon receiving a series of coordinates from the cartographer, an area specified by the cartographer according to a routing function or routing pattern developed by the cartographer. Displays the microscope field of view.
[0027]
A preferred sorting station is the ACCELL® specimen sorting station manufactured by AccuMed International, Chicago, Illinois. FIG. 3 shows an example of this station 110, which includes an automated electronic and optical imaging microscope (or video microscope) 210, which includes a motorized stage 214, a motorized focus driver (not shown). , And a motorized turret 220. The automatic microscope 210 is preferably an Olympus BX-40 microscope available from Olympus Optics of Tokyo, Japan, and preferably includes a pair of lenses 216 that can be individually selected by electric control. The sorting station 110 includes an automatic slide feeder 218, a slide holder 219, a bar code reading printer 221, and a light source 222. The motorized stage 214 moves along the direction indicated by the Y-axis in FIG. The slide holder 219 is connected to an electric stage 214 and is electrically driven to move along the direction indicated by the X-axis in FIG. The control board inside the station 110 receives an external signal for controlling the operation and operation of the electric stage and the slide holder, and automatically moves the specimen slide 102 relative to the microscope lens 216 in two dimensions.
[0028]
In the preferred embodiment, the camera of the image capture device 100 is attached to the video window of the microscope 210 so that images of cells are captured and the slide 102 does not have to be moved between the microscope and the camera. Apart from that, in other embodiments, the image capture device is integrated with the microscope, completely eliminating the intervention of the video window. Mapper 104 is in turn coupled to the sorting station by a direct physical data link or by a data network such as a local area network. Neither the physical structure of the cartographer, image grabber, and sorting station nor the manner in which the cartographer is coupled to the sorting station is critical, but such an arrangement physically sorts the cartographer. The mapper is separated from the station and exchanges information with a plurality of sorting stations. Alternative arrangements for combining the mapper and sorting station, such as a direct serial link, will be apparent to those skilled in the art from the disclosure of the present invention.
[0029]
A cytotechnologist trying to use the sorting station to inspect the slide inserts the slide or group of slides into the slide carrier and then loads it into the automatic slide feeder 218. The apparatus pulls the slide from the automatic feeder and scans the bar code on the slide with a bar code reader 221. The identity of the slide determined by the scanned bar code is used by the device to retrieve coordinate information from the cartograph 104. The slide is then transported from the automatic feeder onto the stage and positioned according to the coordinates received from the cartograph.
[0030]
The cytotechnologist can set the speed at which he or she examines the field of view that appears at the sorting station 110. A cytotechnologist can, for example, accelerate, decelerate, or stop an automated inspection process. The automatic inspection method can be freely changed by the user. This scheme includes, for example, proceeding to the next, stopping and repeating the sorting, continuous sorting, and slow sorting. In addition, the cytotechnologist can choose to switch to a manual inspection system at any time to inspect the surrounding area of the slide in question, for example, without being limited to the basic form of the established inspection path. Fortunately, the sorting station 110 allows the cytotechnologist to return to the basic form of the automated test path when the cytotechnologist is about to leave the established test path. In this way, the station 110 helps to ensure that the cytotechnologist does not overlook any specimen area, including areas that are very critical for accurate diagnosis.
[0031]
In a preferred embodiment, the sorting device, partially limited by the mapper 104 and the sorting station 110, stores information obtained from the sorting process and a priori information (such as patient demographics and medical history data). And a useful combination with a database management system (DMS) to facilitate the presentation of appropriate conclusions to expert pathologists to assist in diagnosis and for diagnosis There is. The DMS preferably takes the form of a programmed general purpose desktop computer with sufficient storage and computing power that can be applied to advanced database applications such as, for example, the Microsoft Windows operating environment and Microsoft Access. When screening a sample, if a cytotechnologist enters records for a certain area of the sample, for example, these records along with the spatial coordinates of the area of interest supplied by the cartographer (in a data base relationship). ) Stored in DMS.
[0032]
In subsequent diagnosis, for example, the pathologist under examination scans the specified slide by scanning a bar code or other identification code attached to the slide to reach the corresponding information stored in the DMS Or a record corresponding to the area of the slide can be reached. In this way, once the sample has been routed to a pathologist for a skilled diagnosis, the pathologist records a variety of a priori information such as cytotechnologist records, patient demographics and medical history, and responses. The sample area or area of interest to be referenced can be referenced, which can be visualized simultaneously or subsequently and examined with the benefit of simultaneously reviewing patient demographics and patient history data.
[0033]
In an improved embodiment of the present invention, an improved “pretreatment” device and a “rapid sorting” device are introduced. FIG. 4 shows the general process flow according to this improved embodiment. FIG. 5 illustrates the functional block diagram of the improved embodiment in more detail.
[0034]
Referring first to FIG. 4, in step 20, a cytological specimen is collected and a sample is prepared from the specimen. In step 22, the sample is optically scanned and a set of image data is captured in a preprocessing device. In step 24, the pre-processing device analyzes the digital image of the sample and normal and atypical healthy cells, well-stained or poorly-stained cells or cell components, well-preserved or poorly stored in the sample. Cells or cell components, subcellular organelles such as nuclei or micronuclei, cytoplasmic regions, cellular debris, cellular debris, adjacent or overlapping cells that appear as clumps or aggregates, and tissue debris Identify cellular objects such as aggregated cells that appear together. The pre-treatment device then eliminates areas that do not contain cellular material and identifies “sortable areas” as described above.
[0035]
In step 26, the pretreatment device calculates for each cellular object the probability that the cellular object is atypical ("probability of cell atypia" or more generally "probability of object atypia"). In step 28, the pretreatment device calculates the probability that the sample as a whole, i.e. the specimen behind it, is atypical (probability of sample atypia) based at least in part on the calculated probability of cell atypia.
[0036]
In step 30, the pre-processing device determines whether the calculated probability of the sample atypia falls within a range that suggests that the sample is “suspicious” or “abnormal”. These ranges can be determined by training the pre-processing equipment for a large number of training and test slides or other control data to mimic the classification ability of a skilled cytopathologist. The range will correspond to diagnostic categories such as “within normal range”, “pre-cancerous” and “cancer”. If the sample atypical probability falls within this range, in step 32 the pretreatment device classifies the sample as “suspicious” or “abnormal”. Similarly, at step 34, the pre-processing device identifies "suspicious" and "atypical" cellular objects in the specimen based on the calculated cell atypical probabilities.
[0037]
Then, in step 36, a cytotechnologist will identify these "suspicious" and "atypical" cellular objects for sorting at a sorting station such as the TRACCELL (R) guided ACCELL (R) work station described above. Submit to. In addition, this screening process is supported by a “scrutinize” device. The preparatory device allows the pretreatment device or other data management device to provide a series of preliminary knowledge about the specimen prior to the actual selection—collecting information (or a priori information) and submitting it to the cytotechnologist. Discontinuous images of the most suspicious cellular object in the sample, or give prior knowledge including repositioning the cells of the microscope for manual inspection prior to sorting-the information is whether the cytology technician is likely to be normal It's designed to help you quickly form expert opinions about suspiciousness. By having knowledge of this a priori information, the cytotechnologist will be able to use more or less time to actually sort the sample.
[0038]
If the probability of sample atypia calculated by the pretreatment device in step 30 does not fall within the range suggesting that the sample is “suspicious” or “abnormal”, then in step 38 the pretreatment device “ Apparently classified as “WNL”. Samples that are apparently classified as -WNL by the pre-processing device are then automatically classified as WNL at step 40 or submitted to a cytotechnologist for selection at step 42.
[0039]
Referring now to FIG. 5, the improved embodiment of the present invention takes the form of a comprehensive laboratory diagnostic device. In FIG. 5, at block 44, a cytological specimen is first taken from the patient and a sample such as a cervical smear is prepared from the specimen. At block 46, the sample is then optically scanned and analyzed by the pretreatment device.
[0040]
In the preferred embodiment, the pre-processing device comprises one or more devices, and preferably includes at least one computer processing device and a set of software routines to perform the various functions described below. In addition, the pre-processing device is preferably coupled to or connected to a file server that provides an electronic gateway for relevant information about the specimen, as described in more detail below. The pre-processing function of the preferred embodiment may be fully automated and performed by one or more pre-processing devices. Instead, as shown in block 47 of FIG. 5, the pre-processing functions may include fully automatic machine processing and manual “processing” or interaction. Machine and manual processing may be performed independently or may interact in an interactive environment that is tightly coupled, for example, by machine-to-human and human-to-machine feedback.
[0041]
In addition, the pre-processing function of the present invention is accomplished by a single pass or multiple passes. For example, the present invention involves first scanning a sample with low spatial and optical resolution and analyzing low resolution images for identification of selectable regions and artifact removal. Similarly, in the first pass, the preprocessed image analysis is performed on a black and white image. The present invention then scans the sample with high resolution and possibly natural colors, for example, to classify the type of cellular material in the sample (such as identifying “suspicious” or “abnormal” cellular objects or regions of interest). Analysis of the image.
[0042]
As shown in functional block 48 of FIG. 5, the pretreatment device acts to some extent as a sample presorter. This pre-processing device preferably fulfills the functions of the above-described TRACCELL® device, such as detecting the slide area to create a map and excluding unsatisfactory samples from further processing. In addition, as shown in block 50 of FIG. 5, the pre-processing device also partially serves as a “suspicious and atypical event detector and analyzer” (also referred to as “atypical analyzer”). As an atypical analyzer, the preprocessor identifies suspicious and abnormal cellular objects in the sample and automatically (i) samples based on statistical analysis of the degree of atypicality of these cellular substances. Classify as either “apparent-WNL” or (ii) “suspicious or abnormal” (apparent-non-WNL).
[0043]
To identify and evaluate cellular objects in the sample, the preprocessor uses statistical and hierarchical pattern recognition techniques to determine the probability that the object is atypical for each cellular object (ie, “cell atypical”). It is desirable to calculate the probability "). In more special cases, the pre-processing device analyzes digital images or individual cellular objects of the sample area already identified in the pre-sorting stage and characterizes the features found in these images in morphological, photometric It is desirable to compare with spectroscopic and other features that have a well-known meaning individually or collectively. These features include, for example, specific dimensions, shape, color, optical density range (gray level range or contrast), optical density distribution (texture), and topology (relative structure relative to other cells) and their characteristic values. Will include combinations.
[0044]
The clearer the extent to which a given cellular object matches or deviates from a known feature or combination of features, the easier the pre-processing device draws a conclusion about the probability of atypical object. For example, if it is known that the shape of a particular cell nucleus is associated with cancer, the pretreatment device will appear very questionable or very cellular objects that are very close to that shape. You will conclude that it is suspicious. The preprocessor will therefore calculate a high probability of atypia for such a target object. Conversely, a deviation of the cellular object from its shape indicates to the pretreatment device that the cellular object does not appear to be questionable. The pre-processing device will calculate a low atypical probability for such a target object.
[0045]
It will be understood that the information (well-known characteristics or “normal” range) referenced by the pretreatment device in the assessment of cell variants is a population-reference and / or specimen-reference. Population-reference information is statistical information such as mean, standard deviation, and statistical moment of inertia drawn from a large sample of patients. Such information will generally indicate that a particular form of cell is seen to have a particular meaning. Population-criteria information also includes information established from verification criteria samples known to be normal for a particular population and / or verification criteria samples known to be abnormal for a particular population. Would include. On the other hand, the specimen-reference information is statistical information extracted from the cells of the specimen in question, and gives a personal or individual reference value for the patient from whom the specimen is collected.
[0046]
In addition, the preprocessor considers other factors associated with the analyte in question in calculating the probability of cell atypia. These factors include, for example, the patient's medical history and demographics that indicate whether the patient from whom the specimen was taken is at particular risk for cancer or other illness or a history of abnormal cervical smear. Will contain data. Such information automatically signals the pre-processing device, and cellular objects that would otherwise be of little interest to the cytologist would be of more interest. Conversely, such information may generally indicate that an object with atypical features is practically normal for a particular patient. For example, many early symptoms of cancer are morphologically essentially the same as cytological changes that normally occur with cell repair, therapeutic treatment (such as radiation therapy) and / or various demographic factors It is well known. Therefore, the pretreatment device would be able to adjust the cell variant calculation based on such a priori information.
[0047]
In a preferred embodiment, the pretreatment device assigns a calculated cell variant probability to each cellular object in the sample. For example, a probability of 1.0 represents the most atypical object (such as apparently a cancer cell), and a probability of 0.0 represents the least atypical or most normal object (such as a healthy cell). ). The preprocessor then stores the coordinates of the cellular object (or the field of view containing the cellular object) having various calculated cell variant probabilities in the buffer storage. In addition, preferably a pre-processing device orders these objects in order of decreasing probability of atypia so that the most suspicious object of the sample can be pointed out immediately. In addition, the preprocessor can calculate and store an indication of the confidence level of each probability calculation, for example 80% that a region is of interest (ie, atypical or complex) An indicator that it is certain, or that the conclusion is reliable by 20%.
[0048]
Based on the calculated value of the probabilities of cellular variants for cellular objects in the sample, the pretreatment device then desirably obtains a statistical atypical distribution for the sample as a whole. This statistical atypical distribution will serve as the basis for a quantitative assessment or calculation of the probability that the sample as a whole (and also the underlying specimen) is atypical (or “probability of specimen atypia”). . For example, the x-axis of the distribution will be the calculated cell variant probability, and the y-axis of the distribution will be the number of cellular objects in the sample with that calculated probability. The preprocessor further weights this statistical distribution based on knowledge of the various factors that affect the calculation results, up to factors that the preprocessor has not yet considered in calculating the individual probabilities of cell variants. Attached or adjusted. For example, the pre-processing device may adjust the probability of specimen atypia calculated based on information about the patient's medical history or demographics.
[0049]
Based on the calculated sample atypical probability represented by the statistical atypical distribution, the pretreatment device determines whether the sample is apparently -WNL. For example, if the probability of a cell variant of a total cellular object in a sample is zero, the statistical variant distribution for that sample will show a zero probability of analyte atypia. On the other hand, if the distribution resembles a Poisson distribution with a maximum value of about 0.9 atypical probability and the base of the distribution extending to 0.0, the pre-processing device will justify that the specimen is abnormal or at least suspicious. I will conclude. Similarly, if the atypical distribution is biased to a high atypical probability even if the maximum value is at a low atypical probability, the pre-treatment device will show that the specimen is apparently non-WNL. And conclude that it seems suspicious or unusual.
[0050]
The path “A” in FIG. 5 represents the specimen that the pretreatment apparatus apparently classifies as -WNL. As shown in block 52, the preprocessor or other machine (or human) automatically classifies and reports some or all of these “apparent-WNL” specimens as WNL. Instead, as shown in block 54, some or all of these “apparent-WNL” specimens are submitted to a cytotechnologist for quality control screening. In this regard, the U.S. Food and Drug Administration is now limited in limited cases and allows fully automatic classification of cervical smear specimens as WNL, but to reduce the probability of false negative conclusions. In addition, it is desirable that at least a part of the specimen that is apparently -WNL by the pretreatment apparatus is selected by a cytotechnologist.
[0051]
Fortunately, cytologists can perform quality control sorting of this “apparent-WNL” sample on an automated microscope workstation such as the ACCELL® workstation described above. In this regard, parts such as TRACCELL (R) of the pretreatment device control the movement of the electric microscope stage of the workstation and transmit the coordinates of the specimen area to the workstation, thereby efficiently examining the cell. The engineer is guided to the selectable area of the specimen. In addition, knowing that the pretreatment device has already investigated the sample and found that it is apparently -WNL, the cytotechnologist is more than possible without the pretreatment device of the present invention. It will be possible to sort the sample quickly.
[0052]
Apparently at block 54-Based on his or her selection of WNL samples, the cytotechnologist may indicate that the specimen is (i) suspicious or abnormal as shown in block 58, or (ii) WNL's as shown in block 60 Determine which one. All apparently WNL specimens that are considered suspicious or abnormal by a cytotechnologist are forwarded to a skilled pathologist for reexamination and final diagnosis. In contrast, apparently-WNL specimens that are considered WNL by a cytotechnologist are classified and reported as WNL. However, as shown by path “C” in FIG. 5, at least 10% of the apparently WNL samples considered by the cytotechnologist to be WNL are re-sorted for further quality control. desirable.
[0053]
According to a preferred embodiment, all specimens that the pretreatment device classifies as “suspicious” or “abnormal” are processed by the atypical analyzer in block 50 to remove “suspect” and “abnormal” cells from “normal cells”. Distinguish and exclude that “normal” cells are subsequently processed. For example, for each cellular object, the pre-processing device is specially designed based on statistical identification experiments to maximize the accuracy of machine classification compared to expert manual diagnosis (i.e. golden rule). It is determined whether or not the calculated probability of cell atypia of the object is greater than the probability level predetermined for the threshold value.
[0054]
If so, the pretreatment device designates the special cellular object as “suspicious” or “abnormal”. However, if this is not the case, the pretreatment device designates the special cellular object as “normal”. In this way, the pretreatment device increases the cell population within each sample so that the cytotechnologist's subsequent analysis focuses only on the cellular objects that are most likely of interest.
[0055]
There may be objects in the sample that the pretreatment device cannot recognize or cannot be automatically classified as "suspicious or abnormal" or "within normal range". In order to reduce the chances of false negative decisions, the pre-treatment device should designate such unknown substances or areas of interest as “suspicious”. Alternatively, however, the pre-treatment device may provide unrecognized material to the human and participate in the human as shown in block 47 of FIG. For example, the preprocessor displays a suspicious object on a color-image monitoring device for consideration by a cytotechnologist, and based on the inspection of the object, the cytotechnologist displays a "normal" or "suspicious or abnormal" As one of the above, the specification about the proposed object is input. The pre-processing device then uses the information provided to the cytotechnologist to classify the object.
[0056]
The path “B” in FIG. 5 represents the specimen that the preprocessing apparatus classifies as “suspicious” or “abnormal”. As shown at block 56, the enriched population of “suspicious” and “abnormal” cellular material in these samples is then submitted to a cytotechnologist for sorting. For apparently-WNL specimens as discussed above, the cytotechnologist performs this sort with the help of an automated microscope workstation, such as a TRACCELL®-guided ACCELL® workstation. In this regard, since the pretreatment device has already identified the most suspicious cellular object in the sample, the pretreatment device efficiently guides the cytologist to sort out only the “worst-example” objects. , Prompting the sorting process.
[0057]
In addition, as noted above, the cytotechnologist's selection of each sample that the pretreatment device has classified as “suspicious or abnormal” should be aided by a “scrutiny” stage. For this purpose, the pre-processing device or other data management device collects various a priori information about the specimen and submits it to the cytotechnologist, who then inspects this a priori information before actually sorting the specimen. I will investigate below. In this follow-up, cytologists should focus their attention on important sample-related information without focusing on information or "noise" that has no meaning regardless of whether the sample is normal or abnormal. It is arranged. The purpose of this follow-up phase is to minimize the possibility of false negative diagnoses in the subsequent selection of cytologists by giving the cytotechnologist's attention to diagnostically meaningful information and providing prior knowledge. . Another purpose of this follow-up phase is to make it easier for cytologists to collect appropriate information about a specimen for examination by a diagnostic pathologist regarding a particular area of the specimen.
[0058]
In the preliminary examination stage, the pretreatment device provides the cytotechnologist with a variety of useful information for consideration by the cytotechnologist. This information may be given to the cytotechnologist in any way and in any convenient manner. Generally speaking, a file server stores one or more “clients” that store some or all of the information that touches the core of specimens selected at a particular cytology or remote laboratory. You may supply the appropriate data to be displayed on the "working workstation". In one embodiment, these preview display workstations may be combined together in an apparatus used as a sorting station, such as an ACCELL® workstation.
[0059]
Regardless of whether it is designed as a stand-alone device or incorporated as part of a sorting workstation, the study workstation contemplated by the present invention is at least one computer or video display device, or an observer is informed about the specimen. It is desirable to include other mechanisms for conveying the appropriate information. The workstation is manually controlled and provides a mechanism that allows cytotechnologists to flag the particularly touching information displayed during the follow-up phase. For example, when a cytotechnologist examines the investigation information, he or she operates a mouse or other selection device at the examination workstation to determine whether the specimen in question is normal, suspicious, or abnormal. Flag some information that appears to be present. If a cytotechnologist flags some information or the preprocessor generates appropriate information about the specimen in question, as described later, that information is useful for re-examination by a pathologist. Automatically added to the electronic database record paired with the specimen.
[0060]
In addition, it is desirable for the workstation to include a bar code reader and a mechanism for scanning bar codes or other indicia to provide the cytotechnologist with inspection data associated with a particular specimen. For this purpose, the examination workstation is eventually combined as part of the sorting station such as ACCELL® described above, and the bar code reader 221 of the ACCELL® sorting station is used for the specimen slide. By reading the bar code attached to, it will help to start the examination of the data that touches the core of the specimen under analysis.
[0061]
In use, the preprocessor can obtain some of its preliminary information from external insurance companies, hospitals, physicians or laboratories databases or external data sources such as direct data entry. In some cases, it is possible to generate survey information based on direct analysis of the specimen. Regardless of the source, some or all of this information will be displayed for investigation by the cytotechnologist during the follow-up phase, with prior knowledge to direct the cytotechnologist's attention to a more diagnostically meaningful aspect of the specimen. give. Therefore, all this information is called "giving background information-information".
[0062]
By way of example, but not by way of limitation, it provides the background knowledge that the present invention provides to cytologists—information includes: (i) patient demographic information, (ii) patient history such as current or previous test results Information, (iii) slides in question (such as specimen validity (eg, cell concentration and sufficiency of staining)), and (iv) specimen images that will be described in more detail later. Let's go. This information can be selectively displayed in one window of the preview display device, as illustrated in FIGS. 6a and 6b by way of example, for consideration by a cytotechnologist, or separately Displayed in the window. In addition, this information includes improved images, annotated images, and comparative displays of combinations of various data types, as well as, for example, narrative descriptions, tables, diagrams, graphs, digitized (electronic) images, And any of a variety of phenotypes, including the field of view of the microscope.
[0063]
The pre-processing device collects some a priori information about the desired specimen and identifies the cellular object of interest within the sample. As described above, for example, the preprocessing device desirably arranges the cellular material in the sample in order according to the probability of cell atypia, and the preprocessing device stores an image of the identified cellular object.
[0064]
In addition, the pre-processing device separates the atypical or suspicious cell or field image from the background image to facilitate a diagnostic display of diagnostically meaningful images. The pre-processing device, for example, identifies the location of cellular objects or fields of view within a particular digital image, eliminates background around the area of interest, and improves the contours of the objects or field images by automatic image processing. The preprocessor combines the electronically separated image with the visual mosaic image to form an artificial or real specimen or composite field of view of cellular objects from the sample without background image or "noise" To do. A composite image simulating this liquid-basic formulation is called “Electric Monolayer Preparation” (EMP). Alternatively, the pre-treatment device may make the background area darker while the area of particular interest is noticeable to effectively maintain the visual status of the cellular object or field of view. This step is required to pass the stored digital image of the specimen one or more times.
[0065]
As another example, the preprocessor uses appropriate a priori information about the spatial distribution of a type of sample to identify and order the probability of atypical cellular material within the sample. There is. For example, if the sample is a liquid-base formulation such as a ThinPrep® slide (eg, as shown in FIG. 9), the plan view of the slide will have a high cell density area (such as a cellular disk), medium Includes areas with a degree of cell density (such as bleed zones), areas with low cell density (such as annular rings), and areas with extremely low cell density (such as areas outside the border pattern). For example, it is known that an object in an annular ring with such a blend is generally more altered (eg, an artifact) than an object in a cellular disk. As a result, if the pretreatment device finds an object with substantially the same cell atypical probability in the cytoplasmic disc and the annular ring, the pretreatment device may actually be abnormal in the object in the cellular disc. Will conclude fairly that it is high.
[0066]
In addition, the pre-processing device is designed to automatically flag some other information as it appears to be relevant to the cytotechnologist's analysis. For example, the pre-processing device is designed to identify specimens that are somehow unsatisfactory due to suspicious collection methods, fixation, or staining. Rather than completely eliminating such specimens, the pretreatment device flags to indicate the fact that the collection was unsatisfactory. As another example, the pretreatment device specifically identifies cell clumps in the specimen and flags such a range of specimens as deemed appropriate.
[0067]
When the preprocessor completes the initial processing of the specimen image and data relating to the core of the specimen under analysis, the work station provides preliminary knowledge for examination by the cytotechnologist—displays information. As indicated above, one such piece of information is a specimen image. In this regard, the pre-treatment device appears to be the “most atypical”, “most suspicious”, or “most complex” area (cellular object or field of view) on the inspection workstation for consideration by cytologists. It is desirable to display. For example, these are stored images of cellular objects that the preprocessor ordered in the highest probability of atypia.
[0068]
As depicted in FIGS. 6 and 7, for example, the preprocessor displays a predetermined number of grids of the most suspicious object or field of view or alternatively a visual mosaic (EMP). These images, as discussed above, either remove the background image or darken to depict either individual cells or examination fields. In addition, the preprocessor will highlight some images on this display based on the automated conclusions of the preprocessor (e.g. color, texture, shading) and identify the attention of the cytologist. Squeeze the substance. Similarly, in a preferred embodiment, the pre-processing device works with a variety of individual images, or is associated with the specimen as a whole, and the graph scale, text, or special appearance of the specimen is viewed as of interest. Other indicators indicating the reliability of the conclusion of the preliminary processing apparatus are displayed.
[0069]
In the preview stage, the cytotechnologist eventually identifies some of these individual objects as suspicious and the cytotechnologist flags the object by clicking, for example, a mouse pointer on the individual image. The cytotechnologist further physically moves the specimen manually or by computer control to visually inspect the flagged object or area of interest. In the preferred embodiment, the cytotechnologist requests the pretreatment device to clearly evaluate a particular cell. In addition, the cytotechnologist inputs relevant information combined with the specimen record draft or the individual cell image for later review by a diagnostic expert.
[0070]
As an added advantage, when the cytotechnologist is examining these individual cells or fields, the cytotechnologist can see an enlarged digital view of the actual microscopic field or specimen area containing the cells, as shown by way of example in FIG. You can click on any of these images or otherwise select them to see the images. In order to provide this function, the pre-processing device communicates with the specimen-map making device (such as the TRACCELL® device) or acts as the mapping device and acts as a sorting station (such as the ACCELL® sorting station). To display the field of view associated with). Alternatively or additionally, the actual microscopic field of view or a region of interest selected from that field of view is displayed directly on the same monitoring device that serves as a preview display device. This method allows a cytotechnologist to quickly examine the actual situation of the cells he sees as potentially suspicious.
[0071]
In addition, as suggested above, the pre-processing device is a series of suitable acquisitions obtained by the pre-processing device for inspection by a cytologist from an external data source and / or based on automatic analysis by itself. Display information that gives prior knowledge. This information is displayed on a display device that is the same as or different from the individual image of the cell or field of view. However, in a preferred embodiment, this information is displayed on the same display device as the individual specimen image, and the cytotechnologist can consider other information regarding the specimen image.
[0072]
One aspect of the information that provides prior knowledge relates to the patient from whom the specimen was taken, including, for example, epidemiological risk factors and abnormal previous laboratory or laboratory test results. When the preparatory device of the present invention is employed in, for example, a lung cancer test (i.e., sputum screening) rather than a cervical smear test, the pretreatment device will effectively display the number of boxes of paper cigarettes that the patient has smoked per year. Let's go. If the patient is inhaling more than a defined number of boxes-years, for example, the information is critically related to whether the specimen is from a patient at high risk of developing lung cancer. Will flag this information. As another example, for cervical smear testing, information about abnormal previous test results includes results of previous cellular DNA tests on patient specimens. In addition, patient information may include, for example, other medical records of the patient, family medical history, and patient demographic information. For example, this information may include patient-specific risk factors for a particular disease based on family history data.
[0073]
Provide background knowledge-Another aspect of the information relates to other test results performed on the same specimen as the one being analyzed, for example, a sample from a portion of the same specimen contained in a separate area of the same slide. ing. As an example, if the sample has already undergone an HPV or cellular DNA polyploid test by a cytology laboratory that is performing a Papanicolaou test selection, the results of this test are reviewed to facilitate testing by a cytotechnologist. Effectively displayed on the workstation. FIG. 6b shows an example of a computerized display of an individual set of nuclear images that appear to be the most interesting within a particular specimen, along with DNA histograms and scatter plots for the same specimen.
[0074]
Give background information-Yet another aspect of the information is about the slide in question. This information includes information flagged or annotated by the preprocessing device based on analysis of the specimen image by the device itself. For example, the pretreatment device includes an indication that a particular area of the sample on the inspection display device will be more of interest because it contains a mass of cells. Information about the slide in question is consistent with standards such as how the slide was handled or mishandled by the Cytology Screening Laboratory, or the specimen is a Bethesda classification code for gynecological specimens. Or is related to. In this regard, information that touches the heart relates to specimen collection, fixation, and / or staining.
[0075]
For example, with respect to sample collection, a sample removed from a patient may contain insufficient vagina, cervical, or internal cervical components. Instead, the sample contains an inadequate number of cells and may therefore be considered unsatisfactory. To make these determinations, the preprocessor automatically analyzes the stored digital image of the sample to determine if the sample lacks cells that are expected to be present in the complete sample.
[0076]
With respect to specimen fixation, those skilled in the art understand that specimens collected for cervical smear testing must generally be immersed or sprayed with alcohol immediately after removal for storage. ing. If the specimen is not soaked or sprayed correctly in the alcohol, air drying will destroy the cell nuclear membrane, or the chromatin structure and distribution will change, causing the cervical smear slide to become blurred and the specimen diagnostic Decrease value. The preprocessor is prepared to automatically analyze the stored digital image of the sample to identify the presence of air-dried artifacts that reflect poor fixation techniques.
[0077]
With regard to staining, the pre-processing device employs an automated digital image analysis technique to determine that the sample is overstained or understained, for example, by being exposed to too much or too little exposure to hematoxylin. Alternatively, the device may determine that the sample is understained due to not being exposed to sufficient hematoxylin. In any case, the pretreatment device displays information identifying the sufficiency of staining for examination by a cytotechnologist. The operator flags such information and thus determines that the analyte in question should not go through the accelerated sorting process.
[0078]
By displaying information about unsatisfactory collection, fixation, or staining, the cytotechnologist can easily identify and focus on poorly prepared samples and make sense for re-examination by the pathologist. Flag some information. In addition, if the cytotechnologist determines, based on this information, that the sample in question is insufficient for further analysis, he or she will stick the sample to be returned without further analysis. Attach or route the sample to an experienced pathologist for diagnosis.
[0079]
In addition, the cytotechnologist gives the preliminary knowledge displayed on the review workstation-when the information is examined, especially when he or she is reviewing a grid of individual cells or field images, the preprocessor is a cytologist. Is contacted with a reference database to help analyze and rank samples (eg, RELATIONAL CYTOPATHOLOGY REFERENCE GUIDE ™ software from AccuMed International). The pre-treatment device is equipped with a database containing information about other specimens or is connected to a proximity or remote database. This database associates certain ancillary information with special cell characteristics, similar to the information provided to cytotechnologists for further investigation. When the cytotechnologist focuses on the individual specimen image of interest, the cytotechnologist queries the relevant database for information about other similar cells, or the preprocessor automatically selects the appropriate database from the database. It is prepared to display information. By doing so, the pre-processing device forms a convenient search filter based on information currently flagged by the cytologist. The pretreatment device can thus efficiently obtain database information about similar cells with similar background information.
[0080]
During the follow-up process, the pretreatment device and the technician can interact and learn from each other, and each other can obtain additional information useful for cytology technicians' subsequent sample selection and possibly also in the pathologist's diagnosis. . In principle, the cytotechnologist provides the prior knowledge displayed by the pre-processing device-benefiting from viewing the information, which focuses the cytotechnologist's attention on the diagnostically meaningful appearance of the specimen It is because it can be made. As a result, if the cytologist is unable to detect or flag anything suspicious or noteworthy about the specimen even after examining the information provided by the preparatory device, the cytologist will You don't have to spend a lot of time watching the slides. Instead, the cytotechnologist considers the specimen probably 90% to 95% normal, and the cytotechnologist may sort the entire slide for cellular abnormalities more quickly.
[0081]
Instead, if the cytotechnologist detects any possible abnormalities in this pre-screening process or flags information that suggests the presence of the anomaly, the cytotechnologist has an average probability of being abnormal In this case, which is greater than the value, it would justify spending more time than average to sort. In this way, the present invention effectively concentrates the cytotechnologist's attention during the actual sorting on the specimen that appears to be the most suspicious or abnormal. On the other hand, the present invention can make it unnecessary for the cytotechnologist to spend an excessive amount of time unnecessarily for selecting a sample that is considered to be within the normal range.
[0082]
In addition, the pretreatment device can learn meaningful information about the sample in question from the cytotechnologist's behavior or action, and the pretreatment device can obtain this information--from the sample and / or from the sample. In addition to other information gathered from external data--can be used for efficient sorting by cytotechnologists. For example, at some stage, this information may be a simple fact that a cytotechnologist has requested an exploded view of a particular specimen region or requested reference database information for comparison with a particular specimen region. Knowing the fact that the cytotechnologist has taken this action for a particular specimen region sends a signal to the pretreatment device that the region is meaningful to the cytotechnologist. This causes an area in the computer to be flagged for examination by a skilled cytopathologist, even if the cytotechnologist has not marked this particular area of interest.
[0083]
In another phase, the preprocessor will obtain information about potentially important areas of the analyte in question by monitoring the cytotechnologist's operating trends in the inspection process. In this regard, some of the cytotechnologist's reactions communicate information about important aspects of the specimen in question, even if they are unconscious. These cytotechnologists' reactions, for example, give the cytotechnologist's eye movements looking at the screen, give special background knowledge-the time that the cytotechnologist's eyes focused on the information, and the cytotechnologist Including pupil dilation. As one example, if the cytotechnologist's eyes are pointed or focused on a special image of the specimen area that is suddenly present, the newly focused area will be an area of diagnostic importance for the specimen.
[0084]
In a preferred embodiment of the present invention, based on the information collected by the pre-processing device through automatic pre-processing as in the case of a cell technician's follow-up, the pre-processing device then selects a “suspicious” or “abnormal” cellular object. A routing function is generated to facilitate automatic microscopic display at the containing field sorting station (these fields or regions are considered to have a probability of atypical cellular objects they contain). As stated above, such a routing function or routing pattern is coupled to the spatial coordinates of the specimen slide recorded during pre-sorting. In the preferred embodiment, the preprocessor is based on a routing pattern in any of a variety of criteria. This evaluation criteria may include, for example, an order in which the pre-established atypia or compositeness is reduced, a region that the cytotechnologist has flagged as of interest when undertaking, and / or cellular debris, or Includes areas that the pretreatment device has determined to be suspicious, such as over- or under-stained areas.
[0085]
In addition, the pre-processor is designed to provide the most efficient physical representation of the routing pattern at the sorting station. For those skilled in the art to understand correctly, microscopic display of specimen regions in order based on the degree of interest (probability of atypia) results in inefficient movement around the slide from one region to the next. It can be. To avoid this result, the pre-processing device may be designed such that the routing function first partitions the specimen region according to the degree of interest and then according to the position of the slide. Alternatively, the device could use EMP or alternative display means to display the most atypical cell image as shown in FIG.
[0086]
For example, if the preprocessor selects 100 regions that appear to be "most atypical", "most suspicious", or "most complex", the preprocessor will order 25 from the beginning of these regions by position. Then, the next 25 is ordered in the same manner according to the position. In this way, the sorting station can use less movement around the slide to sort out a portion of the area designated by the pretreatment device.
[0087]
As another example, if the sample is a liquid-based preparation such as a ThinPrep® slide, to find the most atypical or suspicious cells on the slide, such as the cells in the field of view shown in FIG. The preprocessor can utilize a priori information about the spatial distribution of the sample, and these areas are first provided to the observer. The present invention can then guide the observer to the remaining field of view containing other (less suspicious) cellular material. This stratified approach allows the sorter to examine the cellular material on the liquid-base formulation in the order that the most abnormal cells are encountered earlier in the sorting process.
[0088]
Giving background knowledge from the preliminary investigation stage—When provided with information, the cytotechnologist then performs the actual screening of “suspicious” or “abnormal” cellular objects in the sample. At this stage, it is desirable for the sorting station to display a microscopic field of view containing “suspicious” or “abnormal” objects at the most efficient speed according to the routing function developed by the preprocessor. However, ultimate control over the display type, such as speed and path, is left in the hands of the cytotechnologist, for example through a control board, keyboard, or other input device provided at the sorting station. In addition to displaying these fields of view through a microscope, the present invention can be extended to display of regions of interest for examination on a computer monitoring device.
[0089]
In a preferred embodiment, for example, the routing function may require the sorting station to display the microscopic areas of the specimen in order of decreasing probability of the area of interest. There is a greater difference between specimen regions that have a higher degree of "atypical", "suspicious", or "complex" than specimen regions that are given a low probability of interest by the preprocessor Recognizing this, the sorting station can be pre-adjusted to follow the path more quickly in areas with high probability and more slowly in other areas. The increase in speed when the less interesting area is displayed for sorting is continuous throughout the sorting process. Alternatively, the sorting station may, for example, display a first group of specimen regions at one speed and then display another group of regions at another speed. However, despite the automatic display according to the routing function, the cytotechnologist chooses to stop or manually change the sorting process at any time and to adjust the sorting time and speed for the user as desired. it can.
[0090]
Similarly, if the cytotechnologist recognizes that the pre-processing device is likely to be more interested in the sample area ordered as having a high level of suspicion, the sorting station will determine the area of the sample to a predetermined degree of interest. It is preferable to adjust so that the display is performed at the minimum time interval of the predetermined time. For example, the sorting station may adjust to display each of the first 10 of the “most atypical”, “most suspicious”, or “most complex” areas for at least 3 seconds. The sorting station then displays the other areas relatively earlier, allowing the cytotechnologist to interrupt the automated sorting process and proceed manually at any time. The device can literally adjust the time allocated to each field based on the probability of containing abnormal cells, the complexity of the field, the concentration of cells, the validity of staining, and many other population characteristic values.
[0091]
The cytotechnologist can route the sorting station through the area of the specimen simultaneously or alternatively, in the desired sequence or at the desired speed. For example, a cytologist may highlight each field of view that includes one of the “most atypical”, “most suspicious”, or “most complex” cells, objects, or fields of view displayed during the follow-up phase. Alternatively, the sorting station can be set to stop the sorting process. In this way, the cytotechnologist can automatically see the status of the specimen region. As another example, a cytotechnologist can conveniently set the sorting workstation to automatically stop at each field containing one or more suspicious cells, or at each complex or flagged field. In addition, the cytologist will only stop at the "most suspicious" cell that the cytotechnologist has flagged in the screening station for the sorting station, or the field of view containing the debris or understained or overstained. It can be set to stop only at a field of view that satisfies a specific evaluation criterion such as a field of view. The device can be set by individual laboratories or users for individual or laboratory-scale customer-oriented operating characteristics. Of course, as will be appreciated by those skilled in the art, the sorting station can be pre-prepared to automatically take a path through one or more of these fields of view, whether or not there is direct input by a cytologist. I can do it.
[0092]
As yet another alternative, the sorting station can pre-set or design a maximum total sorting time with the input of a cytotechnologist. This is an important requirement if this device is used to re-sort all or many WNL specimens apparently for quality control. For example, given a routing function developed by the pre-processing device, the sorting station can display the specimen region at a speed designed not to exceed a preset maximum time. However, it is desirable that again the cytotechnologist is authorized to interrupt this automatic sorting at any time or to continue automatic sorting without a maximum time limit.
[0093]
The present invention further contemplates that the information gathered in the preliminary examination stage is provided to the cytotechnologist in the sorting stage along with the specimen image to be sorted. For example, the sorting station can display appropriate information about a particular field of view on the monitoring device in text or drawing next to or over the actual field of view. In addition, the sorting station can display or indicate an indicator such as a sliding bar scale that indicates the reliability of the pre-processing device in the ordering of the target probabilities, similar to the situation at the stage of examination described above.
[0094]
Based on the screening of the cytological technician at block 56 of the “suspicious” or “abnormal” specimen, the cytotechnician determines whether the specimen is suspected or abnormal as indicated at (i) block 58, or (ii) at block 62. Determine if it is WNL as shown. All such specimens that a cytotechnologist considers suspicious or abnormal are routed to a skilled pathologist for examination and final diagnosis. In contrast, as shown by path “D” in FIG. 5, 100% of the specimens that a cytotechnologist considers WNL are the same or re-sorted by another cytotechnician to ensure quality control. Is desirable, especially if the pre-treatment device has found some reason for indicating the sample as “suspicious” or “abnormal”. After this rescreen, if the cytotechnologist still considers these specimens to be WNL, then the specimen is classified and reported as WNL.
[0095]
The preferred embodiment of the present invention has been shown and described. However, it will be understood that changes and modifications can be made to the present invention without departing from the spirit and scope of the invention as defined by the following claims.
[Brief description of the drawings]
Preferred embodiments of the invention have been described with reference to the accompanying drawings. here:
FIG. 1 is a block diagram illustrating the steps of a preliminary embodiment of the present invention.
FIG. 2 is a flowchart showing components and functions performed in the preliminary sorting and sorting apparatus according to the present invention.
FIG. 3 illustrates an automatic microscope-subject selection station employed in the present invention.
FIG. 4 is a flowchart showing a process flow in a preferred embodiment of the present invention.
FIG. 5 is a functional block diagram of a process flow in a preferred embodiment of the present invention.
FIG. 6a illustrates a multi-item information window displayed on a surveillance monitor according to a preferred embodiment of the present invention.
FIG. 6b illustrates a multi-item information window displayed on the surveillance monitor according to a preferred embodiment of the present invention.
FIG. 7 illustrates individually isolated cell images in an electronic single layer formulation (EMP) displayed in a preferred embodiment of the present invention.
FIG. 8 is a view illustrating a microscopic field including a cellular material of interest in a preferred embodiment of the present invention.
FIG. 9 illustrates a conventional “cervical smear” cervical cytology slide.
FIG. 10 illustrates a Cytyc ThinPrep® cervical cytology slide.
FIG. 11 shows the area within the cellular disc (CD) occupied by cells or other light-absorbing objects overlaid on the CD to show typical field size, number, and position. It is shown with a drawn circle.
FIG. 12 illustrates the coordinates and field of view within the ThinPrep® cytoplasmic disc for identifying suspected or abnormal cellular material.

Claims (16)

A method for inspecting a plurality of samples, each sample comprising the following steps:
(1) capturing a set of digital data symbolizing the image of the sample, wherein the sample represents a plurality of objects;
(2) analyzing the digital data to determine whether each object is suspicious or abnormal;
(3) determining, based on the analysis, whether the sample contains an object that is suspicious or abnormal;
(4) If it is determined that the sample does not contain an object that is suspect or abnormal, do not perform the following steps:
(5) If it is determined that the sample includes an object that is suspected or abnormal, the following process is performed on the object that is suspected or abnormal.
The method comprising presenting the information to be given a set of prior knowledge including the object which has been determined to be the most doubtful abnormally (a) having the observer, before SL sample by thereby continuing the observer That screening is carried out ,
(B) performing a step of identifying an order of regions of the sample that contain suspicious or non-normal objects that are presented to the observer for the screening, the steps comprising: wherein is intended to determine the presentation rate of one region of the regions, the presentation rate things are changed for each of the regions according to the degree of possibility of including at least the region is suspect or non-normal objects ,
(C) Presenting the region to the observer based on the step. Receiving input from the observer regarding information providing the background knowledge ;
Inspection method according to claim 1, further comprising a and carrying out the said process based on the entering force. 3. The inspection method according to claim 2, wherein receiving an input from the observer includes observing a behavior pattern of the observer when the observer views information giving the preliminary knowledge . The behavior pattern is a behavior pattern selected from the group consisting of movement of the observer's eyes, gaze on a specific part based on information giving preliminary knowledge , and enlargement of the size of the observer's eyes. The inspection method according to claim 3, wherein the pattern indicates an interest of the observer in the object.   The behavior pattern includes the observer requesting new information about the object, and requesting new information about the object is that the object is an object of interest of the observer. The inspection method according to claim 3, wherein the inspection method is a signal indicating   The inspection method according to claim 5, wherein the new information includes an image in which the object is destroyed.   6. The inspection method according to claim 5, wherein the new information includes reference database information for comparing the objects.   The inspection method according to claim 1, wherein the sample includes a cytological sample. 2. The information providing the prior knowledge further includes information selected from a group consisting of patient-specific information, related test information, slide operation information, and sample suitability information. Inspection method. 2. The inspection method according to claim 1, wherein the information giving the preliminary knowledge includes history information of a sample provided from a reference database. The sample is a cytological sample;
The mosaic target of a plurality of cells that have been separated from the background image a set of information that gives background information is provided to the viewer containing the information the object is determined to be suspicious or abnormal The inspection method according to claim 1, further comprising displaying.   The automatically providing the region to the observer based on the step controls an automated microscope stage and subsequently places the region into the field of view via a microscope optics. The inspection method according to 1.   The inspection method according to claim 1, wherein the sample is a Papanicolaou test sample collected from a patient.   The inspection method according to claim 13, wherein the Papanicolaou test sample is selected from the group consisting of a Papanicolaou test specimen and a preparation.   The inspection method according to claim 1, further comprising automatically classifying a part of the sample determined not to contain the suspicious or abnormal object as a normal range.   2. The method according to claim 1, further comprising: permitting the viewer to correct a presentation order and a presentation speed when the region is automatically presented to the viewer based on the step. The inspection method described.

Images