[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Previous Issue
Volume 15, March-1
You seem to have javascript disabled. Please note that many of the page functionalities won't work as expected without javascript enabled.
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
4.7
3.0
Journals
Diagnostics
Volume 15
Issue 6
share announcement

Diagnostics, Volume 15, Issue 6 (March-2 2025) – 15 articles

  • Issues are regarded as officially published after their release is announced to the table of contents alert mailing list.
  • You may sign up for e-mail alerts to receive table of contents of newly released issues.
  • PDF is the official format for papers published in both, html and pdf forms. To view the papers in pdf format, click on the "PDF Full-text" link, and use the free Adobe Reader to open them.
Order results
Result details
Section
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
18 pages, 19699 KiB  
Article
Enhancing Left Ventricular Segmentation in Echocardiograms Through GAN-Based Synthetic Data Augmentation and MultiResUNet Architecture
by Vikas Kumar, Nitin Mohan Sharma, Prasant K. Mahapatra, Neeti Dogra, Lalit Maurya, Fahad Ahmad, Neelam Dahiya and Prashant Panda
Diagnostics 2025, 15(6), 663; https://doi.org/10.3390/diagnostics15060663 (registering DOI) - 9 Mar 2025
Abstract
Background: Accurate segmentation of the left ventricle in echocardiograms is crucial for the diagnosis and monitoring of cardiovascular diseases. However, this process is hindered by the limited availability of high-quality annotated datasets and the inherent complexities of echocardiogram images. Traditional methods often [...] Read more.
Background: Accurate segmentation of the left ventricle in echocardiograms is crucial for the diagnosis and monitoring of cardiovascular diseases. However, this process is hindered by the limited availability of high-quality annotated datasets and the inherent complexities of echocardiogram images. Traditional methods often struggle to generalize across varying image qualities and conditions, necessitating a more robust solution. Objectives: This study aims to enhance left ventricular segmentation in echocardiograms by developing a framework that integrates Generative Adversarial Networks (GANs) for synthetic data augmentation with a MultiResUNet architecture, providing a more accurate and reliable segmentation method. Methods: We propose a GAN-based framework that generates synthetic echocardiogram images and their corresponding segmentation masks, augmenting the available training data. The synthetic data, along with real echocardiograms from the EchoNet-Dynamic dataset, were used to train the MultiResUNet architecture. MultiResUNet incorporates multi-resolution blocks, residual connections, and attention mechanisms to effectively capture fine details at multiple scales. Additional enhancements include atrous spatial pyramid pooling (ASPP) and scaled exponential linear units (SELUs) to further improve segmentation accuracy. Results: The proposed approach significantly outperforms existing methods, achieving a Dice Similarity Coefficient of 95.68% and an Intersection over Union (IoU) of 91.62%. This represents improvements of 2.58% in Dice and 4.84% in IoU over previous segmentation techniques, demonstrating the effectiveness of GAN-based augmentation in overcoming data scarcity and improving segmentation performance. Conclusions: The integration of GAN-generated synthetic data and the MultiResUNet architecture provides a robust and accurate solution for left ventricular segmentation in echocardiograms. This approach has the potential to enhance clinical decision-making in cardiovascular medicine by improving the accuracy of automated diagnostic tools, even in the presence of limited and complex training data. Full article
(This article belongs to the Special Issue Artificial Intelligence in Cardiovascular Diseases (2024))
►▼ Show Figures

Figure 1

Figure 1
<p>Procedural framework of the Methodology.</p> Full article ">Figure 2
<p>GAN Architecture.</p> Full article ">Figure 3
<p>The architecture of the MultiResUNet Model.</p> Full article ">Figure 4
<p>ResPath Block Structure.</p> Full article ">Figure 5
<p>Synthetic images with corresponding masks Generated through GAN.</p> Full article ">Figure 6
<p>Dice Coefficient vs. Epoch and IoU vs. Epoch Graph during training (<b>left two plots</b>) and validation (<b>right two plots</b>), respectively.</p> Full article ">Figure 7
<p>ROC curve for all test images.</p> Full article ">Figure 8
<p>(<b>a</b>) 2D projection plot of high-dimensional image features before and after GAN-based approach (<b>b</b>) Local Outlier Factor (LOF) anomaly detection. The blue points correspond to the original dataset, while the red points represent the synthetic images generated by the GAN. The black markers represent outliers detected using LOF.</p> Full article ">Figure 9
<p>Boxplots of 10 fold cross validation results of proposed approach.</p> Full article ">Figure 10
<p>MultiResUNet result vs. Expert Annotated LV Segmentation. Green represents true-positive pixels, red indicates false-positive pixels, and blue highlights false-negative pixels.</p> Full article ">
11 pages, 1153 KiB  
Article
PromarkerD Versus Standard of Care Biochemical Measures for Assessing Future Renal Function Decline in Type 2 Diabetes
by Kirsten E. Peters, Isabella A. Joubert, Scott D. Bringans, Wendy A. Davis, Richard J. Lipscombe and Timothy M. E. Davis
Diagnostics 2025, 15(6), 662; https://doi.org/10.3390/diagnostics15060662 (registering DOI) - 9 Mar 2025
Abstract
Background/Objectives: The current standard of care for assessing chronic kidney disease complicating diabetes (DKD) includes measurement of estimated glomerular filtration rate (eGFR) and urinary albumin:creatinine ratio (uACR) but both tests have limitations. The present study compared the biomarker-based Promarker®D test with [...] Read more.
Background/Objectives: The current standard of care for assessing chronic kidney disease complicating diabetes (DKD) includes measurement of estimated glomerular filtration rate (eGFR) and urinary albumin:creatinine ratio (uACR) but both tests have limitations. The present study compared the biomarker-based Promarker®D test with conventional biochemical measures for predicting future kidney function decline in adults with type 2 diabetes (T2D). Methods: Baseline concentrations of apolipoprotein A-IV, CD5 antigen-like protein and insulin-like growth factor binding protein 3 were combined with age, serum HDL cholesterol and eGFR to generate PromarkerD risk scores for incident DKD/eGFR decline ≥ 30% (the primary endpoint) in 857 adults with T2D (mean age 65.4 years, 54% males). Logistic regression modelling was used to compare the association of (i) PromarkerD, (ii) eGFR, (iii) uACR, and (iv) eGFR plus uACR with this outcome during 4 years of follow-up. Results: Study participants were classified by PromarkerD as low (63%), moderate (13%), or high risk (24%) for kidney function decline at baseline. Over a mean 4.2 years, 12.5% developed the primary endpoint. PromarkerD scores showed significantly higher predictive performance (area under the receiver operating characteristic curve (AUC) 0.88 (95% confidence interval (CI) 0.85–0.91)) compared to conventional biochemical measures (AUC = 0.63–0.82). There was a progressive increase in risk with moderate and high risk by PromarkerD exhibiting greater odds of the primary endpoint compared to those at low risk (odds ratios (OR) (95% CI) 8.11 (3.99–16.94) and 21.34 (12.03–40.54), respectively, both p < 0.001). Conclusions: PromarkerD more accurately identifies adults with T2D at risk of kidney function decline than current usual care biochemical tests. Full article
(This article belongs to the Special Issue Current Issues on Kidney Diseases Diagnosis and Management 2025)
►▼ Show Figures

Figure 1

Figure 1
<p>Individual participants ordered by increasing PromarkerD risk score (abscissa) plotted against the magnitude of the score (ordinate). The participants represented by grey dots did not develop the primary endpoint and those represented by red dots did. Jittering is applied to improve the visualization of overlapping points.</p> Full article ">Figure 2
<p>Odds Ratios (ORs) and 95% confidence interval (CI) for models predicting the primary renal outcome. The vertical dashed line at OR = 1 indicates no effect. The bars represent different models, specifically PromarkerD score as a continuous variable and as risk categories, eGFR (inverted for ease of comparison), uACR and KDIGO risk categories.</p> Full article ">Figure 3
<p>Areas under the curve (AUC) of the receiver operating characteristic for PromarkerD, eGFR, uACR, and eGFR + uACR in predicting the primary renal outcome (left-hand panel) and for PromarkerD in predicting primary and secondary outcomes (right-hand panel). The dashed diagonal line represents the line of no discrimination (AUC = 0.5). In the left-hand panel, the curves for eGFR and eGFR + uACR were colinear and so only the eGFR curve is shown.</p> Full article ">
13 pages, 6116 KiB  
Article
Diagnostic and Clinical Impact of Double-Balloon Enteroscopy in Small-Bowel Inflammatory Lesions: A Retrospective Cohort Study in a Turkish Population
by Suleyman Dolu, Mehmet Emin Arayici, Soner Onem, Huseyin Dongelli and Mesut Akarsu
Diagnostics 2025, 15(6), 661; https://doi.org/10.3390/diagnostics15060661 (registering DOI) - 9 Mar 2025
Abstract
Background/Aims: Small-bowel inflammatory lesions are challenging to diagnose thanks to their anatomical complexity and the limitations of conventional imaging. Double-balloon enteroscopy (DBE) allows for direct visualization, biopsy, and therapeutic intervention. This study evaluated the diagnostic yield and clinical impact of DBE in small-bowel [...] Read more.
Background/Aims: Small-bowel inflammatory lesions are challenging to diagnose thanks to their anatomical complexity and the limitations of conventional imaging. Double-balloon enteroscopy (DBE) allows for direct visualization, biopsy, and therapeutic intervention. This study evaluated the diagnostic yield and clinical impact of DBE in small-bowel inflammatory lesions, particularly in differentiating Crohn’s disease (CD) from other etiologies. Methods: This retrospective study included 258 patients who underwent DBE for suspected small-bowel inflammatory lesions at Dokuz Eylül University Hospital (2010–2024). Patients were categorized into the CD and non-CD groups. The clinical, radiological, and endoscopic findings were also analyzed. Statistical comparisons were performed to assess the differences in presentation and DBE findings between the groups. Results: The mean patient age was 48.2 ± 17.3 years. Abdominal pain (47.7%) and diarrhea (31.8%) were the most common symptoms. The DBE findings included ulcers (45.0%), superficial mucosal changes (23.3%), and strictures (9.7%). The ileum was the most commonly affected site (31.7%). CD was diagnosed in 27.5% of the patients, while other etiologies included non-steroidal anti-inflammatory drug-induced (NSAID) enteropathy (12.0%) and malignancies (15.9%). Ulcers were significantly more frequent in patients with CD than in those without (60.3% vs. 39.0%, p = 0.002). Conclusions: DBE plays a crucial role in diagnosing small-bowel inflammatory lesions, distinguishing CD from other conditions, and guiding clinical management. It remains essential for cases requiring histopathological confirmation, offering superior diagnostic accuracy compared to noninvasive imaging. Full article
(This article belongs to the Special Issue Inflammatory Pathologies)
►▼ Show Figures

Figure 1

Figure 1
<p>Flowchart of the study.</p> Full article ">Figure 2
<p>Endoscopic images of small-bowel inflammatory lesions: (<b>A</b>) celiac disease, scalloping of jejunal fold; (<b>B</b>) celiac disease, ulcerative jejunitis; (<b>C</b>) Crohn’s disease, ileal stenosis; (<b>D</b>) Crohn’s disease, jejunal stenosis; (<b>E</b>) adenocarcinoma, jejunum; (<b>F</b>) lymphoma, jejunum; (<b>G</b>) non-steroidal anti-inflammatory drug-induced enteropathy, ileum; (<b>H</b>) graft-versus-host disease, jejunum; (<b>I</b>) autoimmune enteropathy, jejunum; (<b>J</b>) eosinophilic enteritis, jejunum; (<b>K</b>) amyloidosis, jejunum; and (<b>L</b>) lymphangiectasia, jejunum.</p> Full article ">
14 pages, 1343 KiB  
Article
Detection of Respiratory Disease Based on Surface-Enhanced Raman Scattering and Multivariate Analysis of Human Serum
by Yulia Khristoforova, Lyudmila Bratchenko, Vitalii Kupaev, Dmitry Senyushkin, Maria Skuratova, Shuang Wang, Petr Lebedev and Ivan Bratchenko
Diagnostics 2025, 15(6), 660; https://doi.org/10.3390/diagnostics15060660 (registering DOI) - 8 Mar 2025
Viewed by 196
Abstract
Background/Objectives: Chronic obstructive pulmonary disease (COPD) is a significant public health concern, affecting millions of people worldwide. This study aims to use Surface-Enhanced Raman Scattering (SERS) technology to detect the presence of respiratory conditions, with a focus on COPD. Methods: The [...] Read more.
Background/Objectives: Chronic obstructive pulmonary disease (COPD) is a significant public health concern, affecting millions of people worldwide. This study aims to use Surface-Enhanced Raman Scattering (SERS) technology to detect the presence of respiratory conditions, with a focus on COPD. Methods: The samples of human serum from 41 patients with respiratory diseases (11 patients with COPD, 20 with bronchial asthma (BA), and 10 with asthma–COPD overlap syndrome) and 103 patients with ischemic heart disease, complicated by chronic heart failure (CHF), were analyzed using SERS. A multivariate analysis of the SERS characteristics of human serum was performed using Partial Least Squares Discriminant Analysis (PLS-DA) to classify the following groups: (1) all respiratory disease patients versus the pathological referent group, which included CHF patients, and (2) patients with COPD versus those with BA. Results: We found that a combination of SERS characteristics at 638 and 1051 cm−1 could help to identify respiratory diseases. The PLS-DA model achieved a mean predictive accuracy of 0.92 for classifying respiratory diseases and the pathological referent group (0.85 sensitivity, 0.97 specificity). However, in the case of differentiating between COPD and BA, the mean predictive accuracy was only 0.61. Conclusions: Therefore, the metabolic and proteomic composition of human serum shows significant differences in respiratory disease patients compared to the pathological referent group, but the differences between patients with COPD and BA are less significant, suggesting a similarity in the serum and general pathogenetic mechanisms of these two conditions. Full article
(This article belongs to the Special Issue Advances in Pulmonary and Critical Care Medicine: Diagnosis and Management)
►▼ Show Figures

Figure 1

Figure 1
<p>Scheme of PLS-DA model building procedure.</p> Full article ">Figure 2
<p>The mean spectra with standard deviation (SD) of human serum of patients with different pathologies: (<b>a</b>) respiratory diseases vs. CHF; (<b>b</b>) different types of respiratory diseases: COPD, BA, and COPD&amp;BA.</p> Full article ">Figure 3
<p>VIPs’ distribution for constructed PLS-DA models: (<b>a</b>) “respiratory diseases vs. CHF”; (<b>b</b>) COPD vs. BA.</p> Full article ">
21 pages, 3286 KiB  
Article
The Micro-Structure of the Celiac Ganglia—A Two-Photon Microscopy Study on Parkinson’s Disease
by Diana-Theodora Morgos, Lucian-George Eftimie, Horia Nicolae, Remus Iulian Nica, Constantin Stefani, Daniela Miricescu, Radu Hristu, George A. Stanciu, Adrian Tulin and Florin Filipoiu
Diagnostics 2025, 15(6), 659; https://doi.org/10.3390/diagnostics15060659 (registering DOI) - 8 Mar 2025
Viewed by 85
Abstract
Background/Objectives: This study explores the micro-structure of celiac ganglia using two-photon microscopy (TPM) to highlight histological features in neurodegenerative conditions. Neurodegenerative diseases like Parkinson’s disease (PD) are linked to dysautonomia, impacting autonomic regulation and leading to significant gastrointestinal and autonomic symptoms. Our research [...] Read more.
Background/Objectives: This study explores the micro-structure of celiac ganglia using two-photon microscopy (TPM) to highlight histological features in neurodegenerative conditions. Neurodegenerative diseases like Parkinson’s disease (PD) are linked to dysautonomia, impacting autonomic regulation and leading to significant gastrointestinal and autonomic symptoms. Our research compares imaging results from TPM and SHG microscopy, visualizing neuronal integrity, collagen distribution, and the architectural organization of celiac ganglia. SHG specifically allows detailed imaging of collagen fibers and neuronal structures, revealing alterations in collagen density and organization that correlate with dysautonomia. Methods: The cross-sectional study was conducted at “Dr. Carol Davila” Central Military Emergency University Hospital, Bucharest, Romania, involving 70 participants diagnosed with PD (Hoehn and Yahr stages 2–4), including 35 with dysautonomia and 35 without. We utilized samples from PD patients with and without dysautonomia, applying immunohistochemical markers for sympathetic neurons. Results: Our findings reveal significant pathological changes in neuronal structure and collagen architecture. Immunohistochemical markers (neuropeptide Y, neurofilament heavy chain (NF-H), and tyrosine hydroxylase) were employed to characterize sympathetic neurons, while TPM and SHG provided high-resolution imaging of neuronal integrity and extracellular matrix composition. Conclusions: These imaging techniques present a promising tool for early diagnosis and assessment of neurodegeneration and dysautonomia in PD patients. Moreover, these techniques may represent a critical bridge between histopathological findings and clinical manifestations, underscoring their role in enhancing our understanding of neurodegeneration and autonomic dysfunction in Parkinson’s disease. Full article
(This article belongs to the Section Medical Imaging and Theranostics)
12 pages, 1395 KiB  
Article
Combinatory Flowcytometric Approach in Pediatric Acute Lymphoid Leukemia Identifies Surrogate Minimal Residual Disease Markers
by Noreen Grace George, Bhavika Rishi, Sanghmitra Ray, Manpreet Kaur, Raj Kamal, Shikha Garg, Sumit Mehndiratta, Nidhi Chopra, Shamsuz Zaman, Amitabh Singh and Aroonima Misra
Diagnostics 2025, 15(6), 658; https://doi.org/10.3390/diagnostics15060658 (registering DOI) - 8 Mar 2025
Viewed by 230
Abstract
Background/Objectives: Minimal residual disease (MRD) refers to the resistant clonal population of leukemia cells that survive induction chemotherapy, serving as a critical indicator of treatment response in pediatric Acute Lymphoid Leukemia (ALL). While flow cytometry (FCM) and molecular methods are standard for [...] Read more.
Background/Objectives: Minimal residual disease (MRD) refers to the resistant clonal population of leukemia cells that survive induction chemotherapy, serving as a critical indicator of treatment response in pediatric Acute Lymphoid Leukemia (ALL). While flow cytometry (FCM) and molecular methods are standard for MRD detection, novel leukemia-associated immunophenotype (LAIP) markers are needed when conventional markers are insufficient. Methods: MRD was assessed in 218 pediatric B-ALL patients using a combinatory approach of Different-from-Normal (DfN) and LAIP strategies. An eight-color flow cytometry panel included routine MRD markers (e.g., CD10, CD19, and CD20) and less commonly used markers (e.g., CD123, CD73, CD86). Cytogenetic and molecular profiling were integrated to evaluate the association between genetic abnormalities and MRD positivity. Results: The combined DfN and LAIP approach enhanced MRD detection sensitivity compared to individual methods. CD7 showed a significant association with MRD positivity (p = 0.003), whereas CD73 (p = 0.000) and CD86 (p = 0.002) correlated with MRD-negative status. CD123 exhibited the highest aberrancy among MRD-positive cases, while CD81 had the lowest. These findings highlight the prognostic potential of CD73 and CD86 for MRD-negative status, complementing the established utility of CD123. Conclusions: Incorporating novel markers (CD123, CD73, CD86, and CD81) into MRD panels enhances detection sensitivity and clinical applicability. These markers are compatible with standard flow cytometry, supporting their integration into routine practice for comprehensive MRD evaluation, ultimately improving therapeutic outcomes in pediatric B-ALL. Full article
(This article belongs to the Special Issue Diagnosis and Management of Hematologic Malignancies)
►▼ Show Figures

Figure 1

Figure 1
<p>This figure outlines the gating strategy and methodology used for MRD detection, utilizing an 8-colour, 3-laser flow cytometry analysis on a BD FACS CANTO<sup>TM</sup> II cytometer, with data analyzed using BD FACS DIVA <sup>TM</sup> v 8.03 software. (<b>A</b>) Time Gate vs. FSC-H: All events were initially monitored to confirm data acquisition stability. (<b>B</b>) Singlet Gate: A gate was applied on FSC-H vs. FSC-A to exclude doublets. (<b>C</b>) Viable Cell Gate: Debris, platelet clumps, and artifacts were excluded based on FSC and SSC properties to isolate viable cells. (<b>D</b>) CD45 Plot: Representative cell populations were visualized on a CD45 dot plot. (<b>E</b>) CD19+ B Cell Gate: B cells were identified using CD19 vs. SSC plots. (<b>F</b>) CD45 vs. CD19 Plot: This step refined the isolation of CD19+ cells for further analysis. (<b>G</b>) CD10 vs. CD19 Plot: Expression patterns of various markers on B cells were examined. (<b>H</b>) CD34 vs. CD19 Plot: B cell subsets, including immature (CD19+CD34+) and mature populations, were analyzed. (<b>I</b>–<b>L</b>): Residual leukemic cells (MRD) were clearly identified within predefined gates. Statistics for MRD events are provided in the figure. (Flow graphs of the patients are given in <a href="#app1-diagnostics-15-00658" class="html-app">Supplementary Images</a>).</p> Full article ">
22 pages, 1226 KiB  
Article
Determining Urinary Bile Acid Profiles to Predict Maternal and Neonatal Outcomes in Patients with Intrahepatic Cholestasis of Pregnancy
by Ping You, Min Ding, Xue Li, Yong Shao, Tingting Jiang, Yuanyuan Jia, Yuxuan Wang and Xiaoqing Zhang
Diagnostics 2025, 15(6), 657; https://doi.org/10.3390/diagnostics15060657 (registering DOI) - 8 Mar 2025
Viewed by 152
Abstract
Abstract: Objective: Intrahepatic cholestasis of pregnancy (ICP) is associated with an elevated risk of adverse perinatal outcomes, including perinatal morbidity and mortality. The objectives of this study were to evaluate the bile acid (BA) metabolism profiles in the urine of [...] Read more.
Abstract: Objective: Intrahepatic cholestasis of pregnancy (ICP) is associated with an elevated risk of adverse perinatal outcomes, including perinatal morbidity and mortality. The objectives of this study were to evaluate the bile acid (BA) metabolism profiles in the urine of patients with ICP and to investigate the association between specific BAs and maternal and neonatal outcomes in patients with ICP. Methods: A total of 127 Chinese women with ICP and 55 healthy pregnant women were enrolled in our retrospective study. Spot urine samples and clinical data were collected from pregnant women from January 2019 to December 2022 at the First Affiliated Hospital of Chongqing Medical University, Chongqing. Based on total bile acid (TBA) levels, the ICP group was subdivided into mild (10–40 μmol/L) and severe (≥40 μmol/L) ICP groups. Patients in the ICP group were further divided into two categories according to neonatal outcomes: an ICP with adverse pregnancy outcomes group and an ICP with non-adverse pregnancy outcomes group. Metabolites from maternal urine were collected and analyzed using ultra-high-performance liquid chromatography–triple quadrupole time-of-flight mass spectroscopy (UPLC-triple TOF-MS). Results: Significant differences were observed between the mild and severe ICP groups in the onset time of symptoms, gestational weeks at time of ICP diagnosis, the duration of using ursodeoxycholic acid (UDCA) drugs during pregnancy, gestational age at delivery, premature delivery, and cesarean delivery. The expression levels of the composition of different urinary bile acids including THCA, TCA, T-ω-MCA, TCA-3-S, TCDCA-3-S, TDCA-3-S, GCDCA-3-S, DCA-3-G and GDCA-3-G were remarkably higher in the ICP with adverse pregnancy outcomes group than those in the ICP with non-adverse pregnancy outcomes group and the control group. The single-parameter model used to predict adverse pregnancy outcomes in ICP had similar areas under the curve (AUCs) of the receiver operating characteristic (ROC), ranging from 0.755 to 0.869. However, an AUC of 0.886 and 95% CI were obtained by the index of combined urinary bile acids in multiple prediction models (95% CI 0.790 to 0.983, p < 0.05). TCA-3-S in the urinary bile acids had a strong positive correlation with the aspartate aminotransferase (AST) level (r = 0.617, p < 0.05). Furthermore, TCDCA-3-S and GCDCA-3-S in the urinary bile acids had a strong positive correlation with the alanine aminotransferase (ALT) level (r = 0.607, p < 0.05; r = 0.611, p < 0.05) and AST level (r = 0.629, p < 0.05; r = 0.619, p < 0.05). Conclusions: Maternal urinary bile acid profiles were prominent for the prognosis of maternal and neonatal outcomes of ICP. Elevated levels of TCA-3-S, TCDCA-3-S, and GCDCA-3-S in urine might be important predictors for indicating adverse pregnancy outcomes in ICP. Full article
(This article belongs to the Section Pathology and Molecular Diagnostics)
15 pages, 848 KiB  
Article
Contemporary Diagnosis, Management, and Early Outcomes in Children with Kawasaki Disease in Romania: A Single-Center Experience
by Cristina Ramona Rădulescu, Anca Cristina Drăgănescu, Diana Maria Băncilă, Anuţa Bilaşco, Mihai-Rareş Bădescu and Doina Anca Pleşca
Diagnostics 2025, 15(6), 656; https://doi.org/10.3390/diagnostics15060656 - 7 Mar 2025
Viewed by 224
Abstract
Background: Kawasaki disease (KD) is an acute inflammatory vasculitis with a particularly high incidence of coronary artery complications and constitutes a significant cause of acquired heart disease in children and young adults. Methods: We conducted a retrospective analysis of consecutive patients aged 0–18 [...] Read more.
Background: Kawasaki disease (KD) is an acute inflammatory vasculitis with a particularly high incidence of coronary artery complications and constitutes a significant cause of acquired heart disease in children and young adults. Methods: We conducted a retrospective analysis of consecutive patients aged 0–18 years hospitalized at the “Prof. Dr. Matei Balş” National Institute of Infectious Diseases in Bucharest with Kawasaki disease over a period of 6 years (2018–2023). Results: A total of 25 children were discharged from hospital with this diagnosis during the analyzed period. The mean age was 2.9 years, and 56% were boys. Fever ≥5 days was present in all cases, and the most frequent additional sign was the presence of oral changes. Patients were treated according to in-effect guidelines with intravenous immunoglobulin (IVIG) (100%) and acetylsalicylic acid (68%). Only two cases were considered IVIG resistant and received a second IVIG infusion. Only mild cardiovascular changes were noted in echocardiography: mild coronary artery dilatation (21.7% of cases), mild valvular regurgitation, and small pericardial effusion. Infants displayed less inflammation and higher percentages of leukocytosis, developed an increase in platelet count sooner, received IVIG faster, and had longer hospital stays. Outcomes were generally favorable, and 92% of children were discharged, while the two remaining patients were transferred to other centers. No deaths were recorded. Conclusions: To our knowledge, this is the largest contemporary Romanian cohort of Kawasaki disease published to date, outlining the local diagnostic process, therapeutic strategies, and early outcomes of Kawasaki disease. Full article
(This article belongs to the Section Clinical Diagnosis and Prognosis)
►▼ Show Figures

Figure 1

Figure 1
<p>Seasonality of Kawasaki disease showing a peak during cold months.</p> Full article ">Figure 2
<p>Frequency of main clinical features in KD cases, <span class="html-italic">n</span> = 25.</p> Full article ">Figure 3
<p>Evaluation of prolonged febrile illness. Values expressed as percent of cases tested and diagnostic yield (percent of positives from number of tests). * Viruses: adenovirus; coronavirus 229; coronavirus HKU1; coronavirus NL63; coronavirus OC43; MERS-CoV; SARS-CoV-2; human metapneumovirus; human rhinovirus/enterovirus; influenza A and B; parainfluenza virus 1, 2, 3, and 4; respiratory syncytial virus. Bacteria: Bordetella parapertussis (IS1001), Bordetella pertussis (ptxP), Chlamydia pneumoniae, Mycoplasma pneumoniae. ** Cytomegalovirus IgM, IgG; Epstein–Barr IgM; adenovirus IgM, Coxsackie virus IgM; urlian IgM; echovirus IgM; herpes simplex virus 1 + 2 IgM, HIV 1 + 2; parvo B19 virus IgM; morbillivirus IgM, IgG; rubella IgM, IgG; SARS-CoV-2. *** IL-1, IL-6, PAI-1, and TNF-alpha.</p> Full article ">Figure 4
<p>Main cardiovascular changes.</p> Full article ">
17 pages, 3023 KiB  
Systematic Review
Accuracy of Artificial Intelligence Models in Detecting Peri-Implant Bone Loss: A Systematic Review
by Maryam H. Mugri
Diagnostics 2025, 15(6), 655; https://doi.org/10.3390/diagnostics15060655 (registering DOI) - 7 Mar 2025
Viewed by 223
Abstract
Background and Objectives: AI is considered one of the most innovative technologies of this century. Its introduction into healthcare has transformed the industry, significantly impacting various aspects such as education, teaching, diagnosis, treatment planning, and patient care. Researchers have tested the accuracy of [...] Read more.
Background and Objectives: AI is considered one of the most innovative technologies of this century. Its introduction into healthcare has transformed the industry, significantly impacting various aspects such as education, teaching, diagnosis, treatment planning, and patient care. Researchers have tested the accuracy of various generations of AI models for detecting peri-implant bone loss using radiographic images. While studies have reported promising outcomes, there remains significant potential for improving these models. This systematic review aims to critically analyze the existing published literature on the accuracy of AI models in detecting peri-implant bone loss and to evaluate the current state of knowledge in this area. Methods: The guidelines established by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) were pivotal and provided a framework for preparing, implementing, and recording this systematic review. The protocol for this review was registered in PROSPERO. Four electronic databases (PubMed, Scopus, Web of Science, and Cochrane) were diligently searched on 5–6 January 2025, targeting articles published between January 2000 and December 2024. The PIRD elements (population, index test, reference test, diagnosis of interest) that helped in structuring the protocol of the present review were: P: X-ray images of humans demonstrating the bone loss around the dental implant; I: Artificial intelligence models used for detecting radiographic peri-implant bone loss; R: Expert opinions and reference standards; D: Radiographic peri-implant bone loss. The Quality Assessment and Diagnostic Accuracy Tool (QUADAS-2) was used to assess the quality of each included study. Results: Seven studies met the selection criteria and were included in the qualitative analysis. A self-designed table was used to tabulate all the relevant study characteristics. The included studies were reported to have a moderate level of certainty of evidence as assessed by the GRADE assessment. In general, all studies included in this review demonstrated a low risk of bias. Overall accuracy of the AI models varied and ranged between 61% and 94.74%. The precision values ranged from 0.63% to 100%. Whereas sensitivity and specificity values range between 67% and 94.44%, and 87% and 100%, respectively. Conclusions: The present systematic review highlights that AI models demonstrate high accuracy in detecting peri-implant bone loss using dento-maxillofacial radiographic images. Thus, AI models can serve as effective tools for the practicing dentist in confirming the diagnosis of peri-implant bone loss, ultimately aiding in accurate treatment planning and improving treatment outcomes. Full article
(This article belongs to the Section Machine Learning and Artificial Intelligence in Diagnostics)
►▼ Show Figures

Figure 1

Figure 1
<p>PRISMA 2020 flow diagram depicting the search strategy.</p> Full article ">Figure 2
<p>Country-wise distribution of the studies.</p> Full article ">Figure 3
<p>Quality assessment of the individual risk of bias domains and applicability using QUADAS-2 tool.</p> Full article ">
22 pages, 516 KiB  
Systematic Review
Positron Emission Tomography–Magnetic Resonance Imaging, a New Hybrid Imaging Modality for Dentomaxillofacial Malignancies—A Systematic Review
by Anastasia Mitsea, Nikolaos Christoloukas, Spyridoula Koutsipetsidou, Periklis Papavasileiou, Georgia Oikonomou and Christos Angelopoulos
Diagnostics 2025, 15(6), 654; https://doi.org/10.3390/diagnostics15060654 - 7 Mar 2025
Viewed by 115
Abstract
Background/Objectives: Emerging hybrid imaging modalities, like Positron Emission Tomography/Computed Tomography (PET/CT) and Positron Emission Tomography/Magnetic Resonance Imaging (PET/MRI), are useful for assessing head and neck cancer (HNC) and its prognosis during follow-up. PET/MRI systems enable simultaneous PET and MRI scans within a [...] Read more.
Background/Objectives: Emerging hybrid imaging modalities, like Positron Emission Tomography/Computed Tomography (PET/CT) and Positron Emission Tomography/Magnetic Resonance Imaging (PET/MRI), are useful for assessing head and neck cancer (HNC) and its prognosis during follow-up. PET/MRI systems enable simultaneous PET and MRI scans within a single session. These combined PET/MRI scanners merge MRI’s better soft tissue contrast and the molecular metabolic information offered by PET. Aim: To review scientific articles on the use of hybrid PET/MRI techniques in diagnosing dentomaxillofacial malignancies. Method: The available literature on the use of PET/MRI for the diagnosis of dentomaxillofacial malignancies in four online databases (Scopus, PubMed, Web of Science, and the Cochrane Library) was searched. Eligible for this review were original full-text articles on PET/MRI imaging, published between January 2010 and November 2024, based on experimental or clinical research involving humans. Results: Out of the 783 articles retrieved, only twelve articles were included in this systematic review. Nearly half of the articles (5 out of 12) concluded that PET/MRI is superior to PET, MRI, and PET/CT imaging in relation to defining malignancies’ size. Six articles found no statistically significant results and the diagnostic accuracy presented was similar in PET/MRI versus MRI and PET/CT images. Regarding the overall risk of bias, most articles had a moderate risk. Conclusions: The use of PET/MRI in HNC cases provides a more accurate diagnosis regarding dimensions of the tumor and thus a more accurate surgical approach if needed. Further prospective studies on a larger cohort of patients are required to obtain more accurate results on the application of hybrid PET/MRI. Full article
(This article belongs to the Special Issue Advances in Dental Imaging, Oral Diagnosis, and Forensic Dentistry)
►▼ Show Figures

Figure 1

Figure 1
<p>PRISMA flow diagram.</p> Full article ">
16 pages, 1104 KiB  
Article
Detection of Fractured Endodontic Instruments in Periapical Radiographs: A Comparative Study of YOLOv8 and Mask R-CNN
by İrem Çetinkaya, Ekin Deniz Çatmabacak and Emir Öztürk
Diagnostics 2025, 15(6), 653; https://doi.org/10.3390/diagnostics15060653 - 7 Mar 2025
Viewed by 105
Abstract
Background/Objectives: Accurate localization of fractured endodontic instruments (FEIs) in periapical radiographs (PAs) remains a significant challenge. This study aimed to evaluate the performance of YOLOv8 and Mask R-CNN in detecting FEIs and root canal treatments (RCTs) and compare their diagnostic capabilities with those [...] Read more.
Background/Objectives: Accurate localization of fractured endodontic instruments (FEIs) in periapical radiographs (PAs) remains a significant challenge. This study aimed to evaluate the performance of YOLOv8 and Mask R-CNN in detecting FEIs and root canal treatments (RCTs) and compare their diagnostic capabilities with those of experienced endodontists. Methods: A data set of 1050 annotated PAs was used. Mask R-CNN and YOLOv8 models were trained and evaluated for FEI and RCT detection. Metrics including accuracy, intersection over union (IoU), mean average precision at 0.5 IoU (mAP50), and inference time were analyzed. Observer agreement was assessed using inter-class correlation (ICC), and comparisons were made between AI predictions and human annotations. Results: YOLOv8 achieved an accuracy of 97.40%, a mAP50 of 98.9%, and an inference time of 14.6 ms, outperforming Mask R-CNN in speed and mAP50. Mask R-CNN demonstrated an accuracy of 98.21%, a mAP50 of 95%, and an inference time of 88.7 ms, excelling in detailed segmentation tasks. Comparative analysis revealed no statistically significant differences in diagnostic performance between the models and experienced endodontists. Conclusions: Both YOLOv8 and Mask R-CNN demonstrated high diagnostic accuracy and reliability, comparable to experienced endodontists. YOLOv8’s rapid detection capabilities make it particularly suitable for real-time clinical applications, while Mask R-CNN excels in precise segmentation. This study establishes a strong foundation for integrating AI into dental diagnostics, offering innovative solutions to improve clinical outcomes. Future research should address data diversity and explore multimodal imaging for enhanced diagnostic capabilities. Full article
(This article belongs to the Special Issue Advances in Medical Image Processing, Segmentation and Classification)
►▼ Show Figures

Figure 1

Figure 1
<p>Representative examples of Mask R-CNN’s performance on periapical radiographs (PAs) for detecting fractured endodontic instruments (FEI) and root canal treatments (RCT). The bounding boxes and associated confidence scores highlight the model’s ability to accurately identify and localize objects. Panels (<b>A1</b>–<b>E1</b>) represent the ground truth annotations marked with blue boxes for FEI and red boxes for RCT, while panels (<b>A2</b>–<b>E2</b>) depict the segmentations generated by the Mask R-CNN model, where FEI is marked with red boxes and RCT with pink boxes.</p> Full article ">Figure 2
<p>Flowchart of Mask R-CNN architecture. CNN extracts feature maps from the input image. The Region Proposal Network generates candidate regions, which are processed through RoI (Region of Interest) Align to ensure accurate spatial alignment. The extracted features are passed through FC (Fully Connected) layers for classification and bounding box regression. Additionally, Conv (Convolutional) layers are used for mask prediction.</p> Full article ">Figure 3
<p>Flowchart of YOLO architecture.</p> Full article ">Figure 4
<p>Saliency map outputs for FEI and RCT detection using YOLO and Mask R-CNN. (<b>A1</b>–<b>D1</b>) Raw periapical radiographs, (<b>A2</b>–<b>D2</b>) corresponding saliency maps. (<b>A</b>) YOLO-based saliency map for FEI detection, (<b>B</b>) YOLO-based saliency map for RCT detection, (<b>C</b>) Mask R-CNN-based saliency map for FEI detection, and (<b>D</b>) Mask R-CNN-based saliency map for RCT detection. The red boxes indicate the regions identified by the models as containing FEI or RCT, highlighting the areas of interest detected by the respective deep learning approaches.</p> Full article ">Figure 5
<p>Comparison of training and validation losses for YOLOv8 (top) and Mask R-CNN (bottom) models. The YOLOv8 graphs depict box loss (<b>A</b>) and class loss (<b>B</b>), illustrating a steady decrease in both training and validation losses with minimal divergence, indicating strong generalization and effective performance in object localization and classification. In contrast, the Mask R-CNN graph (<b>C</b>) shows the total loss across training and validation, with training loss decreasing rapidly and validation loss stabilizing with slight fluctuations, reflecting its ability to perform detailed segmentation tasks. Overall, YOLOv8 demonstrates faster convergence and smoother loss reduction, while Mask R-CNN exhibits robustness in tasks requiring precise segmentation.</p> Full article ">
14 pages, 2771 KiB  
Article
Prognostic Significance of DNAJB4 Expression in Gastric Cancer: Correlation with CD31, Caspase-3, and Tumor Progression
by Chiao-Yin Cheng, Yen-Lin Chen, Hua Ho, Chun-Yen Huang, Sheng-En Chu and Yao-Jen Liang
Diagnostics 2025, 15(6), 652; https://doi.org/10.3390/diagnostics15060652 - 7 Mar 2025
Viewed by 87
Abstract
Background/Objectives: Gastric cancer is one of the most common and lethal cancers worldwide, with particularly high incidence and mortality rates in East Asia and Europe. DNAJB4 has been shown to have prognostic implications in other cancer types; however, its expression patterns and role [...] Read more.
Background/Objectives: Gastric cancer is one of the most common and lethal cancers worldwide, with particularly high incidence and mortality rates in East Asia and Europe. DNAJB4 has been shown to have prognostic implications in other cancer types; however, its expression patterns and role in gastric cancer have not been extensively studied. This study aimed to analyze DNAJB4 expression in gastric cancer and explore its association with clinical characteristics, molecular markers, and patient outcomes. Methods: We selected suitable tumor samples from 189 gastric cancer patients who had not undergone chemotherapy or radiotherapy, with 188 patients ultimately included in the analysis. Tissue microarray and immunohistochemistry were used to evaluate DNAJB4 expression, and the samples were divided into high- and low-expression groups based on the H-score. Multivariate logistic regression and survival analysis were conducted to identify influencing factors. Results: High DNAJB4 expression was significantly correlated with increased CD31 levels but was inversely associated with advanced cancer stages. Subgroup analysis revealed that in patients with advanced gastric cancer, high DNAJB4 expression was associated with increased caspase-3 levels and with elevated CD31 and decreased E-cadherin levels. Conclusions: High DNAJB4 expression was associated with both angiogenesis and apoptosis, indicating its complex role in gastric cancer progression. Although DNAJB4 promoted angiogenesis by increasing CD31 levels, it may also enhance apoptosis in tumor cells through caspase-3-induced apoptosis. Full article
(This article belongs to the Special Issue Advances in the Diagnosis of Gastrointestinal Diseases—2nd Edition)
►▼ Show Figures

Figure 1

Figure 1
<p>Differential staining intensities of DNAJB4 in tumor paraffin sections under immunohistochemistry (grades 0 to 3).</p> Full article ">Figure 2
<p>Immunohistochemistry of CD31, pERK, pSTAT3, pAXL, EhpA5, and β-catenin in the high and low DNAJB4 expression groups. (<b>A</b>) Low DNAJB4 expression. (<b>B</b>) High DNAJB4 expression. Scale bar, 20 μm for 400× magnification.</p> Full article ">Figure 3
<p>Kaplan–Meier analysis survival curves in the high and low DNAJB4 expression groups. Pink, high DNAJB4 expression group; purple, low DNAJB4 expression group. * <span class="html-italic">p</span> &lt; 0.05.</p> Full article ">
13 pages, 466 KiB  
Review
Hyperglycemia and Lung Cancer—A Possible Relationship
by Spasoje Popevic, Nikola Maric, Branislav Ilic, Slobodan Belic, Ivana Sekulovic Radovanovic, Sanja Dimic-Janjic and Mihailo Stjepanovic
Diagnostics 2025, 15(6), 651; https://doi.org/10.3390/diagnostics15060651 - 7 Mar 2025
Viewed by 112
Abstract
Glucose is the main source of energy in human cells. Elevated levels of glucose are one of the most common metabolic disorders, and it has been shown to have a significant, mostly negative, effect on multiple chronic and acute diseases. Lung cancer remains [...] Read more.
Glucose is the main source of energy in human cells. Elevated levels of glucose are one of the most common metabolic disorders, and it has been shown to have a significant, mostly negative, effect on multiple chronic and acute diseases. Lung cancer remains one of the biggest challenges for treatment in modern medicine, with a high prevalence, incidence and mortality. Hyperglycemia is not uncommon in patients with lung cancer; however, it is usually overlooked. Patients with unregulated glycemia and lung cancer have been shown to have worse outcomes, reduced therapeutic effect and more complications during treatment. Studies have identified multiple molecular pathways common in both hyperglycemia and lung cancer; however, no clear correlation has been identified. By understanding these signaling pathways, we can influence the outcome therapeutically and thereby improve the survival of patients with lung cancer. Full article
(This article belongs to the Special Issue Hyperglycemia in Respiratory Diseases—Impact and Challenges)
►▼ Show Figures

Figure 1

Figure 1
<p>Signs and symptoms of hyperglycemia and LC, with overlap.</p> Full article ">
13 pages, 221 KiB  
Article
New Predictive Diagnostic Method for Cardiac Dynamics Based on Probability Distributions
by Javier Rodríguez Velásquez, Leonardo Juan Ramírez López and Sofia García Torres
Diagnostics 2025, 15(6), 650; https://doi.org/10.3390/diagnostics15060650 - 7 Mar 2025
Viewed by 185
Abstract
Background/Objectives: Probability theory and dynamic systems have enabled the development of diagnostic support tools that simplify Holter evaluation. Method: A study was conducted on 80 Holter tests over 21 h with patients over 21 years old. Four prototypes were selected based on normal, [...] Read more.
Background/Objectives: Probability theory and dynamic systems have enabled the development of diagnostic support tools that simplify Holter evaluation. Method: A study was conducted on 80 Holter tests over 21 h with patients over 21 years old. Four prototypes were selected based on normal, chronic, acute, and pacemaker diagnoses. An induction was created using the heart rate ranges of the prototypes, from 55 to 105, as the general probability space. Probability theory was applied to the frequency repetition ranges of 1000 to 2000 and 2001 to 3000. A blinded study was conducted with the remaining Holter tests, applying the same methodology used for the prototypes. A physical/mathematical induction was performed for the prototypes, and the other Holter tests were analyzed in a blinded study. Results: The results were compared to the predictions of the prototypes, and sensitivity, specificity, and the kappa coefficient were calculated. In the 1000–2000 range, the repetition counts for normal dynamics were 14 to 11, for chronic cases 31 to 21, for acute cases 11 to 9, and for pacemaker dynamics 5 to 4. In the 2001–3000 range, the repetitions for normal dynamics were 3 to 0, for chronic cases 14 to 10, for acute cases 6 to 3, and for pacemaker dynamics 2. The cumulative probabilities loaded for the 1000–2000 range were as follows: normal dynamics, 0.46 to 0.35; chronic dynamics, 0.48 to 0.35; acute cases, 0.6 to 0.5; and pacemaker dynamics, 0.6 to 0.5. In the 2001–3000 range, the cumulative probabilities loaded for normal dynamics were 1 to 0; for chronic cases, 0.7 to 0.54; for acute cases, 0.75 to 0.46; and for pacemaker dynamics, 1. The frequencies observed in the repetition ranges for 1000–2000 were normal, 95 to 55; chronic, 105 to 65; acute, 100 to 75; and pacemaker, 75 to 60. For the 2001–3000 range, the frequencies were normal, 95 to 65; chronic, 85 to 65; acute, 100 to 80; and pacemaker, 65 to 60. The probabilities were less than 0.3 for normal dynamics and greater than 0.3 for chronic, acute, and pacemaker dynamics across different frequency ranges, differentiating the dynamics. Conclusions: The epidemiological study results for sensitivity, specificity, and kappa coefficient were all 1. To conclude, a diagnostic support tool was developed for cardiac dynamics with clinical applications based on the appearance of frequency ranges and probability theory, enabling differentiation of normal, chronic, acute, and pacemaker dynamics. Full article
(This article belongs to the Special Issue Novel Methods and Devices for Monitoring and Diagnosing in Critical Care)
12 pages, 2818 KiB  
Article
Photon-Counting CT Scan Phantom Study: Stability of Radiomics Features
by Lama Dawi, Kodjodenis Amouzouga, Serge Muller, Cyril Nallet, Arnaud Dupont, Benoit Vielliard, Cedric Croisille, Aurelie Moussier, Gabriel Garcia, François Bidault, Remy Barbe, Salma Moalla, Thibaut Pierre, Corinne Balleyguier, Jules Dupont and Nathalie Lassau
Diagnostics 2025, 15(6), 649; https://doi.org/10.3390/diagnostics15060649 - 7 Mar 2025
Viewed by 211
Abstract
Background/Objectives: To evaluate and optimize the reconstruction parameters of images acquired with a photon-counting CT scanner to achieve a stable radiomics signal. Methods: Radiomics is a quantitative imaging biomarker correlated to survival in oncology patients. Implementing radiomics in clinical routine remains [...] Read more.
Background/Objectives: To evaluate and optimize the reconstruction parameters of images acquired with a photon-counting CT scanner to achieve a stable radiomics signal. Methods: Radiomics is a quantitative imaging biomarker correlated to survival in oncology patients. Implementing radiomics in clinical routine remains challenging due to the feature’s instability. Photon-counting CT scans use innovative technology directly converting photons into electrical signals resulting in higher-resolution images with reduced artifacts. This study used two organic phantoms: a natural wet sponge and a dry sausage. UHR images were acquired using a NAEOTOM Alpha photon-counting CT scan (Siemens) with a 0.4 mm slice thickness and 0.3 × 0.3 mm pixel size. Tube current and voltage were fixed at 112 mA and 120 KvP. A total of 24 reconstruction parameter sets were obtained by combining different values of kernel (Br), quantitative iterative reconstruction (QIR), spectral reconstruction (keV), and matrix size. Ten successive acquisitions were obtained on both phantoms. In total, 93 radiomic features were extracted on an ROI using the default parameters of Pyradiomic 3.0.1. Each feature’s stability was evaluated using the coefficient of variation (CV) within each parameter set. Results: Of the 24 reconstruction parameter sets, 5 were selected based on best image quality by seven radiologists and three radiology technologists. Radiomics features were considered stable on a set when CV was less than 15%. Feature stability was impacted by reconstruction parameters and the phantom used. The most stable combination included 90 and 65 stable features of the 93 tested on the sausage and sponge respectively. It was configured with Br36, QIR 4, 60 keV, and a 1024 × 1024 matrix size. Conclusions: Images obtained on photon-counting CT scans offer promising radiomic feature stability with optimal parameter configurations that could be applied in a clinical setting. Full article
(This article belongs to the Section Medical Imaging and Theranostics)
►▼ Show Figures

Figure 1

Figure 1
<p>Images of phantom 1 (<b>top</b>) and phantom 2 (<b>bottom</b>) acquired in UHR mode. Regions of interest (ROI) were drawn to exhibit heterogeneity in density (blue area on right images). ROI include the inner part of the dry sausage (<b>top-right</b>). ROI exclude air bubbles in the wet sponge (<b>bottom-right</b>).</p> Full article ">Figure 2
<p>Stability of radiomics features illustrated through the distribution of coefficients of variation (CV). Each blue bar corresponds to the occurrence of CV values within a CV interval of 0.05. The red line corresponds to a 0.15 CV threshold. Stable features have a CV lower than 0.15 (<b>left</b>: phantom 1, <b>right</b>: phantom 2, <b>top-to-bottom</b>: sets {1,2,3,4,5} of reconstruction parameters). UHR mode.</p> Full article ">Figure 2 Cont.
<p>Stability of radiomics features illustrated through the distribution of coefficients of variation (CV). Each blue bar corresponds to the occurrence of CV values within a CV interval of 0.05. The red line corresponds to a 0.15 CV threshold. Stable features have a CV lower than 0.15 (<b>left</b>: phantom 1, <b>right</b>: phantom 2, <b>top-to-bottom</b>: sets {1,2,3,4,5} of reconstruction parameters). UHR mode.</p> Full article ">Figure 3
<p>Stability of radiomics features illustrated through the distribution of coefficients of variation (CV). Each blue bar corresponds to the occurrence of CV values within a CV interval of 0.05. The red line corresponds to a 0.15 CV threshold. Stable features have a CV lower than 0.15 (phantom 2). STD mode. We observe that 100% of the CV values are below the 15% threshold in STD mode as opposed to UHR mode.</p> Full article ">
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-15
Previous Issue
Back to TopTop