Artificial Intelligence in Medicine

Heart disease accounts for millions of deaths worldwide annually, representing a major public health concern. Large-scale heart disease screening can yield significant benefits both in terms of lives saved and economic costs. In this study, we ...

• Customized algorithms show unique contributions in disease screening tasks.
• Models tailored for each patient protect privacy and enable more accurate predictions.
• Unique algorithm design enables real-time model updates and better ...

In recent years, there has been a considerable focus on developing effective methods for monitoring health care processes. Utilizing Statistical Process Monitoring (SPM) approaches, particularly risk-adjusted control charts, has emerged as a ...

• Proposing an ML-driven SPM framework for monitoring health care processes
• Designing a GAM-based risk-adjusted control chart using stacked autoencoders
• Reducing the complexity of GAMs by omitting recurrent parameter estimation ...

Early detection and accurate identification of thyroid nodules are the major challenges in controlling and treating thyroid cancer that can be difficult even for expert physicians. Currently, many computer-aided diagnosis (CAD) systems have been ...

• A self-distillation-driven network is designed for thyroid nodule identification.
• Deformable attention captures the representations with diverse geometric variations.
• Flexible inter-layer interaction and feature aggregation via ...

Clinical prediction models tend only to incorporate structured healthcare data, ignoring information recorded in other data modalities, including free-text clinical notes. Here, we demonstrate how multimodal models that effectively leverage both ...

• We describe a method for multimodal fine-tuning of clinical language models
• The method is used for training multimodal models for predicting COVID-19 outcomes
• This yields multimodal models with better predictive performance than ...

Unsupervised domain adaptation (UDA) plays a crucial role in transferring knowledge gained from a labeled source domain to effectively apply it in an unlabeled and diverse target domain. While UDA commonly involves training on data from both ...

• We revisit the challenges of Source-free Unsupervised Domain Adaptation (SFUDA) segmentation, which easily be dominated by easy class and with noisy pseudo-label.
• We propose an efficient pseudo label denoising framework, which ...

Medical report generation is an integral part of computer-aided diagnosis aimed at reducing the workload of radiologists and physicians and alerting them of misdiagnosis risks. In general, medical report generation is an image captioning task. ...

• Knowledge enhancement using medical knowledge in dictionary form and historical knowledge.
• Multilevel alignment method reduces the modal difference between text and image.
• Experiments on IU X-ray and MIMIC-CXR datasets and ...

Physical disabilities become more common with advancing age. Rehabilitation restores function, maintaining independence for longer. However, the poor availability and accessibility of rehabilitation limits its clinical impact. ...

• AI may improve the accessibility of rehabilitation programmes.
• Clinical evaluation of artificial intelligence (AI) in rehabilitation is lacking.
• Identification of implementation challenges for AI in rehabilitation is required.

The preoperative evaluation of myometrial tumors is essential to avoid delayed treatment and to establish the appropriate surgical approach. Specifically, the differential diagnosis of leiomyosarcoma (LMS) is particularly challenging due to the ...

Display Omitted

• Leiomyosarcoma (LMS) diagnosis is complex due to clinical overlap between fibroids and LMS and class imbalance.
• A machine learning (ML) methodology is proposed for preoperative LMS vs. leiomyomas diagnosis.
• XAI is embedded to ...

Healthcare needs in rural areas differ significantly from those in urban areas. Addressing the healthcare challenges in rural communities is of paramount importance, as these regions often lack access to adequate healthcare facilities. Moreover, ...

• Suggesting methodical and systematic review of intelligent IoT-based healthcare systems in rural societies and governance
• Suggesting an inductive overview of the selected articles
• To outline the important areas where IoT-based ...

In the era of healthcare digital transformation, using electronic health record (EHR) data to generate various endpoint estimates for active monitoring is highly desirable in chronic disease management. However, traditional predictive ...

• A novel real-time endpoint prediction provides native missing parameterization and survival distribution during follow-up.
• An integrated training strategy to directly accommodate censored patients.
• Verified top-tier point estimate ...

Ventilator-associated pneumonia (VAP) is a leading cause of morbidity and mortality in intensive care units (ICUs). Early identification of patients at risk of VAP enables early intervention, which in turn improves patient outcomes. We ...

• A novel set of criteria to mark presumed ventilator-associated pneumonia (VAP) in electronic health records,
• A machine learning model to evaluate the risk of a VAP event at least 24 hours prior to any clinical suspicion of the event,

Multi-armed bandits are very simple and powerful methods to determine actions to maximize a reward in a limited number of trials. An early phase in dose-finding clinical trials needs to identify the maximum tolerated dose among multiple doses by ...

• Difficult to select an optimal dose from a limited sample size.
• Multi-armed bandit is superior methods to determine actions in a limited number.
• Application of multi-arm bandit to dose-finding designs to improve performance.
• ...

Hospital patients can have catheters and lines inserted during the course of their admission to give medicines for the treatment of medical issues, especially the central venous catheter (CVC). However, malposition of CVC will lead to many ...

• An automatic catheter status prediction framework based on radiographs is proposed.
• A high-resolution network structure is adopted to maintain high-resolution features.
• The multi-task learning framework improves the catheter status ...

Pangolin is the most popular tool for SARS-CoV-2 lineage assignment. During COVID-19, healthcare professionals and policymakers required accurate and timely lineage assignment of SARS-CoV-2 genomes for pandemic response. Therefore, tools such as ...

• SARS-CoV-2 lineage assignment tool, Pangolin, is vulnerable to adversarial attacks.
• The lineage assignments of a variety of lineages were altered in our evaluation.
• Experiments show that an attack requires very few base pairs ...

Automatic segmentation of the three substructures of glomerular filtration barrier (GFB) in transmission electron microscopy (TEM) images holds immense potential for aiding pathologists in renal disease diagnosis. However, the labor-intensive ...

• The first study of self-supervised representation learning for GFB ultrastructure segmentation.
• A hybrid pretext task GCLR integrating two pixel-level subtasks: global clustering and local restoration.
• No image-level subtask, ...

There is a need for a simple yet comprehensive tool to produce and edit pedagogical anatomy video courses, given the widespread usage of multimedia and 3D content in anatomy instruction. Anatomy teachers have minimal control over the present ...

• Anatomy teachers did not have creative control over the making of digital media used in their lessons.
• Our system takes text written by the teacher in domain-specific language to make an animated video lesson.
• Videos are retimed to ...

A disease is an abnormal condition that negatively impacts the functioning of the human body. Pathology determines the causes behind the disease and identifies its development mechanism and functional consequences. Each disease has different ...

• Google introduced federated learning in 2016 to intensify data privacy.
• Review of the models trained through federated learning, published between 2019 and 2023.
• A meta-analysis of the existing disease detection models trained ...

Hospitals use medical cyber-physical systems (MCPS) more often to give patients quality continuous care. MCPS isa life-critical, context-aware, networked system of medical equipment. It has been challenging to achieve high assurance in system ...

Twelve lead electrocardiogram signals capture unique fingerprints about the body’s biological processes and electrical activity of heart muscles. Machine learning and deep learning-based models can learn the embedded patterns in the ...

• ECG-based age and gender estimation methods are comprehensively reviewed and analyzed.
• Methods are stratified based on the type of methodology adopted and the nature of the study (conventional/deep-learning-based and technical/...

Natural language processing (NLP) combined with machine learning (ML) techniques are increasingly used to process unstructured/free-text patient-reported outcome (PRO) data available in electronic health records (EHRs). This systematic ...

• This systematic review summarizes the NLP/ML applications for analyzing PROs in clinical narratives of EHRs.
• Most studies used NLP/ML techniques to extract unstructured PROs or predict disease progression.
• Studies used different ...