Artificial intelligence (AI) is a promising and powerful technology with increasing use in orthopedics. The global morbidity of knee arthroplasty is expanding. This study investigated the use of AI algorithms to review radiographs of knee arthroplasty.

The Ovid-Embase, Web of Science, Cochrane Library, PubMed, China National Knowledge Infrastructure (CNKI), WeiPu (VIP), WanFang, and China Biology Medicine (CBM) databases were systematically screened from inception to March 2024 (PROSPERO study protocol registration: CRD42024507549). The quality assessment of the diagnostic accuracy studies tool assessed the risk of bias.

A total of 21 studies were included in the analysis. Of these, 10 studies identified and classified implant brands, 6 measured implant size and component alignment, 3 detected implant loosening, and 2 diagnosed prosthetic joint infections (PJI). For classifying and identifying implant brands, 5 studies demonstrated near-perfect prediction with an area under the curve (AUC) ranging from 0.98 to 1.0, and 10 achieved accuracy (ACC) between 96-100%. Regarding implant measurement, one study showed an AUC of 0.62, and two others exhibited over 80% ACC in determining component sizes. Moreover, Artificial intelligence showed good to excellent reliability across all angles in three separate studies (Intraclass Correlation Coefficient > 0.78). In predicting PJI, one study achieved an AUC of 0.91 with a corresponding ACC of 90.5%, while another reported a positive predictive value ranging from 75% to 85%. For detecting implant loosening, the AUC was found to be at least as high as 0.976 with ACC ranging from 85.8% to 97.5%.

These studies show that AI is promising in recognizing implants in knee arthroplasty. Future research should follow a rigorous approach to AI development, with comprehensive and transparent reporting of methods and the creation of open-source software programs and commercial tools that can provide clinicians with objective clinical decisions.

Foundation artificial intelligence (AI) models are capable of complex tasks that involve text, medical images, and many other types of data, but have not yet been customized for procedural medicine. This report reviews prior work in deep learning related to interventional radiology (IR), identifying barriers to generalization and deployment at scale. Moreover, this report outlines the potential design of an "IR-GPT" foundation model to provide a unified platform for AI in IR, including data collection, annotation, and training methods-while also contextualizing challenges and highlighting potential downstream applications.

Computed tomography (CT) imaging of parotid pleomorphic adenoma (PA) has been widely reported, nonetheless few reports have estimated the capsule characteristics of PA at length. This study aimed to establish and validate CT-based intratumoral and peritumoral radiomics models to clarify the characteristics between parotid PA with and without complete capsule.

In total, data of 129 patients with PA were randomly assigned to a training and test set at a ratio of 7:3. Quantitative radiomics features of the intratumoral and peritumoral regions of 2 mm and 5 mm on CT images were extracted, and radiomics models of Tumor, External2, External5, Tumor+ External2, and Tumor+External5 were constructed and used to train six different machine learning algorithms. Meanwhile, the prediction performances of different radiomics models (Tumor, External2, External5, Tumor+External2, Tumor+External5) based on single phase (plain, arterial, and venous phase) and multiphase (three-phase combination) were compared. The receiver operating characteristic (ROC) curve analysis and the area under the curve (AUC) were used to evaluate the prediction performance of each model.

Among all the established machine learning prediction radiomics models, the model based on a three-phase combination had better prediction performance, and the model using a combination of intratumoral and peritumoral radiomics features achieved a higher AUC than the model with only intratumoral or peritumoral radiomics features, and the Tumor+External2 model based on LR was the optimal model, the AUC of the test set was 0.817 (95% CI = 0.712, 0.847), and its prediction performance was significantly higher (p < 0.05, DeLong's test) than that with the Tumor model based on LDA (AUC of 0.772), the External2 model based on LR (AUC of 0.751), and the External5 model based on SVM (AUC of 0.667). And the Tumor+External2 model based on LR had a higher AUC than the Tumor+External5 model based on LDA (AUC = 0.817 vs. 0.796), but no statistically significant difference (P = 0.667).

The intratumoral and peritumoral radiomics model based on multiphasic CT images could accurately predict capsular characteristics of parotid of PA preoperatively, which may help in making treatment strategies before surgery, as well as avoid intraoperative tumor spillage and residuals.

Emergency/trauma radiology artificial intelligence (AI) is maturing along all stages of technology readiness, with research and development (R&D) ranging from data curation and algorithm development to post-market monitoring and retraining.

To develop an expert consensus document on best research practices and methodological priorities for emergency/trauma radiology AI.

A Delphi consensus exercise was conducted by the ASER AI/ML expert panel between 2022-2024. In phase 1, a steering committee (7 panelists) established key themes- curation; validity; human factors; workflow; barriers; future avenues; and ethics- and generated an edited, collated long-list of statements. In phase 2, two Delphi rounds using anonymous RAND/UCLA Likert grading were conducted with web-based data capture (round 1) and a bespoke excel document with literature hyperlinks (round 2). Between rounds, editing and knowledge synthesis helped maximize consensus. Statements reaching ≥80% agreement were included in the final document.

Delphi rounds 1 and 2 consisted of 81 and 78 items, respectively.18/21 expert panelists (86%) responded to round 1, and 15 to round 2 (17% drop-out). Consensus was reached for 65 statements. Observations were summarized and contextualized. Statements with unanimous consensus centered around transparent methodologic reporting; testing for generalizability and robustness with external data; and benchmarking performance with appropriate metrics and baselines. A manuscript draft was circulated to panelists for editing and final approval.

The document is meant as a framework to foster best-practices and further discussion among researchers working on various aspects of emergency and trauma radiology AI.

New technologies developed for triage systems can have positive effects on health care professionals. The research was conducted to identify and visualize the studies conducted between 2001 and 2024 on triage and digital triage systems in emergency departments and reveals global trends on this subject.

The data were obtained from the "Web of Science Core Collection" database on February 8th, 2024. Performance analysis, scientific mapping, and bibliometric analyses were performed using the VOSviewer (1.6.15) software program. Data from 236 publications were analyzed in the study.

The most publications were by Alcock J (n = 3), the most publications by country were published in the USA (n = 114), Harvard University (n = 19) was the institution that published the most, the United States Department of Health Human Services (n = 25) supported publications among the funding institutions, and the most publications were published in the Emergency Medicinal Journal (n = 8).

The results obtained from the study reveal the triage and digital triage systems used in emergency services, provide a general perspective on the subject, and guide future research on this subject.

This narrative review examines the transformative role of Artificial Intelligence (AI) and Machine Learning (ML) in organ retrieval and transplantation. AI and ML technologies enhance donor-recipient matching by integrating and analyzing complex datasets encompassing clinical, genetic, and demographic information, leading to more precise organ allocation and improved transplant success rates. In surgical planning, AI-driven image analysis automates organ segmentation, identifies critical anatomical features, and predicts surgical outcomes, aiding pre-operative planning and reducing intraoperative risks. Predictive analytics further enable personalized treatment plans by forecasting organ rejection, infection risks, and patient recovery trajectories, thereby supporting early intervention strategies and long-term patient management. AI also optimizes operational efficiency within transplant centers by predicting organ demand, scheduling surgeries efficiently, and managing inventory to minimize wastage, thus streamlining workflows and enhancing resource allocation. Despite these advancements, several challenges hinder the widespread adoption of AI and ML in organ transplantation. These include data privacy concerns, regulatory compliance issues, interoperability across healthcare systems, and the need for rigorous clinical validation of AI models. Addressing these challenges is essential to ensuring the reliable, safe, and ethical use of AI in clinical settings. Future directions for AI and ML in transplantation medicine include integrating genomic data for precision immunosuppression, advancing robotic surgery for minimally invasive procedures, and developing AI-driven remote monitoring systems for continuous post-transplantation care. Collaborative efforts among clinicians, researchers, and policymakers are crucial to harnessing the full potential of AI and ML, ultimately transforming transplantation medicine and improving patient outcomes while enhancing healthcare delivery efficiency.

Artificial intelligence (AI) has made significant advances in radiology. Nonetheless, challenges in AI development, validation, and reproducibility persist, primarily due to the lack of high-quality, large-scale, standardized data across the world. Addressing these challenges requires comprehensive standardization of medical imaging data and seamless integration with structured medical data.Developed by the Observational Health Data Sciences and Informatics community, the OMOP Common Data Model enables large-scale international collaborations with structured medical data. It ensures syntactic and semantic interoperability, while supporting the privacy-protected distribution of research across borders. The recently proposed Medical Imaging Common Data Model is designed to encompass all DICOM-formatted medical imaging data and integrate imaging-derived features with clinical data, ensuring their provenance.The harmonization of medical imaging data and its seamless integration with structured clinical data at a global scale will pave the way for advanced AI research in radiology. This standardization will enable federated learning, ensuring privacy-preserving collaboration across institutions and promoting equitable AI through the inclusion of diverse patient populations. Moreover, it will facilitate the development of foundation models trained on large-scale, multimodal datasets, serving as powerful starting points for specialized AI applications. Objective and transparent algorithm validation on a standardized data infrastructure will enhance reproducibility and interoperability of AI systems, driving innovation and reliability in clinical applications.

Recent healthcare advancements highlight the potential of Artificial Intelligence (AI) - and especially, among its subfields, Machine Learning (ML) - in enhancing Breast Cancer (BC) clinical care, leading to improved patient outcomes and increased radiologists' efficiency. While medical imaging techniques have significantly contributed to BC detection and diagnosis, their synergy with AI algorithms has consistently demonstrated superior diagnostic accuracy, reduced False Positives (FPs), and enabled personalized treatment strategies. Despite the burgeoning enthusiasm for leveraging AI for early and effective BC clinical care, its widespread integration into clinical practice is yet to be realized, and the evaluation of AI-based health technologies in terms of health and economic outcomes remains an ongoing endeavor.

This scoping review aims to investigate AI (and especially ML) applications that have been implemented and evaluated across diverse clinical tasks or decisions in breast imaging and to explore the current state of evidence concerning the assessment of AI-based technologies for BC clinical care within the context of Health Technology Assessment (HTA).

We conducted a systematic literature search following the Preferred Reporting Items for Systematic review and Meta-Analysis Protocols (PRISMA-P) checklist in PubMed and Scopus to identify relevant studies on AI (and particularly ML) applications in BC detection and diagnosis. We limited our search to studies published from January 2015 to October 2023. The Minimum Information about CLinical Artificial Intelligence Modeling (MI-CLAIM) checklist was used to assess the quality of AI algorithms development, evaluation, and reporting quality in the reviewed articles. The HTA Core Model® was also used to analyze the comprehensiveness, robustness, and reliability of the reported results and evidence in AI-systems' evaluations to ensure rigorous assessment of AI systems' utility and cost-effectiveness in clinical practice.

Of the 1652 initially identified articles, 104 were deemed eligible for inclusion in the review. Most studies examined the clinical effectiveness of AI-based systems (78.84%, n= 82), with one study focusing on safety in clinical settings, and 13.46% (n=14) focusing on patients' benefits. Of the studies, 31.73% (n=33) were ethically approved to be carried out in clinical practice, whereas 25% (n=26) evaluated AI systems legally approved for clinical use. Notably, none of the studies addressed the organizational implications of AI systems in clinical practice. Of the 104 studies, only two of them focused on cost-effectiveness analysis, and were analyzed separately. The average percentage scores for the first 102 AI-based studies' quality assessment based on the MI-CLAIM checklist criteria were 84.12%, 83.92%, 83.98%, 74.51%, and 14.7% for study design, data and optimization, model performance, model examination, and reproducibility, respectively. Notably, 20.59% (n=21) of these studies relied on large-scale representative real-world breast screening datasets, with only 10.78% (n =11) studies demonstrating the robustness and generalizability of the evaluated AI systems.

In bridging the gap between cutting-edge developments and seamless integration of AI systems into clinical workflows, persistent challenges encompass data quality and availability, ethical and legal considerations, robustness and trustworthiness, scalability, and alignment with existing radiologists' workflow. These hurdles impede the synthesis of comprehensive, robust, and reliable evidence to substantiate these systems' clinical utility, relevance, and cost-effectiveness in real-world clinical workflows. Consequently, evaluating AI-based health technologies through established HTA methodologies becomes complicated. We also highlight potential significant influences on AI systems' effectiveness of various factors, such as operational dynamics, organizational structure, the application context of AI systems, and practices in breast screening or examination reading of AI support tools in radiology. Furthermore, we emphasize substantial reciprocal influences on decision-making processes between AI systems and radiologists. Thus, we advocate for an adapted assessment framework specifically designed to address these potential influences on AI systems' effectiveness, mainly addressing system-level transformative implications for AI systems rather than focusing solely on technical performance and task-level evaluations.

Artificial intelligence (AI) has emerged as a transformative technology in the field of healthcare, offering significant advancements in various medical disciplines, including obstetrics. The integration of artificial intelligence into 3D/4D ultrasound analysis of fetal facial profiles presents numerous benefits. By leveraging machine learning and deep learning algorithms, AI can assist in the accurate and efficient interpretation of complex 3D/4D ultrasound data, enabling healthcare providers to make more informed decisions and deliver better prenatal care. One such innovation that has significantly improved the analysis of fetal facial profiles is the integration of AI in 3D/4D ultrasound imaging. In conclusion, the integration of artificial intelligence in the analysis of 3D/4D ultrasound data for fetal facial profiles offers numerous benefits, including improved accuracy, consistency, and efficiency in prenatal diagnosis and care.

Myopic maculopathy (MM) is a major cause of vision impairment globally. Artificial intelligence (AI) and deep learning (DL) algorithms for detecting MM from fundus images could potentially improve diagnosis and assist screening in a variety of health care settings.

To evaluate DL algorithms for MM classification and segmentation and compare their performance with that of ophthalmologists.

The Myopic Maculopathy Analysis Challenge (MMAC) was an international competition to develop automated solutions for 3 tasks: (1) MM classification, (2) segmentation of MM plus lesions, and (3) spherical equivalent (SE) prediction. Participants were provided 3 subdatasets containing 2306, 294, and 2003 fundus images, respectively, with which to build algorithms. A group of 5 ophthalmologists evaluated the same test sets for tasks 1 and 2 to ascertain performance. Results from model ensembles, which combined outcomes from multiple algorithms submitted by MMAC participants, were compared with each individual submitted algorithm. This study was conducted from March 1, 2023, to March 30, 2024, and data were analyzed from January 15, 2024, to March 30, 2024.

DL algorithms submitted as part of the MMAC competition or ophthalmologist interpretation.

MM classification was evaluated by quadratic-weighted κ (QWK), F1 score, sensitivity, and specificity. MM plus lesions segmentation was evaluated by dice similarity coefficient (DSC), and SE prediction was evaluated by R2 and mean absolute error (MAE).

The 3 tasks were completed by 7, 4, and 4 teams, respectively. MM classification algorithms achieved a QWK range of 0.866 to 0.901, an F1 score range of 0.675 to 0.781, a sensitivity range of 0.667 to 0.778, and a specificity range of 0.931 to 0.945. MM plus lesions segmentation algorithms achieved a DSC range of 0.664 to 0.687 for lacquer cracks (LC), 0.579 to 0.673 for choroidal neovascularization, and 0.768 to 0.841 for Fuchs spot (FS). SE prediction algorithms achieved an R2 range of 0.791 to 0.874 and an MAE range of 0.708 to 0.943. Model ensemble results achieved the best performance compared to each submitted algorithms, and the model ensemble outperformed ophthalmologists at MM classification in sensitivity (0.801; 95% CI, 0.764-0.840 vs 0.727; 95% CI, 0.684-0.768; P = .006) and specificity (0.946; 95% CI, 0.939-0.954 vs 0.933; 95% CI, 0.925-0.941; P = .009), LC segmentation (DSC, 0.698; 95% CI, 0.649-0.745 vs DSC, 0.570; 95% CI, 0.515-0.625; P < .001), and FS segmentation (DSC, 0.863; 95% CI, 0.831-0.888 vs DSC, 0.790; 95% CI, 0.742-0.830; P < .001).

In this diagnostic study, 15 AI models for MM classification and segmentation on a public dataset made available for the MMAC competition were validated and evaluated, with some models achieving better diagnostic performance than ophthalmologists.

Purpose.Positron emission tomography (PET) image quality can be affected by artifacts emanating from PET, computed tomography (CT), or artifacts due to misalignment between PET and CT images. Automated detection of misalignment artifacts can be helpful both in data curation and in facilitating clinical workflow. This study aimed to develop an explainable machine learning approach to detect misalignment artifacts in PET/CT imaging.Approach.This study included 1216 PET/CT images. All images were visualized and images with respiratory misalignment artifact (RMA) detected. Using previously trained models, four organs including the lungs, liver, spleen, and heart were delineated on PET and CT images separately. Data were randomly split into cross-validation (80%) and test set (20%), then two segmentations performed on PET and CT images were compared and the comparison metrics used as predictors for a random forest framework in a 10-fold scheme on cross-validation data. The trained models were tested on 20% test set data. The model's performance was calculated in terms of specificity, sensitivity, F1-Score and area under the curve (AUC).Main results.Sensitivity, specificity, and AUC of 0.82, 0.85, and 0.91 were achieved in ten-fold data split. F1_score, sensitivity, specificity, and AUC of 84.5 vs 82.3, 83.9 vs 83.8, 87.7 vs 83.5, and 93.2 vs 90.1 were achieved for cross-validation vs test set, respectively. The liver and lung were the most important organs selected after feature selection.Significance.We developed an automated pipeline to segment four organs from PET and CT images separately and used the match between these segmentations to decide about the presence of misalignment artifact. This methodology may follow the same logic as a reader detecting misalignment through comparing the contours of organs on PET and CT images. The proposed method can be used to clean large datasets or integrated into a clinical scanner to indicate artifactual cases.

The need for increased clinical efficacy and efficiency has been the main force in developing artificial intelligence (AI) tools in medical imaging. The INCISIVE project is a European Union-funded initiative aiming to revolutionize cancer imaging methods using AI technology. It seeks to address limitations in imaging techniques by developing an AI-based toolbox that improves accuracy, specificity, sensitivity, interpretability, and cost-effectiveness.

To ensure the successful implementation of the INCISIVE AI service, a study was conducted to understand the needs, challenges, and expectations of health care professionals (HCPs) regarding the proposed toolbox and any potential implementation barriers.

A mixed methods study consisting of 2 phases was conducted. Phase 1 involved user experience (UX) design workshops with users of the INCISIVE AI toolbox. Phase 2 involved a Delphi study conducted through a series of sequential questionnaires. To recruit, a purposive sampling strategy based on the project's consortium network was used. In total, 16 HCPs from Serbia, Italy, Greece, Cyprus, Spain, and the United Kingdom participated in the UX design workshops and 12 completed the Delphi study. Descriptive statistics were performed using SPSS (IBM Corp), enabling the calculation of mean rank scores of the Delphi study's lists. The qualitative data collected via the UX design workshops was analyzed using NVivo (version 12; Lumivero) software.

The workshops facilitated brainstorming and identification of the INCISIVE AI toolbox's desired features and implementation barriers. Subsequently, the Delphi study was instrumental in ranking these features, showing a strong consensus among HCPs (W=0.741, P<.001). Additionally, this study also identified implementation barriers, revealing a strong consensus among HCPs (W=0.705, P<.001). Key findings indicated that the INCISIVE AI toolbox could assist in areas such as misdiagnosis, overdiagnosis, delays in diagnosis, detection of minor lesions, decision-making in disagreement, treatment allocation, disease prognosis, prediction, treatment response prediction, and care integration throughout the patient journey. Limited resources, lack of organizational and managerial support, and data entry variability were some of the identified barriers. HCPs also had an explicit interest in AI explainability, desiring feature relevance explanations or a combination of feature relevance and visual explanations within the toolbox.

The results provide a thorough examination of the INCISIVE AI toolbox's design elements as required by the end users and potential barriers to its implementation, thus guiding the design and implementation of the INCISIVE technology. The outcome offers information about the degree of AI explainability required of the INCISIVE AI toolbox across the three services: (1) initial diagnosis; (2) disease staging, differentiation, and characterization; and (3) treatment and follow-up indicated for the toolbox. By considering the perspective of end users, INCISIVE aims to develop a solution that effectively meets their needs and drives adoption.

With profound effects on patient care, the role of artificial intelligence (AI) in radiomics has become a disruptive force in contemporary medicine. Radiomics, the quantitative feature extraction and analysis from medical images, offers useful imaging biomarkers that can reveal important information about the nature of diseases, how well patients respond to treatment and patient outcomes. The use of AI techniques in radiomics, such as machine learning and deep learning, has made it possible to create sophisticated computer-aided diagnostic systems, predictive models, and decision support tools. The many uses of AI in radiomics are examined in this review, encompassing its involvement of quantitative feature extraction from medical images, the machine learning, deep learning and computer-aided diagnostic (CAD) systems approaches in radiomics, and the effect of radiomics and AI on improving workflow automation and efficiency, optimize clinical trials and patient stratification. This review also covers the predictive modeling improvement by machine learning in radiomics, the multimodal integration and enhanced deep learning architectures, and the regulatory and clinical adoption considerations for radiomics-based CAD. Particular emphasis is given to the enormous potential for enhancing diagnosis precision, treatment personalization, and overall patient outcomes.

The National Institutes of Health-US Food and Drug Administration Joint Leadership Council Next-Generation Sequencing and Radiomics Working Group was formed by the National Institutes of Health-Food and Drug Administration Joint Leadership Council to promote the development and validation of innovative next-generation sequencing tests, radiomic tools, and associated data analysis and interpretation enhanced by artificial intelligence and machine learning technologies. A 2-day workshop was held on September 29-30, 2021, to convene members of the scientific community to discuss how to overcome the "ground truth" gap that has frequently been acknowledged as 1 of the limiting factors impeding high-quality research, development, validation, and regulatory science in these fields. This report provides a summary of the resource gaps identified by the working group and attendees, highlights existing resources and the ways they can potentially be employed to accelerate growth in these fields, and presents opportunities to support next-generation sequencing and radiomic tool development and validation using technologies such as artificial intelligence and machine learning.

The rapid advancement of artificial intelligence (AI) has sparked renewed discussions on its trustworthiness and the concept of eXplainable AI (XAI). Recent research in neuroscience has emphasized the relevance of XAI in studying cognition. This scoping review aims to identify and analyze various XAI methods used to study the mechanisms and features of cognitive function and dysfunction. In this study, the collected evidence is qualitatively assessed to develop an effective framework for approaching XAI in cognitive neuroscience. Based on the Joanna Briggs Institute and preferred reporting items for systematic reviews and meta-analyses extension for scoping review guidelines, we searched for peer-reviewed articles on MEDLINE, Embase, Web of Science, Cochrane Central Register of Controlled Trials, and Google Scholar. Two reviewers performed data screening, extraction, and thematic analysis in parallel. Twelve eligible experimental studies published in the past decade were included. The results showed that the majority (75%) focused on normal cognitive functions such as perception, social cognition, language, executive function, and memory, while others (25%) examined impaired cognition. The predominant XAI methods employed were intrinsic XAI (58.3%), followed by attribution-based (41.7%) and example-based (8.3%) post hoc methods. Explainability was applied at a local (66.7%) or global (33.3%) scope. The findings, predominantly correlational, were anatomical (83.3%) or nonanatomical (16.7%). In conclusion, while these XAI techniques were lauded for their predictive power, robustness, testability, and plausibility, limitations included oversimplification, confounding factors, and inconsistencies. The reviewed studies showcased the potential of XAI models while acknowledging current challenges in causality and oversimplification, particularly emphasizing the need for reproducibility.

Cardiovascular diseases remain the leading cause of global mortality, underscoring the critical need for accurate and timely diagnosis. This narrative review examines the current applications and future potential of artificial intelligence (AI) and machine learning (ML) in cardiovascular imaging. We discuss the integration of these technologies across various imaging modalities, including echocardiography, computed tomography, magnetic resonance imaging, and nuclear imaging techniques. The review explores AI-assisted diagnosis in key areas such as coronary artery disease detection, valve disorders assessment, cardiomyopathy classification, arrhythmia detection, and prediction of cardiovascular events. AI demonstrates promise in improving diagnostic accuracy, efficiency, and personalized care. However, significant challenges persist, including data quality standardization, model interpretability, regulatory considerations, and clinical workflow integration. We also address the limitations of current AI applications and the ethical implications of their implementation in clinical practice. Future directions point towards advanced AI architectures, multimodal imaging integration, and applications in precision medicine and population health management. The review emphasizes the need for ongoing collaboration between clinicians, data scientists, and policymakers to realize the full potential of AI in cardiovascular imaging while ensuring ethical and equitable implementation. As the field continues to evolve, addressing these challenges will be crucial for the successful integration of AI technologies into cardiovascular care, potentially revolutionizing diagnostic capabilities and improving patient outcomes.

This systematic review aims to address the research gap in the performance of computational algorithms for the digital image analysis of HER2 images in clinical settings. While numerous studies have explored various aspects of these algorithms, there is a lack of comprehensive evaluation regarding their effectiveness in real-world clinical applications. We conducted a search of the Web of Science and PubMed databases for studies published from 31 December 2013 to 30 June 2024, focusing on performance effectiveness and components such as dataset size, diversity and source, ground truth, annotation, and validation methods. The study was registered with PROSPERO (CRD42024525404). Key questions guiding this review include the following: How effective are current computational algorithms at detecting HER2 status in digital images? What are the common validation methods and dataset characteristics used in these studies? Is there standardization of algorithm evaluations of clinical applications that can improve the clinical utility and reliability of computational tools for HER2 detection in digital image analysis? We identified 6833 publications, with 25 meeting the inclusion criteria. The accuracy rate with clinical datasets varied from 84.19% to 97.9%. The highest accuracy was achieved on the publicly available Warwick dataset at 98.8% in synthesized datasets. Only 12% of studies used separate datasets for external validation; 64% of studies used a combination of accuracy, precision, recall, and F1 as a set of performance measures. Despite the high accuracy rates reported in these studies, there is a notable absence of direct evidence supporting their clinical application. To facilitate the integration of these technologies into clinical practice, there is an urgent need to address real-world challenges and overreliance on internal validation. Standardizing study designs on real clinical datasets can enhance the reliability and clinical applicability of computational algorithms in improving the detection of HER2 cancer.

The advent of artificial intelligence (AI) is revolutionizing medicine, particularly radiology. With the development of newer models, AI applications are demonstrating improved performance and versatile utility in the clinical setting. Thoracic imaging is an area of profound interest, given the prevalence of chest imaging and the significant health implications of thoracic diseases. This review aims to highlight the promising applications of AI within thoracic imaging. It examines the role of AI, including its contributions to improving diagnostic evaluation and interpretation, enhancing workflow, and aiding in invasive procedures. Next, it further highlights the current challenges and limitations faced by AI, such as the necessity of 'big data', ethical and legal considerations, and bias in representation. Lastly, it explores the potential directions for the application of AI in thoracic radiology.

Patient-reported outcome measures (PROM) play a critical role in evaluating the success of treatment interventions for musculoskeletal conditions. However, predicting which patients will benefit from treatment interventions is complex and influenced by a multitude of factors. Artificial intelligence (AI) may better anticipate the propensity to achieve clinically meaningful outcomes through leveraging complex predictive analytics that allow for personalized medicine. This article provides a contemporary review of current applications of AI developed to predict clinically significant outcome (CSO) achievement after musculoskeletal treatment interventions.

The highest volume of literature exists in the subspecialties of total joint arthroplasty, spine, and sports medicine, with only three studies identified in the remaining orthopedic subspecialties combined. Performance is widely variable across models, with most studies only reporting discrimination as a performance metric. Given the complexity inherent in predictive modeling for this task, including data availability, data handling, model architecture, and outcome selection, studies vary widely in their methodology and results. Importantly, the majority of studies have not been externally validated or demonstrate important methodological limitations, precluding their implementation into clinical settings. A substantial body of literature has accumulated demonstrating variable internal validity, limited scope, and low potential for clinical deployment. The majority of studies attempt to predict the MCID-the lowest bar of clinical achievement. Though a small proportion of models demonstrate promise and highlight the utility of AI, important methodological limitations need to be addressed moving forward to leverage AI-based applications for clinical deployment.

This work puts forth and demonstrates the utility of a reporting framework for collecting and evaluating annotations of medical images used for training and testing artificial intelligence (AI) models in assisting detection and diagnosis. AI has unique reporting requirements, as shown by the AI extensions to the Consolidated Standards of Reporting Trials (CONSORT) and Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) checklists and the proposed AI extensions to the Standards for Reporting Diagnostic Accuracy (STARD) and Transparent Reporting of a Multivariable Prediction model for Individual Prognosis or Diagnosis (TRIPOD) checklists. AI for detection and/or diagnostic image analysis requires complete, reproducible, and transparent reporting of the annotations and metadata used in training and testing data sets. In an earlier work by other researchers, an annotation workflow and quality checklist for computational pathology annotations were proposed. In this manuscript, we operationalize this workflow into an evaluable quality checklist that applies to any reader-interpreted medical images, and we demonstrate its use for an annotation effort in digital pathology. We refer to this quality framework as the Collection and Evaluation of Annotations for Reproducible Reporting of Artificial Intelligence (CLEARR-AI).

Artificial intelligence (AI), particularly deep learning, has made enormous strides in medical imaging analysis. In the field of musculoskeletal radiology, deep-learning models are actively being developed for the identification and evaluation of bone fractures. These methods provide numerous benefits to radiologists such as increased diagnostic accuracy and efficiency while also achieving standalone performances comparable or superior to clinician readers. Various algorithms are already commercially available for integration into clinical workflows, with the potential to improve healthcare delivery and shape the future practice of radiology. In this systematic review, we explore the performance of current AI methods in the identification and evaluation of fractures, particularly those in the ankle, wrist, hip, and ribs. We also discuss current commercially available products for fracture detection and provide an overview of the current limitations of this technology and future directions of the field.

Machine Learning models need large amounts of annotated data for training. In the field of medical imaging, labeled data is especially difficult to obtain because the annotations have to be performed by qualified physicians. Natural Language Processing (NLP) tools can be applied to radiology reports to extract labels for medical images automatically. Compared to manual labeling, this approach requires smaller annotation efforts and can therefore facilitate the creation of labeled medical image data sets. In this article, we summarize the literature on this topic spanning from 2013 to 2023, starting with a meta-analysis of the included articles, followed by a qualitative and quantitative systematization of the results. Overall, we found four types of studies on the extraction of labels from radiology reports: those describing systems based on symbolic NLP, statistical NLP, neural NLP, and those describing systems combining or comparing two or more of the latter. Despite the large variety of existing approaches, there is still room for further improvement. This work can contribute to the development of new techniques or the improvement of existing ones.

Prostate-specific membrane antigen (PSMA) PET/CT imaging is widely used for quantitative image analysis, especially in radioligand therapy (RLT) for metastatic castration-resistant prostate cancer (mCRPC). Unknown features influencing PSMA biodistribution can be explored by analyzing segmented organs at risk (OAR) and lesions. Manual segmentation is time-consuming and labor-intensive, so automated segmentation methods are desirable. Training deep-learning segmentation models is challenging due to the scarcity of high-quality annotated images. Addressing this, we developed shifted windows UNEt TRansformers (Swin UNETR) for fully automated segmentation. Within a self-supervised framework, the model's encoder was pre-trained on unlabeled data. The entire model was fine-tuned, including its decoder, using labeled data.

In this work, 752 whole-body [68Ga]Ga-PSMA-11 PET/CT images were collected from two centers. For self-supervised model pre-training, 652 unlabeled images were employed. The remaining 100 images were manually labeled for supervised training. In the supervised training phase, 5-fold cross-validation was used with 64 images for model training and 16 for validation, from one center. For testing, 20 hold-out images, evenly distributed between two centers, were used. Image segmentation and quantification metrics were evaluated on the test set compared to the ground-truth segmentation conducted by a nuclear medicine physician.

The model generates high-quality OARs and lesion segmentation in lesion-positive cases, including mCRPC. The results show that self-supervised pre-training significantly improved the average dice similarity coefficient (DSC) for all classes by about 3%. Compared to nnU-Net, a well-established model in medical image segmentation, our approach outperformed with a 5% higher DSC. This improvement was attributed to our model's combined use of self-supervised pre-training and supervised fine-tuning, specifically when applied to PET/CT input. Our best model had the lowest DSC for lesions at 0.68 and the highest for liver at 0.95.

We developed a state-of-the-art neural network using self-supervised pre-training on whole-body [68Ga]Ga-PSMA-11 PET/CT images, followed by fine-tuning on a limited set of annotated images. The model generates high-quality OARs and lesion segmentation for PSMA image analysis. The generalizable model holds potential for various clinical applications, including enhanced RLT and patient-specific internal dosimetry.

We aimed to develop a diagnostic deep learning model using contrast-enhanced CT images and to investigate whether cervical lymphadenopathies can be diagnosed with these deep learning methods without radiologist interpretations and histopathological examinations.

A total of 400 patients who underwent surgery for lymphadenopathy in the neck between 2010 and 2022 were retrospectively analyzed. They were examined in four groups of 100 patients: the granulomatous diseases group, the lymphoma group, the squamous cell tumor group, and the reactive hyperplasia group. The diagnoses of the patients were confirmed histopathologically. Two CT images from all the patients in each group were used in the study. The CT images were classified using ResNet50, NASNetMobile, and DenseNet121 architecture input.

The classification accuracies obtained with ResNet50, DenseNet121, and NASNetMobile were 92.5%, 90.62, and 87.5, respectively.

Deep learning is a useful diagnostic tool in diagnosing cervical lymphadenopathy. In the near future, many diseases could be diagnosed with deep learning models without radiologist interpretations and invasive examinations such as histopathological examinations. However, further studies with much larger case series are needed to develop accurate deep-learning models.

This paper provides an in-depth analysis of the clinical applications of artificial intelligence (AI) in Nuclear Medicine, focusing on three key areas: neurology, cardiology, and oncology. Beginning with neurology, specifically Alzheimer's disease and Parkinson's disease, the paper examines reviews on diagnosis and treatment planning. The same pattern is followed in cardiology studies. In the final section on oncology, the paper explores the various AI applications in multiple cancer types, including lung, head and neck, lymphoma, and pancreatic cancer.

Positron emission tomography (PET) is vital for diagnosing diseases and monitoring treatments. Conventional image reconstruction (IR) techniques like filtered backprojection and iterative algorithms are powerful but face limitations. PET IR can be seen as an image-to-image translation. Artificial intelligence (AI) and deep learning (DL) using multilayer neural networks enable a new approach to this computer vision task. This review aims to provide mutual understanding for nuclear medicine professionals and AI researchers. We outline fundamentals of PET imaging as well as state-of-the-art in AI-based PET IR with its typical algorithms and DL architectures. Advances improve resolution and contrast recovery, reduce noise, and remove artifacts via inferred attenuation and scatter correction, sinogram inpainting, denoising, and super-resolution refinement. Kernel-priors support list-mode reconstruction, motion correction, and parametric imaging. Hybrid approaches combine AI with conventional IR. Challenges of AI-assisted PET IR include availability of training data, cross-scanner compatibility, and the risk of hallucinated lesions. The need for rigorous evaluations, including quantitative phantom validation and visual comparison of diagnostic accuracy against conventional IR, is highlighted along with regulatory issues. First approved AI-based applications are clinically available, and its impact is foreseeable. Emerging trends, such as the integration of multimodal imaging and the use of data from previous imaging visits, highlight future potentials. Continued collaborative research promises significant improvements in image quality, quantitative accuracy, and diagnostic performance, ultimately leading to the integration of AI-based IR into routine PET imaging protocols.

Background and Objective - In the field of medical image analysis, achieving high accuracy is not enough; ensuring well-calibrated predictions is also crucial. Confidence scores of a deep neural network play a pivotal role in explainability by providing insights into the model's certainty, identifying cases that require attention, and establishing trust in its predictions. Consequently, the significance of a well-calibrated model becomes paramount in the medical imaging domain, where accurate and reliable predictions are of utmost importance. While there has been a significant effort towards training modern deep neural networks to achieve high accuracy on medical imaging tasks, model calibration and factors that affect it remain under-explored. Methods - To address this, we conducted a comprehensive empirical study that explores model performance and calibration under different training regimes. We considered fully supervised training, which is the prevailing approach in the community, as well as rotation-based self-supervised method with and without transfer learning, across various datasets and architecture sizes. Multiple calibration metrics were employed to gain a holistic understanding of model calibration. Results - Our study reveals that factors such as weight distributions and the similarity of learned representations correlate with the calibration trends observed in the models. Notably, models trained using rotation-based self-supervised pretrained regime exhibit significantly better calibration while achieving comparable or even superior performance compared to fully supervised models across different medical imaging datasets. Conclusion - These findings shed light on the importance of model calibration in medical image analysis and highlight the benefits of incorporating self-supervised learning approach to improve both performance and calibration.

Two data-driven algorithms were developed for detecting and characterizing Inferior Vena Cava (IVC) filters on abdominal computed tomography to assist healthcare providers with the appropriate management of these devices to decrease complications: one based on 2-dimensional data and transfer learning (2D + TL) and an augmented version of the same algorithm which accounts for the 3-dimensional information leveraging recurrent convolutional neural networks (3D + RCNN). The study contains 2048 abdominal computed tomography studies obtained from 439 patients who underwent IVC filter placement during the 10-year period from January 1st, 2009, to January 1st, 2019. Among these, 399 patients had retrievable filters, and 40 had non-retrievable filter types. The reference annotations for the filter location were obtained through a custom-developed interface. The ground truth annotations for the filter types were determined based on the electronic medical record and physician review of imaging. The initial stage of the framework returns a list of locations containing metallic objects based on the density of the structure. The second stage processes the candidate locations and determines which one contains an IVC filter. The final stage of the pipeline classifies the filter types as retrievable vs. non-retrievable. The computational models are trained using Tensorflow Keras API on an Nvidia Quadro GV100 system. We utilized a fine-tuning supervised training strategy to conduct our experiments. We find that the system achieves high sensitivity on detecting the filter locations with a high confidence value. The 2D + TL model achieved a sensitivity of 0.911 and a precision of 0.804, and the 3D + RCNN model achieved a sensitivity of 0.923 and a precision of 0.853 for filter detection. The system confidence for the IVC location predictions is high: 0.993 for 2D + TL and 0.996 for 3D + RCNN. The filter type prediction component of the system achieved 0.945 sensitivity, 0.882 specificity, and 0.97 AUC score with 2D + TL and 0. 940 sensitivity, 0.927 specificity, and 0.975 AUC score with 3D + RCNN. With the intent to create tools to improve patient outcomes, this study describes the initial phase of a computational framework to support healthcare providers in detecting patients with retained IVC filters, so an individualized decision can be made to remove these devices when appropriate, to decrease complications. To our knowledge, this is the first study that curates abdominal computed tomography (CT) scans and presents an algorithm for automated detection and characterization of IVC filters.

Cardiovascular disease (CVD) related mortality and morbidity heavily strain society. The relationship between external risk factors and our genetics have not been well established. It is widely acknowledged that environmental influence and individual behaviours play a significant role in CVD vulnerability, leading to the development of polygenic risk scores (PRS). We employed the PRISMA search method to locate pertinent research and literature to extensively review artificial intelligence (AI)-based PRS models for CVD risk prediction. Furthermore, we analyzed and compared conventional vs. AI-based solutions for PRS. We summarized the recent advances in our understanding of the use of AI-based PRS for risk prediction of CVD. Our study proposes three hypotheses: i) Multiple genetic variations and risk factors can be incorporated into AI-based PRS to improve the accuracy of CVD risk predicting. ii) AI-based PRS for CVD circumvents the drawbacks of conventional PRS calculators by incorporating a larger variety of genetic and non-genetic components, allowing for more precise and individualised risk estimations. iii) Using AI approaches, it is possible to significantly reduce the dimensionality of huge genomic datasets, resulting in more accurate and effective disease risk prediction models. Our study highlighted that the AI-PRS model outperformed traditional PRS calculators in predicting CVD risk. Furthermore, using AI-based methods to calculate PRS may increase the precision of risk predictions for CVD and have significant ramifications for individualized prevention and treatment plans.

In recent years, there has been a significant increase in the development of artificial intelligence (AI) algorithms aimed at reviewing radiographs after total joint arthroplasty (TJA). This disruptive technology is particularly promising in the context of preoperative planning for revision TJA. Yet, the efficacy of AI algorithms regarding TJA implant analysis has not been examined comprehensively.

PubMed, EBSCO, and Google Scholar electronic databases were utilized to identify all studies evaluating AI algorithms related to TJA implant analysis between 1 January 2000, and 27 February 2023 (PROSPERO study protocol registration: CRD42023403497). The mean methodological index for non-randomized studies score was 20.4 ± 0.6. We reported the accuracy, sensitivity, specificity, positive predictive value, and area under the curve (AUC) for the performance of each outcome measure.

Our initial search yielded 374 articles, and a total of 20 studies with three main use cases were included. Sixteen studies analyzed implant identification, two addressed implant failure, and two addressed implant measurements. Each use case had a median AUC and accuracy above 0.90 and 90%, respectively, indicative of a well-performing AI algorithm. Most studies failed to include explainability methods and conduct external validity testing.

These findings highlight the promising role of AI in recognizing implants in TJA. Preliminary studies have shown strong performance in implant identification, implant failure, and accurately measuring implant dimensions. Future research should follow a standardized guideline to develop and train models and place a strong emphasis on transparency and clarity in reporting results.

Level III.

Daphnia magna is one species of water flea that has been used for a long time for ecotoxicity studies. In addition, Daphnia has a myogenic heart that is very useful for cardiotoxicity studies. Previous attempts to calculate the cardiac parameter endpoints in Daphnia suffer from the drawback of tedious operation and high variation due to manual counting errors. Even the previous method that utilized deep learning to help the process suffer from either overestimation of parameters or the need for specialized equipment to perform the analysis. In this study, we utilized DeepLabCut software previously used for animal pose tracking and demonstrated that ResNet_152 was the best fit for training the network. The trained network also showed comparable results with ImageJ and Kymograph, which was mostly done manually. In addition to that, several macro scripts in either Excel or Python format were developed to help summarize the data for faster analysis. The trained network was then challenged to analyze the potential cardiotoxicity of imidacloprid and pendimethalin in D. magna, and it showed that both pesticides cause alteration in their cardiac performance. Overall, this method provides a simple and automatic method to analyze the cardiac performance of Daphnia by utilizing DeepLabCut. The method proposed in this paper can contribute greatly to scientists conducting fast and accurate cardiotoxicity measurements when using Daphnia as a model.

Training deep Convolutional Neural Networks (CNNs) presents challenges in terms of memory requirements and computational resources, often resulting in issues such as model overfitting and lack of generalization. These challenges can only be mitigated by using an excessive number of training images. However, medical image datasets commonly suffer from data scarcity due to the complexities involved in their acquisition, preparation, and curation. To address this issue, we propose a compact and hybrid machine learning architecture based on the Morphological and Convolutional Neural Network (MCNN), followed by a Random Forest classifier. Unlike deep CNN architectures, the MCNN was specifically designed to achieve effective performance with medical image datasets limited to a few hundred samples. It incorporates various morphological operations into a single layer and uses independent neural networks to extract information from each signal channel. The final classification is obtained by utilizing a Random Forest classifier on the outputs of the last neural network layer. We compare the classification performance of our proposed method with three popular deep CNN architectures (ResNet-18, ShuffleNet-V2, and MobileNet-V2) using two training approaches: full training and transfer learning. The evaluation was conducted on two distinct medical image datasets: the ISIC dataset for melanoma classification and the ORIGA dataset for glaucoma classification. Results demonstrate that the MCNN method exhibits reliable performance in melanoma classification, achieving an AUC of 0.94 (95% CI: 0.91 to 0.97), outperforming the popular CNN architectures. For the glaucoma dataset, the MCNN achieved an AUC of 0.65 (95% CI: 0.53 to 0.74), which was similar to the performance of the popular CNN architectures. This study contributes to the understanding of mathematical morphology in shallow neural networks for medical image classification and highlights the potential of hybrid architectures in effectively learning from medical image datasets that are limited by a small number of case samples.

Image segmentation is an important process for quantifying characteristics of malignant bone lesions, but this task is challenging and laborious for radiologists. Deep learning has shown promise in automating image segmentation in radiology, including for malignant bone lesions. The purpose of this review is to investigate deep learning-based image segmentation methods for malignant bone lesions on Computed Tomography (CT), Magnetic Resonance Imaging (MRI), and Positron-Emission Tomography/CT (PET/CT).

The literature search of deep learning-based image segmentation of malignant bony lesions on CT and MRI was conducted in PubMed, Embase, Web of Science, and Scopus electronic databases following the guidelines of Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). A total of 41 original articles published between February 2017 and March 2023 were included in the review.

The majority of papers studied MRI, followed by CT, PET/CT, and PET/MRI. There was relatively even distribution of papers studying primary vs. secondary malignancies, as well as utilizing 3-dimensional vs. 2-dimensional data. Many papers utilize custom built models as a modification or variation of U-Net. The most common metric for evaluation was the dice similarity coefficient (DSC). Most models achieved a DSC above 0.6, with medians for all imaging modalities between 0.85-0.9.

Deep learning methods show promising ability to segment malignant osseous lesions on CT, MRI, and PET/CT. Some strategies which are commonly applied to help improve performance include data augmentation, utilization of large public datasets, preprocessing including denoising and cropping, and U-Net architecture modification. Future directions include overcoming dataset and annotation homogeneity and generalizing for clinical applicability.

Differentiating benign renal oncocytic tumors and malignant renal cell carcinoma (RCC) on imaging and histopathology is a critical problem that presents an everyday clinical challenge. This manuscript aims to demonstrate a novel methodology integrating metabolomics with radiomics features (RF) to differentiate between benign oncocytic neoplasia and malignant renal tumors. For this purpose, thirty-three renal tumors (14 renal oncocytic tumors and 19 RCC) were prospectively collected and histopathologically characterised. Matrix-assisted laser desorption/ionisation mass spectrometry imaging (MALDI-MSI) was used to extract metabolomics data, while RF were extracted from CT scans of the same tumors. Statistical integration was used to generate multilevel network communities of -omics features. Metabolites and RF critical for the differentiation between the two groups (delta centrality > 0.1) were used for pathway enrichment analysis and machine learning classifier (XGboost) development. Receiver operating characteristics (ROC) curves and areas under the curve (AUC) were used to assess classifier performance. Radiometabolomics analysis demonstrated differential network node configuration between benign and malignant renal tumors. Fourteen nodes (6 RF and 8 metabolites) were crucial in distinguishing between the two groups. The combined radiometabolomics model achieved an AUC of 86.4%, whereas metabolomics-only and radiomics-only classifiers achieved AUC of 72.7% and 68.2%, respectively. Analysis of significant metabolite nodes identified three distinct tumour clusters (malignant, benign, and mixed) and differentially enriched metabolic pathways. In conclusion, radiometabolomics integration has been presented as an approach to evaluate disease entities. In our case study, the method identified RF and metabolites important in differentiating between benign oncocytic neoplasia and malignant renal tumors, highlighting pathways differentially expressed between the two groups. Key metabolites and RF identified by radiometabolomics can be used to improve the identification and differentiation between renal neoplasms.

In this study, we present a self-supervised learning (SSL)-based model that enables anatomical structure-based unsupervised anomaly detection (UAD). The model employs an anatomy-aware pasting (AnatPaste) augmentation tool that uses a threshold-based lung segmentation pretext task to create anomalies in normal chest radiographs used for model pretraining. These anomalies are similar to real anomalies and help the model recognize them. We evaluate our model using three open-source chest radiograph datasets. Our model exhibits area under curves of 92.1%, 78.7%, and 81.9%, which are the highest among those of existing UAD models. To the best of our knowledge, this is the first SSL model to employ anatomical information from segmentation as a pretext task. The performance of AnatPaste shows that incorporating anatomical information into SSL can effectively improve accuracy.

precision-medicine quantitative tools for cross-sectional imaging require painstaking labeling of targets that vary considerably in volume, prohibiting scaling of data annotation efforts and supervised training to large datasets for robust and generalizable clinical performance. A straight-forward time-saving strategy involves manual editing of AI-generated labels, which we call AI-collaborative labeling (AICL). Factors affecting the efficacy and utility of such an approach are unknown. Reduction in time effort is not well documented. Further, edited AI labels may be prone to automation bias.

In this pilot, using a cohort of CTs with intracavitary hemorrhage, we evaluate both time savings and AICL label quality and propose criteria that must be met for using AICL annotations as a high-throughput, high-quality ground truth.

57 CT scans of patients with traumatic intracavitary hemorrhage were included. No participant recruited for this study had previously interpreted the scans. nnU-net models trained on small existing datasets for each feature (hemothorax/hemoperitoneum/pelvic hematoma; n = 77-253) were used in inference. Two common scenarios served as baseline comparison- de novo expert manual labeling, and expert edits of trained staff labels. Parameters included time effort and image quality graded by a blinded independent expert using a 9-point scale. The observer also attempted to discriminate AICL and expert labels in a random subset (n = 18). Data were compared with ANOVA and post-hoc paired signed rank tests with Bonferroni correction.

AICL reduced time effort 2.8-fold compared to staff label editing, and 8.7-fold compared to expert labeling (corrected p < 0.0006). Mean Likert grades for AICL (8.4, SD:0.6) were significantly higher than for expert labels (7.8, SD:0.9) and edited staff labels (7.7, SD:0.8) (corrected p < 0.0006). The independent observer failed to correctly discriminate AI and human labels.

For our use case and annotators, AICL facilitates rapid large-scale curation of high-quality ground truth. The proposed quality control regime can be employed by other investigators prior to embarking on AICL for segmentation tasks in large datasets.

Machine learning (ML) for diagnosis of thyroid nodules on ultrasound is an active area of research. However, ML tools require large, well-labeled datasets, the curation of which is time-consuming and labor-intensive. The purpose of our study was to develop and test a deep-learning-based tool to facilitate and automate the data annotation process for thyroid nodules; we named our tool Multistep Automated Data Labelling Procedure (MADLaP). MADLaP was designed to take multiple inputs including pathology reports, ultrasound images, and radiology reports. Using multiple step-wise 'modules' including rule-based natural language processing, deep-learning-based imaging segmentation, and optical character recognition, MADLaP automatically identified images of a specific thyroid nodule and correctly assigned a pathology label. The model was developed using a training set of 378 patients across our health system and tested on a separate set of 93 patients. Ground truths for both sets were selected by an experienced radiologist. Performance metrics including yield (how many labeled images the model produced) and accuracy (percentage correct) were measured using the test set. MADLaP achieved a yield of 63 % and an accuracy of 83 %. The yield progressively increased as the input data moved through each module, while accuracy peaked part way through. Error analysis showed that inputs from certain examination sites had lower accuracy (40 %) than the other sites (90 %, 100 %). MADLaP successfully created curated datasets of labeled ultrasound images of thyroid nodules. While accurate, the relatively suboptimal yield of MADLaP exposed some challenges when trying to automatically label radiology images from heterogeneous sources. The complex task of image curation and annotation could be automated, allowing for enrichment of larger datasets for use in machine learning development.