Search Articles

View query in Help articles search

Search Results (1 to 10 of 102 Results)

Download search results: CSV END BibTex RIS


Evaluating the Diagnostic Accuracy of ChatGPT-4 Omni and ChatGPT-4 Turbo in Identifying Melanoma: Comparative Study

Evaluating the Diagnostic Accuracy of ChatGPT-4 Omni and ChatGPT-4 Turbo in Identifying Melanoma: Comparative Study

The diagnostic accuracies of GPT-4 Turbo and GPT-4o were 0.546 (95% CI 0.515‐0.577) and 0.577 (95% CI 0.547‐0.608), respectively. There was no significant difference in accuracy between the two models (P=.10). GPT-4 Turbo demonstrated a sensitivity of 76.3%, specificity of 32.9%, and false-positive rate of 67.1% (Table 1).

Samantha S. Sattler, Nitin Chetla, Matthew Chen, Tamer Rajai Hage, Joseph Chang, William Young Guo, Jeremy Hugh

JMIR Dermatol 2025;8:e67551

Complete Blood Count and Monocyte Distribution Width–Based Machine Learning Algorithms for Sepsis Detection: Multicentric Development and External Validation Study

Complete Blood Count and Monocyte Distribution Width–Based Machine Learning Algorithms for Sepsis Detection: Multicentric Development and External Validation Study

Noteworthy, MDW received approval from the Food and Drug Administration and European Community In-Vitro Diagnostic Medical Device as an early sepsis indicator in adult patients in the ED. MDW has great potential to be used in clinical practice because it is timely and fruitfully measured without further samples, additional costs, or even a different request from clinicians. Thus, it represents a cost-effective tool for early sepsis screening.

Andrea Campagner, Luisa Agnello, Anna Carobene, Andrea Padoan, Fabio Del Ben, Massimo Locatelli, Mario Plebani, Agostino Ognibene, Maria Lorubbio, Elena De Vecchi, Andrea Cortegiani, Elisa Piva, Donatella Poz, Francesco Curcio, Federico Cabitza, Marcello Ciaccio

J Med Internet Res 2025;27:e55492

Real-World Insights Into Dementia Diagnosis Trajectory and Clinical Practice Patterns Unveiled by Natural Language Processing: Development and Usability Study

Real-World Insights Into Dementia Diagnosis Trajectory and Clinical Practice Patterns Unveiled by Natural Language Processing: Development and Usability Study

Reference 32: Automatic identification of heart failure diagnostic criteria, using text analysis of clinical processing annotation tool to facilitate phenotyping of cognitive status in electronic health records: diagnosticdiagnostic

Hunki Paek, Richard H Fortinsky, Kyeryoung Lee, Liang-Chin Huang, Yazeed S Maghaydah, George A Kuchel, Xiaoyan Wang

JMIR Aging 2025;8:e65221

Novel Procedures for Evaluating Autism Online in a Culturally Diverse Population of Children: Protocol for a Mixed Methods Pathway Development Study

Novel Procedures for Evaluating Autism Online in a Culturally Diverse Population of Children: Protocol for a Mixed Methods Pathway Development Study

Families in rural or underserved areas can now access diagnostic services more easily through remote assessments [11]. In addition, telehealth can significantly reduce wait times by enabling quicker initial screenings and follow-ups, which is especially beneficial given the high demand for autism assessments and the limited number of specialists available [11]. Our recent systematic review found that Telehealth assessments are as accurate as in-person assessments, with over 80% diagnostic agreement [12].

Venus Mirzaei, Jeanne Wolstencroft, Georgia Lockwood Estrin, Eleanor Buckley, Shermina Sayani, Panos Katakis, Reena Anand, Tessa Squire, Eleanor Short, Paige Frankson, David Skuse, Michelle Heys

JMIR Res Protoc 2025;14:e55741

Detection of Alzheimer Disease in Neuroimages Using Vision Transformers: Systematic Review and Meta-Analysis

Detection of Alzheimer Disease in Neuroimages Using Vision Transformers: Systematic Review and Meta-Analysis

DL and ML approaches have been widely applied to various diagnostic tasks, although the diagnostic accuracy of DL remains uncertain in some areas. To address this gap, meta-analysis methods have been proposed [4] to provide a more comprehensive understanding. For example, studies have estimated the diagnostic accuracy of DL models for COVID-19 detection [5] and evaluated ML models for osteoporosis diagnosis in the hip bone [6].

Vivens Mubonanyikuzo, Hongjie Yan, Temitope Emmanuel Komolafe, Liang Zhou, Tao Wu, Nizhuan Wang

J Med Internet Res 2025;27:e62647

The Evolution of Uroflowmetry and Bladder Diary and the Emerging Trend of Using Home Devices From Hospital to Home

The Evolution of Uroflowmetry and Bladder Diary and the Emerging Trend of Using Home Devices From Hospital to Home

The diagnostic accuracy of uroflowmetry is largely affected by threshold values [12,13], especially with physiological compensatory processes, detrusor underactivity, or an underfilled bladder [14]. Although uroflowmetry can be used to monitor treatment outcomes [15] and to correlate symptoms with these objective findings [12,16], its clinical value is still limited, as it is unable to differentiate between the possible underlying mechanisms.

Ming-wei Li, Yao-Chou Tsai, Stephen Shei-Dei Yang, Yuan-Hung Pong, Yu-Ting Tsai, Vincent Fang-Sheng Tsai

Interact J Med Res 2025;14:e66694

Machine Learning and Deep Learning for Diagnosis of Lumbar Spinal Stenosis: Systematic Review and Meta-Analysis

Machine Learning and Deep Learning for Diagnosis of Lumbar Spinal Stenosis: Systematic Review and Meta-Analysis

A total of 2 reviewers (TW and NF) used the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2), which is a tool for assessing the quality of primary diagnostic accuracy studies, to independently assess the risk of bias for each eligible study [39]. The QUADAS-2 criteria assessed the risk of bias in 4 domains: patient selection, index test, reference standard, and flow and timing. Any disagreements were resolved by discussion with a third author.

Tianyi Wang, Ruiyuan Chen, Ning Fan, Lei Zang, Shuo Yuan, Peng Du, Qichao Wu, Aobo Wang, Jian Li, Xiaochuan Kong, Wenyi Zhu

J Med Internet Res 2024;26:e54676

Large Language Models in Gastroenterology: Systematic Review

Large Language Models in Gastroenterology: Systematic Review

This, pivotal in diagnosing and treating digestive tract diseases, faces challenges like diagnostic variability and labor-intensive documentation. Gastrointestinal endoscopy, traditionally reliant on the expertise of specialists to interpret complex visual data and execute precise interventions, can greatly benefit from the automation and analytical capabilities provided by LLMs [4,5].

Eun Jeong Gong, Chang Seok Bang, Jae Jun Lee, Jonghyung Park, Eunsil Kim, Subeen Kim, Minjae Kimm, Seoung-Ho Choi

J Med Internet Res 2024;26:e66648

Evaluating Bard Gemini Pro and GPT-4 Vision Against Student Performance in Medical Visual Question Answering: Comparative Case Study

Evaluating Bard Gemini Pro and GPT-4 Vision Against Student Performance in Medical Visual Question Answering: Comparative Case Study

With the growing dependence on diagnostic imaging, highlighted by a notable increase in imaging procedures in hospitals, the ability to understand and interpret these images is becoming increasingly important [8]. Additionally, artificial intelligence (AI) presents a valuable opportunity to enhance the training and learning experience of medical trainees [9].

Jonas Roos, Ron Martin, Robert Kaczmarczyk

JMIR Form Res 2024;8:e57592