A5043523524- Artificial Intelligence in Healthcare and Education
- Radiomics and Machine Learning in Medical Imaging
- COVID-19 diagnosis using AI
- Radiology practices and education
- Lung Cancer Diagnosis and Treatment
- Machine Learning in Healthcare
- Cardiac Imaging and Diagnostics
- Cardiac Valve Diseases and Treatments
- AI in cancer detection
- Phonocardiography and Auscultation Techniques
- Ultrasound in Clinical Applications
- Cerebrovascular and Carotid Artery Diseases
- Acute Ischemic Stroke Management
- Photoacoustic and Ultrasonic Imaging
- Infective Endocarditis Diagnosis and Management
- Advanced X-ray and CT Imaging
- Medical Imaging Techniques and Applications
- Explainable Artificial Intelligence (XAI)
- Fetal and Pediatric Neurological Disorders
- COVID-19 Clinical Research Studies
- Acute Myocardial Infarction Research
- Retinal Imaging and Analysis
- Atrial Fibrillation Management and Outcomes
- Glioma Diagnosis and Treatment
- Retinal Diseases and Treatments
Osaka Metropolitan University
2022-2025
Tokyo Metropolitan University
2023-2024
Osaka City University Hospital
2021-2022
Osaka City University
2021-2022
Chest radiography is a common and widely available examination. Although cardiovascular structures-such as cardiac shadows vessels-are visible on chest radiographs, the ability of these radiographs to estimate function valvular disease poorly understood. Using datasets from multiple institutions, we aimed develop validate deep-learning model simultaneously detect functions radiographs.
To compare the diagnostic accuracy of Generative Pre-trained Transformer (GPT)-4-based ChatGPT, GPT-4 with vision (GPT-4V) based and radiologists in musculoskeletal radiology.
Abstract Background GPT-4-based ChatGPT demonstrates significant potential in various industries; however, its clinical applications remain largely unexplored. Methods We employed the New England Journal of Medicine (NEJM) quiz “Image Challenge” from October 2021 to March 2023 assess ChatGPT's capabilities. The quiz, designed for healthcare professionals, tests ability analyze scenarios and make appropriate decisions. evaluated performance on NEJM analyzing accuracy rate by questioning type...
Large language models like GPT-4 have demonstrated potential for diagnosis in radiology. Previous studies investigating this primarily utilized quizzes from academic journals. This study aimed to assess the diagnostic capabilities of GPT-4-based Chat Generative Pre-trained Transformer (ChatGPT) using actual clinical radiology reports brain tumors and compare its performance with that neuroradiologists general radiologists.
We investigated the performance improvement of physicians with varying levels chest radiology experience when using a commercially available artificial intelligence (AI)-based computer-assisted detection (CAD) software to detect lung cancer nodules on radiographs from multiple vendors.
Background Carbon 11 (11C)–methionine is a useful PET radiotracer for the management of patients with glioma, but radiation exposure and lack molecular imaging facilities limit its use. Purpose To generate synthetic methionine images from contrast-enhanced (CE) MRI through an artificial intelligence (AI)–based image-to-image translation model to compare performance grading prognosis gliomas that real PET. Materials Methods An AI-based CE was developed validated who underwent both at...
Abstract Background GPT-4-based ChatGPT demonstrates significant potential in various industries; however, its clinical applications remain largely unexplored. Methods We employed the New England Journal of Medicine (NEJM) quiz “Image Challenge” from October 2021 to March 2023 assess ChatGPT’s capabilities. The quiz, designed for healthcare professionals, tests ability analyze scenarios and make appropriate decisions. evaluated performance on NEJM analyzing accuracy rate by questioning type...
Background Digital subtraction angiography (DSA) generates an image by subtracting a mask from dynamic angiogram. However, patient movement-caused misregistration artifacts can result in unclear DSA images that interrupt procedures. Purpose To train and to validate deep learning (DL)-based model produce DSA-like cerebral angiograms directly then quantitatively visually evaluate these for clinical usefulness. Materials Methods A retrospective development validation study was conducted on...
Chest radiographs are widely available and cost-effective; however, their usefulness as a biomarker of ageing using multi-institutional data remains underexplored. The aim this study was to develop from chest radiography examine the correlation between diseases.In retrospective, study, we trained, tuned, externally tested an artificial intelligence (AI) model estimate age healthy individuals biomarker. For modelling phase used consecutively collected May 22, 2008, Dec 28, 2021, three...
Abstract Objective To compare the diagnostic accuracy of Generative Pre-trained Transformer (GPT)-4 based ChatGPT, GPT-4 with vision (GPT-4V) and radiologists in musculoskeletal radiology. Materials Methods We included 106 “Test Yourself” cases from Skeletal Radiology between January 2014 September 2023. input medical history imaging findings into ChatGPT images GPT-4V then both generated a diagnosis for each case. Two (a radiology resident board-certified radiologist) independently provided...
This study created an image-to-image translation model that synthesizes diffusion tensor images (DTI) from conventional weighted images, and validated the similarities between original synthetic DTI. Thirty-two healthy volunteers were prospectively recruited. DTI DWI obtained with six three directions of motion probing gradient (MPG), respectively. The identical imaging plane was paired for synthesized one direction MPG DWI. process repeated times in respective directions. Regions interest...
Background: Large-scale radiographic datasets often include errors in labels such as body–part or projection, which can undermine automated image analysis. Purpose: To develop and externally validate two deep learning models—one for categorizing radiographs by body-part, another identifying projection rotation of chest radiographs—using large, diverse datasets. Methods: We retrospectively collected from multiple institutions public repositories. For the first model (Xp-Bodypart-Checker), we...
Abstract Aims We aimed to develop models detect aortic stenosis (AS) from chest radiographs—one of the most basic imaging tests—with artificial intelligence. Methods and results used 10 433 retrospectively collected digital radiographs 5638 patients train, validate, test three deep learning models. Chest were who had also undergone echocardiography at a single institution between July 2016 May 2019. These labelled corresponding assessments as AS-positive or AS-negative. The separated on...