Kunal Sukhija
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Kunal Sukhija has a total of 20 co-authors. Here are the top co-authors by number of shared publications:
- Mihir Tandon
Albany Medical College, Albany, NY, United States
7 shared publications - Nitin Chetla
University of Virginia School of Medicine, Charlottesville, VA, United States
7 shared publications - Sai Samayamanthula
University of Virginia School of Medicine, Charlottesville, VA, United States
5 shared publications - Adarsh Mallepally
Virginia Commonwealth University School of Medicine, Richmond, VA, USA
4 shared publications - Botan Zebari
St. James School of Medicine, Binghamton, NY, United States
4 shared publications - John Chen
Albany Medical College, Albany, NY, United States
4 shared publications
Latest Publications
- Open Access
- 12 September 2025
- Published by JMIR Publications Inc. in JMIR Biomedical Engineering
- Vol. 10 (1) , e82980
Correction: Can Artificial Intelligence Diagnose Knee Osteoarthritis?
[This corrects the article DOI: 10.2196/67481.].
- 12 September 2025
- Published by JMIR Publications Inc. in JMIR Biomedical Engineering
- Vol. 10 (1) , e82980
Open AccessCorrection: Can Artificial Intelligence Diagnose Knee Osteoarthritis?
[This corrects the article DOI: 10.2196/67481.].
- Open Access
- 25 August 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Correction: Can Artificial Intelligence Diagnose Knee Osteoarthritis? (Preprint)
Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author).
- 25 August 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessCorrection: Can Artificial Intelligence Diagnose Knee Osteoarthritis? (Preprint)
Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author).
- Open Access
- 24 June 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Sparse Mixture‑of‑Experts Transformer Performance for Initial MRI Alzheimer's Detection (Preprint)
UNSTRUCTURED While current diagnostic methods for Alzheimer’s disease (AD) rely on cognitive assessments, such as the Mini-Mental State Examination (MMSE) and Clinical Dementia Rating (CDR), this study evaluates the performance of a sparse Mixture-of-Experts (MoE) multimodal transformer model in detecting probable AD from structural MRI data. The model API was queried iteratively with standardized prompts to classify scans as “probable AD present” (A) or “absent” (B). Performance metrics (accuracy, precision, recall, F1-score) were calculated against clinical benchmarks (CDR/MMSE). The MoE achieved 67% accuracy (precision: 64% for AD, 71% for non-AD; recall: 77% for AD, 57% for non-AD). The 23% false‑negative and 30% false‑positive rates reflect prompt sensitivity and dataset homogeneity of predominantly mild AD (CDR=1). The sparse MoE architecture shows promise as a tool for prioritizing suspected AD cases, offering better computational efficiency than convolutional neural networks (CNNs). However, performance inconsistencies underscore the need for larger, diverse datasets (e.g., CSF tau, APOE ε4 status, amyloid-PET, plasma biomarkers) to improve clinical reliability.
- 24 June 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessSparse Mixture‑of‑Experts Transformer Performance for Initial MRI Alzheimer's Detection (Preprint)
UNSTRUCTURED While current diagnostic methods for Alzheimer’s disease (AD) rely on cognitive assessments, such as the Mini-Mental State Examination (MMSE) and Clinical Dementia Rating (CDR), this study evaluates the performance of a sparse Mixture-of-Experts (MoE) multimodal transformer model in detecting probable AD from structural MRI data. The model API was queried iteratively with standardized prompts to classify scans as “probable AD present” (A) or “absent” (B). Performance metrics (accuracy, precision, recall, F1-score) were calculated against clinical benchmarks (CDR/MMSE). The MoE achieved 67% accuracy (precision: 64% for AD, 71% for non-AD; recall: 77% for AD, 57% for non-AD). The 23% false‑negative and 30% false‑positive rates reflect prompt sensitivity and dataset homogeneity of predominantly mild AD (CDR=1). The sparse MoE architecture shows promise as a tool for prioritizing suspected AD cases, offering better computational efficiency than convolutional neural networks (CNNs). However, performance inconsistencies underscore the need for larger, diverse datasets (e.g., CSF tau, APOE ε4 status, amyloid-PET, plasma biomarkers) to improve clinical reliability.
- Open Access
- 23 April 2025
- Published by JMIR Publications Inc. in JMIR Biomedical Engineering
- Vol. 10, e67481
Can Artificial Intelligence Diagnose Knee Osteoarthritis?
This study analyzed the capability of GPT-4o to properly identify knee osteoarthritis and found that the model had good sensitivity but poor specificity in identifying knee osteoarthritis; patients and clinicians should practice caution when using GPT-4o for image analysis in knee osteoarthritis.
- 23 April 2025
- Published by JMIR Publications Inc. in JMIR Biomedical Engineering
- Vol. 10, e67481
Open AccessCan Artificial Intelligence Diagnose Knee Osteoarthritis?
This study analyzed the capability of GPT-4o to properly identify knee osteoarthritis and found that the model had good sensitivity but poor specificity in identifying knee osteoarthritis; patients and clinicians should practice caution when using GPT-4o for image analysis in knee osteoarthritis.
- Open Access
- 10 January 2025
- Published by JMIR Publications Inc. in JMIR AI
- Vol. 4 (1) , e67621
Evaluating ChatGPT’s Efficacy in Pediatric Pneumonia Detection From Chest X-Rays: Comparative Analysis of Specialized AI Models
- 10 January 2025
- Published by JMIR Publications Inc. in JMIR AI
- Vol. 4 (1) , e67621
Open AccessEvaluating ChatGPT’s Efficacy in Pediatric Pneumonia Detection From Chest X-Rays: Comparative Analysis of Specialized AI Models
- Open Access
- 16 October 2024
- Published by JMIR Publications Inc. in JMIR Preprints
Evaluating ChatGPT’s Efficacy in Pediatric Pneumonia Detection From Chest X-Rays: Comparative Analysis of Specialized AI Models (Preprint)
Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author). Background: Machine learning and artificial intelligence have made large impacts in the field of medicine, especially the study of radiology. One of these impacts include the potential ability to help diagnose disease in patients using machine-learning image recognition on patient diagnostic images. The importance of AI in medical imaging can be seen through its use to increase radiology efficiency, reduce diagnostic errors, or even as far as filling in the role of a radiologist when one is not available. Objective: Machine learning and artificial intelligence (AI) have indeed shown promise in medicine, particularly in radiology. While AI, especially through image recognition, has demonstrated the ability to assist in diagnosing diseases from patient diagnostic images, its reliability and long-term role are far from certain. The goal of this study is to understand its diagnostic capability and reliability in pediatric pneumonia chest x-rays. Methods: To better understand how effective Artificial Intelligence would be at diagnostic imaging, ChatGPT was asked to categorize 1000 pediatric lung x-ray images into either normal or pneumonia groups and measure the accuracy based on confirmed diagnosis. The study did not require Institutional Review Board (IRB) approval because it did not involve patient consent or direct interaction with individuals. Results: The results show that while ChatGPT-4 Turbo and ChatGPT-4o have a high specificity rate and sensitivity rate, respectively, the accuracy of the LLM is poor compared to a pneumonia-specific machine-learning algorithm. Conclusions: ChatGPT-4 has limitations when being used for diagnosing pneumonia from chest X-ray radiographs as shown by this research. The model's strong bias towards a non-pneumonia diagnosis, limited ability to distinguish between the two classes, need for extensive prompt engineering, and lack of specialized medical knowledge suggest that it may not be suitable for clinical use in its current form.
- 16 October 2024
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessEvaluating ChatGPT’s Efficacy in Pediatric Pneumonia Detection From Chest X-Rays: Comparative Analysis of Specialized AI Models (Preprint)
Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author). Background: Machine learning and artificial intelligence have made large impacts in the field of medicine, especially the study of radiology. One of these impacts include the potential ability to help diagnose disease in patients using machine-learning image recognition on patient diagnostic images. The importance of AI in medical imaging can be seen through its use to increase radiology efficiency, reduce diagnostic errors, or even as far as filling in the role of a radiologist when one is not available. Objective: Machine learning and artificial intelligence (AI) have indeed shown promise in medicine, particularly in radiology. While AI, especially through image recognition, has demonstrated the ability to assist in diagnosing diseases from patient diagnostic images, its reliability and long-term role are far from certain. The goal of this study is to understand its diagnostic capability and reliability in pediatric pneumonia chest x-rays. Methods: To better understand how effective Artificial Intelligence would be at diagnostic imaging, ChatGPT was asked to categorize 1000 pediatric lung x-ray images into either normal or pneumonia groups and measure the accuracy based on confirmed diagnosis. The study did not require Institutional Review Board (IRB) approval because it did not involve patient consent or direct interaction with individuals. Results: The results show that while ChatGPT-4 Turbo and ChatGPT-4o have a high specificity rate and sensitivity rate, respectively, the accuracy of the LLM is poor compared to a pneumonia-specific machine-learning algorithm. Conclusions: ChatGPT-4 has limitations when being used for diagnosing pneumonia from chest X-ray radiographs as shown by this research. The model's strong bias towards a non-pneumonia diagnosis, limited ability to distinguish between the two classes, need for extensive prompt engineering, and lack of specialized medical knowledge suggest that it may not be suitable for clinical use in its current form.
- Open Access
- 12 October 2024
- Published by JMIR Publications Inc. in JMIR Preprints
Can Artificial Intelligence Diagnose Knee Osteoarthritis? (Preprint)
UNSTRUCTURED This study analyzed the capability of GPT-4o to properly identify knee osteoarthritis and found that the model had good sensitivity but poor specificity in identifying knee osteoarthritis; patients and clinicians should practice caution when using GPT-4o for image analysis in knee osteoarthritis.
- 12 October 2024
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessCan Artificial Intelligence Diagnose Knee Osteoarthritis? (Preprint)
UNSTRUCTURED This study analyzed the capability of GPT-4o to properly identify knee osteoarthritis and found that the model had good sensitivity but poor specificity in identifying knee osteoarthritis; patients and clinicians should practice caution when using GPT-4o for image analysis in knee osteoarthritis.