Mohammad Beheshti
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Mohammad Beheshti has a total of 23 co-authors. Here are the top co-authors by number of shared publications:
- Iris Zachary
Missouri Cancer Registry and Research Center, Department of Public Health, University of Missouri, Columbia, USA
2 shared publications - Suzanne Austin Boren
MU Institute for Data Science and Informatics, University of Missouri, Columbia, MO 65211, USA
2 shared publications - Ali Sharifi Kia
Insight Lab, Department of Computer Science, Western University, London, ON N6A 3K7, Canada
1 shared publication - Amir Erfan Zareei Shams Abadi
MU Institute for Data Science and Informatics, University of Missouri, Columbia, MO 65211, USA
1 shared publication Brandon Henslee
Division of Urology, Department of Surgery, University of Missouri at Columbia, Columbia, Missouri
1 shared publication- Butros M. Dahu
Institute for Data Science and Informatics, University of Missouri, Columbia, MO 65211, USA
1 shared publication
Latest Publications
- Open Access
- 14 November 2026
- Published by Springer Nature in BMC Medical Informatics and Decision Making
Leveraging open-weight language models for automated patient record linkage
- 14 November 2026
- Published by Springer Nature in BMC Medical Informatics and Decision Making
Open AccessLeveraging open-weight language models for automated patient record linkage
- Open Access
- 18 September 2026
- Published by MDPI AG in Information
- Vol. 17 (9) , 917
Entity Resolution Using Transformer-Based Language Models: A Systematic Scoping Review
Entity resolution (ER) is fundamental to integrating heterogeneous data, which traditional approaches address through deterministic rule-based methods and probabilistic record linkage. We conducted a systematic scoping review following PRISMA-ScR to characterize the use of transformer-based language models for ER. Five databases were searched, and 155 studies were included for synthesis. The literature expanded sharply after 2023, with 55% of included studies published in 2025 or 2026. General-purpose matching was the most common entity focus, followed by product/e-commerce. Healthcare applications were especially scarce, with only one study applying ER to the patient/healthcare domain. Among studies that performed blocking, embedding-based similarity was most common, followed by string-based approaches. Classification-head approaches remained the most common matching approach, followed by prompt-based approaches. Encoder-only models remained the most widely evaluated architecture, while decoder-only models grew increasingly prominent from 2023 onward. Full-parameter fine-tuning was the predominant learning strategy, followed by zero-shot and few-shot prompting. Among studies classified as general-purpose, nearly one-third were evaluated on only one or two entity types, limiting their generalizability. Reported best F1 scores varied across benchmarks, with no consistent advantage for encoder-only versus decoder-only architectures. These findings support the need for more diverse, end-to-end evaluations that consider efficiency and robustness alongside accuracy.
- 18 September 2026
- Published by MDPI AG in Information
- Vol. 17 (9) , 917
Open AccessEntity Resolution Using Transformer-Based Language Models: A Systematic Scoping Review
Entity resolution (ER) is fundamental to integrating heterogeneous data, which traditional approaches address through deterministic rule-based methods and probabilistic record linkage. We conducted a systematic scoping review following PRISMA-ScR to characterize the use of transformer-based language models for ER. Five databases were searched, and 155 studies were included for synthesis. The literature expanded sharply after 2023, with 55% of included studies published in 2025 or 2026. General-purpose matching was the most common entity focus, followed by product/e-commerce. Healthcare applications were especially scarce, with only one study applying ER to the patient/healthcare domain. Among studies that performed blocking, embedding-based similarity was most common, followed by string-based approaches. Classification-head approaches remained the most common matching approach, followed by prompt-based approaches. Encoder-only models remained the most widely evaluated architecture, while decoder-only models grew increasingly prominent from 2023 onward. Full-parameter fine-tuning was the predominant learning strategy, followed by zero-shot and few-shot prompting. Among studies classified as general-purpose, nearly one-third were evaluated on only one or two entity types, limiting their generalizability. Reported best F1 scores varied across benchmarks, with no consistent advantage for encoder-only versus decoder-only architectures. These findings support the need for more diverse, end-to-end evaluations that consider efficiency and robustness alongside accuracy.
- Top Cited PapersOpen Access
- 17 January 2025
- Published by MDPI AG in Informatics
- Vol. 12 (1) , 9
Evaluating the Reliability of ChatGPT for Health-Related Questions: A Systematic Review
The rapid advancement of large language models like ChatGPT has significantly impacted natural language processing, expanding its applications across various fields, including healthcare. However, there remains a significant gap in understanding the consistency and reliability of ChatGPT’s performance across different medical domains. We conducted this systematic review according to an LLM-assisted PRISMA setup. The high-recall search term “ChatGPT” yielded 1101 articles from 2023 onwards. Through a dual-phase screening process, initially automated via ChatGPT and subsequently manually by human reviewers, 128 studies were included. The studies covered a range of medical specialties, focusing on diagnosis, disease management, and patient education. The assessment metrics varied, but most studies compared ChatGPT’s accuracy against evaluations by clinicians or reliable references. In several areas, ChatGPT demonstrated high accuracy, underscoring its effectiveness. However, performance varied, and some contexts revealed lower accuracy. The mixed outcomes across different medical domains emphasize the challenges and opportunities of integrating AI like ChatGPT into healthcare. The high accuracy in certain areas suggests that ChatGPT has substantial utility, yet the inconsistent performance across all applications indicates a need for ongoing evaluation and refinement. This review highlights ChatGPT’s potential to improve healthcare delivery alongside the necessity for continued research to ensure its reliability.
- 17 January 2025
- Published by MDPI AG in Informatics
- Vol. 12 (1) , 9
Top Cited PapersOpen AccessEvaluating the Reliability of ChatGPT for Health-Related Questions: A Systematic Review
The rapid advancement of large language models like ChatGPT has significantly impacted natural language processing, expanding its applications across various fields, including healthcare. However, there remains a significant gap in understanding the consistency and reliability of ChatGPT’s performance across different medical domains. We conducted this systematic review according to an LLM-assisted PRISMA setup. The high-recall search term “ChatGPT” yielded 1101 articles from 2023 onwards. Through a dual-phase screening process, initially automated via ChatGPT and subsequently manually by human reviewers, 128 studies were included. The studies covered a range of medical specialties, focusing on diagnosis, disease management, and patient education. The assessment metrics varied, but most studies compared ChatGPT’s accuracy against evaluations by clinicians or reliable references. In several areas, ChatGPT demonstrated high accuracy, underscoring its effectiveness. However, performance varied, and some contexts revealed lower accuracy. The mixed outcomes across different medical domains emphasize the challenges and opportunities of integrating AI like ChatGPT into healthcare. The high accuracy in certain areas suggests that ChatGPT has substantial utility, yet the inconsistent performance across all applications indicates a need for ongoing evaluation and refinement. This review highlights ChatGPT’s potential to improve healthcare delivery alongside the necessity for continued research to ensure its reliability.
- Open Access
- 1 September 2024
- Published by Wolters Kluwer Health in JU Open Plus
- Vol. 2 (9)
Transition in Quality Of Life Among Older Patients With Bladder Cancer and Its Association With Treatment: A Clustering Approach
Purpose: To classify older patients with bladder cancer (BC) based on quality-of-life (QoL) measures and explore the clusters transitions over a 2-year period and patient characteristics, BC stages, and treatment. Patients with BC report a significant decline in functional status and QoL. The association of this transition and patient characteristics and treatment over time is not known. We aim to explore this association. Methods: Data were obtained from patients with BC in the SEER-MHOS database (1998-2021). Unsupervised cluster analysis was employed to identify homogeneous clusters of patients with similar functionality based on age, physical and mental component score, and ability to perform activities of daily living. Univariate and multivariable analyses were performed to assess the association of patient characteristics and treatment on cluster transitions over a 2-year interval. Results: The final cohort consisted of 2934 (age 65 or older) older patients with BC. We identified 3 distinct clusters: 67.9% in Cluster 1; 25.4% in Cluster 2; and 6.7% in Cluster 3. After 2 years, 11.7% patients showed improvement and 14% experienced a decline. Cancer/treatment characteristics were not associated with cluster transition. Younger age and being married were identified as factors associated with improvement. Older age, non–homeownership, and last-year fall were associated with transition to less healthy cluster. Conclusions: Over a 2-year interval after BC diagnosis, a quarter of patients will have substantial changes in QoL. No significant association was observed between cluster transition and BC characteristics. Age as well as marital, fall, and homeowner status do impact this transition.
- 1 September 2024
- Published by Wolters Kluwer Health in JU Open Plus
- Vol. 2 (9)
Open AccessTransition in Quality Of Life Among Older Patients With Bladder Cancer and Its Association With Treatment: A Clustering Approach
Purpose: To classify older patients with bladder cancer (BC) based on quality-of-life (QoL) measures and explore the clusters transitions over a 2-year period and patient characteristics, BC stages, and treatment. Patients with BC report a significant decline in functional status and QoL. The association of this transition and patient characteristics and treatment over time is not known. We aim to explore this association. Methods: Data were obtained from patients with BC in the SEER-MHOS database (1998-2021). Unsupervised cluster analysis was employed to identify homogeneous clusters of patients with similar functionality based on age, physical and mental component score, and ability to perform activities of daily living. Univariate and multivariable analyses were performed to assess the association of patient characteristics and treatment on cluster transitions over a 2-year interval. Results: The final cohort consisted of 2934 (age 65 or older) older patients with BC. We identified 3 distinct clusters: 67.9% in Cluster 1; 25.4% in Cluster 2; and 6.7% in Cluster 3. After 2 years, 11.7% patients showed improvement and 14% experienced a decline. Cancer/treatment characteristics were not associated with cluster transition. Younger age and being married were identified as factors associated with improvement. Older age, non–homeownership, and last-year fall were associated with transition to less healthy cluster. Conclusions: Over a 2-year interval after BC diagnosis, a quarter of patients will have substantial changes in QoL. No significant association was observed between cluster transition and BC characteristics. Age as well as marital, fall, and homeowner status do impact this transition.
- Open Access
- 8 January 2022
- Published by Skyler Publication in Frontiers in Health Informatics
- Vol. 11 (1) , 120
Health Information Systems Evaluation Criteria: Overview of Systematic Reviews
Introduction: Health information systems play an important role in improving the quality of patient care and patient safety. to ensure their effectiveness and efficiency, they need to be evaluated. Although HIS evaluation has been investigated in many studies, there is no consensus on which aspects of HIS to evaluate. The aim of this study is to identify the indicators for the evaluation of health information systems and to provide an overview of the criteria devised and studies conducted.Methods: An umbrella review was performed exploring databases PubMed, Science direct, Web of Science, Science, and IEEE while following the PRISMA protocol. Articles were reviewed by two authors independently using the covidence tool to check the inclusion criteria and to extract the data items. Risk of bias was assessed using ROBIS and AMSTAR.Result: All included studies showed a high risk of bias according to ROBIS criteria. The extracted evaluation criteria were classified into 13 categories. Most of the studies believe that a more reliable and standardized tool is needed for the evaluation of health information systems. Two studies mentioned that surveys and questionnaires were the most commonly used method for evaluation of the systems. Summative evaluation was the most used method in two studies and the least used method in another one study.Conclusion: All the included studies had high risk of bias. Accordingly, further research and evidence is needed in this field. Most of the studies highlighted the need for more reliable and standardized tools for evaluation of health information systems.
- 8 January 2022
- Published by Skyler Publication in Frontiers in Health Informatics
- Vol. 11 (1) , 120
Open AccessHealth Information Systems Evaluation Criteria: Overview of Systematic Reviews
Introduction: Health information systems play an important role in improving the quality of patient care and patient safety. to ensure their effectiveness and efficiency, they need to be evaluated. Although HIS evaluation has been investigated in many studies, there is no consensus on which aspects of HIS to evaluate. The aim of this study is to identify the indicators for the evaluation of health information systems and to provide an overview of the criteria devised and studies conducted.Methods: An umbrella review was performed exploring databases PubMed, Science direct, Web of Science, Science, and IEEE while following the PRISMA protocol. Articles were reviewed by two authors independently using the covidence tool to check the inclusion criteria and to extract the data items. Risk of bias was assessed using ROBIS and AMSTAR.Result: All included studies showed a high risk of bias according to ROBIS criteria. The extracted evaluation criteria were classified into 13 categories. Most of the studies believe that a more reliable and standardized tool is needed for the evaluation of health information systems. Two studies mentioned that surveys and questionnaires were the most commonly used method for evaluation of the systems. Summative evaluation was the most used method in two studies and the least used method in another one study.Conclusion: All the included studies had high risk of bias. Accordingly, further research and evidence is needed in this field. Most of the studies highlighted the need for more reliable and standardized tools for evaluation of health information systems.