Mert Zure

Artificial intelligence (AI) language models are increasingly being explored as tools to support medical education and clinical care. Evaluating their performance on valid and reliable assessments such as board certification exams may provide insight into their potential integration into real-world medical settings. This study evaluated the accuracy, consistency, and difficulty assessment of four advanced AI models using board-level geriatrics questions. Four AI models—Grok-3, ChatGPT-4o, Microsoft Copilot, and Google Gemini 2.0 Flash—were tested on 300 text-based multiple-choice questions from the BoardVitals geriatrics certification question bank. The questions were equally divided into easy, medium, and hard categories. Each model was asked to classify the question’s difficulty and provide an answer twice. Model responses were evaluated for accuracy, consistency between attempts, quality of explanations, and alignment with the difficulty ratings predefined by BoardVitals. GPT-4o demonstrated the highest overall accuracy (85.3%), followed by Grok-3 (82.0%), Copilot (78.7%), and Gemini (74.0%). All models performed best on easy questions, and showed a decrease in accuracy as the difficulty increased (p < 0.001). GPT-4o exhibited the highest consistency (96.3%), followed by Grok-3 (95.0%), Copilot (90.7%), and Gemini (81.3%). While their overall performance surpassed the average success rates of human users in the database, the agreement between model-assigned and reference difficulty ratings was moderate (mean κ = 0.41). GPT-4o received the highest mean quality score (4.68 ± 0.84), followed by Grok-3 (4.59 ± 0.98), Copilot (4.30 ± 1.07), and Gemini (3.88 ± 1.53). Advanced AI models demonstrate strong performance on geriatrics board-level content, suggesting potential applications as educational support tools. However, performance on multiple-choice examinations does not equate to clinical utility. Significant limitations include struggles with complex scenarios, difficulty in metacognitive assessment of question complexity, and variable explanation quality. These findings emphasize that AI integration into geriatric education and practice requires careful human oversight, explicit acknowledgment of limitations, and continued validation in diverse real-world contexts.

SUMMARY OBJECTIVE: Osteoporosis is a prevalent but often overlooked condition in hemiplegic patients that increases the risk of fracture. The aim of this study was to evaluate osteoporosis knowledge and awareness levels and determine associated factors in adult hemiplegic individuals. METHODS: A cross-sectional study was conducted with 88 hemiplegic participants (45 males, 43 females) in a tertiary rehabilitation clinic. Demographic and clinical data were documented. Cognitive function was assessed using the Mini-Mental State Examination, while osteoporosis knowledge was evaluated through the 31-item Osteoporosis Awareness Scale in addition to a structured question ("Have you ever heard of osteoporosis?"). Statistical analyses included independent t-tests, one-way analysis of variance with Tukey post-hoc testing, Pearson correlations, and multiple linear regression (significance: p<0.05). RESULTS: The mean Osteoporosis Awareness Scale score was 72.55±32.29 (range: 32–124). Seventy-two (81.8%) participants reported having heard of osteoporosis. In univariate analysis, Osteoporosis Awareness Scale scores were higher in females (76.58±33.81 vs. 68.73±30.52, p=0.036), postmenopausal women (78.92±33.45 vs. 58.40±26.31, p=0.042), those with higher education levels (p<0.001), and individuals with better cognitive function (p<0.001). Multiple linear regression confirmed that education level (p<0.001) and Mini-Mental State Examination score (p=0.002) were significant independent predictors of Osteoporosis Awareness Scale scores. Gender (p=0.052) and menopausal status (p=0.087) were not significant in the multivariate model. CONCLUSION: Education level and cognitive function were significant independent predictors of osteoporosis knowledge in hemiplegic patients. Gender and menopausal status were not significant after adjustment for education and cognitive function. These findings highlight the need for targeted educational strategies in rehabilitation settings, particularly for patients with lower educational backgrounds and cognitive impairment.

Objective: This study aims to examine and compare the effects of prolotherapy combined with exercise and exercise treatment alone on pain and functional status in individuals with lateral epicondylitis(LE). Methods: This retrospective cohort study included participants with chronic LE who were treated with prolotherapy combined with exercise treatment and exercise treatment alone. The participants were evaluated at baseline, immediately after treatment, and at the first, third, and sixth month after treatment using the visual analog scale (VAS) for pain levels; the Disabilities of the Arm, Shoulder, and Hand Questionnaire (DASH); and its work module (DASH-work) for disability levels. Results: A total of 20 participants were included in the exercise group (EG) and 19 in the prolotherapy with exercise group (PrEG). Disability levels, work-related disability, and pain levels significantly decreased over the follow-up period, irrespective of group allocation (DASH: p < 0.001, η ² = 0.56; DASH-work: p < 0.001, η ² = 0.46; VAS: p < 0.001, η ² = 0.64). There was a significant change in time by group interaction for DASH, DASH-work, and VAS scores ( p < 0.001) concerning PrEG and EG. Conclusions: Prolotherapy combined with exercise was more effective in reducing pain and general/work-related disability levels compared to exercise treatment alone. In the short term, better pain control and functionality were seen with prolotherapy, and these effects were sustained during long-term follow-up.

Background/Aims: Individuals increasingly turn to artificial intelligence (AI) chatbots for health-related information; however, the accuracy and usability of their responses remain uncertain. This study assessed the quality, comprehensiveness, and readability of responses from 6 AI chatbots—ChatGPT-3.5, ChatGPT-4o (OpenAI), Copilot AI (Microsoft), Perplexity AI (Perplexity.AI), Gemini AI (Google), and ChatSonic AI (Writesonic)—to the most commonly searched fibromyalgia-related queries. Materials and Methods: The top 10 most frequently searched fibromyalgia-related questions from the past 2 years were retrieved from the Google Trends database. Each chatbot was queried separately, and a total of 60 responses (10 per chatbot) were assessed both qualitatively and quantitatively by 2 reviewers, focusing on content quality, accuracy, readability, and alignment with evidence-based guidelines. Results: ChatGPT-3.5 had the lowest Ensuring Quality Information for Patients score (20.6 ± 4.5), indicating very low quality information, while Gemini achieved the highest (40.5 ± 5), which was still classified as low quality. Understandability was moderate for Copilot, Gemini, and Perplexity (67.2) but lowest for ChatGPT-3.5 (43.2 ± 10.2). Actionability was weak and the misinformation assessment revealed a moderate level across all chatbots. Readability scores indicated university-level complexity, with ChatGPT-4o having the lowest Reading Ease score (11.3 ± 11.2) and Copilot the highest (30.3 ± 13.2). Conclusion: While AI chatbots provide accessible health information, their accuracy and depth vary. Gemini, Copilot, and Perplexity AI showed better quality, but citation inconsistencies, readability challenges, and misinformation risks highlight the need for refinement beyond the hype. Clinicians should guide fibromyalgia patients in critically assessing AI-generated health content. Future research should explore improvements in AI chatbot applicability for medical inquiries.

Article: The effect of blood flow restricted aerobic exercise training on pain, functional status, quality of life and hormonal response to exercise in fibromyalgia patients: a randomized double-blind study - European Journal of Physical and Rehabilitation Medicine 2025 May 28 - Minerva Medica - Journals

Ultrasound-guided injections have become increasingly popular in musculoskeletal practice due to their precision and minimally invasive nature. Cadaveric musculoskeletal ultrasound-guided injection workshops offer unique hands-on training, allowing physicians to refine their skills and improve procedural confidence in a realistic anatomical context. This study evaluates the effectiveness of such workshops in enhancing participants’ confidence in performing musculoskeletal ultrasound-guided injections. A prospective cross-sectional study was conducted with 109 participants attending three one-day workshops. The workshops included theoretical sessions and hands-on practice on ultrasound-guided injections of the hip, knee, and ankle joints using cadaveric specimens. Participants completed pre- and post-workshop surveys assessing their self-reported confidence in performing injections. Changes in confidence were analyzed by comparing pre- and post-course responses to evaluate the effectiveness of the training. Participants showed significant improvements in confidence for all three joints, with the largest gain observed in hip injections. Residents exhibited lower pre- and post-course confidence but demonstrated greater improvement compared to specialists. Prior exposure to ultrasound training was associated with higher pre-course confidence but not post-course; implying that those who had not attended previous ultrasonography courses showed greater improvement and that cadaveric courses compensate for the lack of previous courses. Access to ultrasound equipment during residency was positively correlated with higher confidence levels. Cadaveric musculoskeletal ultrasound-guided injection workshops significantly enhance procedural confidence among physicians, particularly for hip and ankle joint injections. These workshops are valuable enough tools to make up for missed preceding ultrasound program attendance, and it could thus call for a reevaluation of how educational resources are distributed to optimize procedural competence and patient safety.

Objectives: This study aims to accurately evaluate pain lasting longer than three months and falls under the category of chronic pain and to determine the risk factors to follow up and treat properly and to develop appropriate diagnostic and treatment algorithms. Patients and methods: Between March 2021 and December 2021, a total of 437 patients (162 males, 275 females; mean age: 44±14.6 years; range, 12 to 82 years) who were referred to the participating centers due to pain complaints and were diagnosed with post-COVID-19 condition according to the criteria defined by the World Health Organization (WHO) were included in the study. The patients were divided into three groups as nociceptive pain, neuropathic pain, and central sensitization, based on the physician's clinical evaluation and the Self-Report Leeds Assessment of Neuropathic Symptoms and Signs (S-LANSS) and Central Sensitization Inventory scores. Results: The most common diagnosis was nociceptive pain followed by central sensitization. Patients with nociceptive pain had less pain. It was found that not exercising regularly, having a chronic disease and being a woman were risk factors for central sensitization, having thyroid disease before COVID-19, and defining the current pain as very severe were risk factors for neuropathic pain. Conclusion: In the evaluation of post-COVID-19 pain, neuropathic pain and central sensitization should be also considered in addition to nociceptive pain and the severity of pain, systemic diseases and physical activity should be questioned.

Objective YouTube, a popular worldwide source to access health information online, may have an impact in enhancing exercise therapy for fibromyalgia patients. This study is aimed to investigate the quality and reliability of exercise videos prepared for fibromyalgia syndrome on YouTube. Methods A thorough search on YouTube using six search terms related to fibromyalgia treatment and exercise was conducted. The videos with content on exercises for fibromyalgia, acceptable audio-video quality, and are in English language were included, whereas duplicates were excluded from the analysis. Quality was determined using the Global Quality Scale (GQS) and JAMA system, and reliability using the mDISCERN criteria. Results Out of a total of 600 videos, 70 videos that met the inclusion criteria were included in the analysis. Physicians were found to produce content mostly in high and medium quality (%70.2), other medical professionals mostly in low and medium quality (93.8%), while non-medical users mostly in low quality (86.7%). Ninety percent of the videos targeted patients, while 10% targeted healthcare professionals as an audience, and the majority (93.7%) of videos targeting patients were of low and medium quality, as most (85.7%) of the videos targeting healthcare professionals were of high quality. Conclusion YouTube can be used as a tool to increase exercise adherence in fibromyalgia patients, as it offers a wide range of content that is easily accessible and useful at times. To implement this tool, physicians need to upload more content online as they stand for pioneers of high-quality information dissemination on the internet. KeyPoints • This research emphasizes the need for high-quality online healthcare information which is accessed by a vast number of individuals and the implementations that effectively address and prevent the spread of misinformation. • Additionally, it presents suggestions for the future regarding the uploading of such content on the internet. • Physicians hold the key responsibility in ensuring the availability of high-quality online health content, thereby carrying significant implications for its overall impact.