Qi Gao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Qi Gao has a total of 35 co-authors. Here are the top co-authors by number of shared publications:
- Caihua Huang
Research and Communication Center of Exercise and Health, Xiamen University of Technology, Xiamen 361024, China
6 shared publications - Xiangxun Lai
Xiamen University of Technology
6 shared publications - Yue Lai
Department of Mathematics and Digital Science, Chengyi College, Jimei University, Xiamen, China
5 shared publications - Bing Ruan
School of Sports Medicine and Rehabilitation, Beijing Sport University, Beijing, China
3 shared publications - Jiacheng Chen
Research and Communication Center for Exercise and Health, Xiamen University of Technology, 600 Ligong Road, Jimei District, Xiamen, Fujian Province, 361024, China, 86 15606951380
3 shared publications - Kaijiang Pan
School of Marxism, Xiamen Ocean Vocational College, Xiamen, Fujian, People’s Republic of China
3 shared publications
Latest Publications
- Open Access
- 9 October 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
Tacrolimus attenuates BCL-2/BAX-mediated chondrocyte apoptosis and cartilage matrix degradation in experimental osteoarthritis
- 9 October 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
Open AccessTacrolimus attenuates BCL-2/BAX-mediated chondrocyte apoptosis and cartilage matrix degradation in experimental osteoarthritis
- Open Access
- 25 September 2026
- Published by Frontiers Media SA in Frontiers in Medicine
- Vol. 13, 1991448
Correction: Comparative effectiveness and dose–response relationship of exercise-based interventions for muscle mass in older women with sarcopenia: a systematic review and network meta-analysis
Sarcopenia is a progressive, generalized skeletal muscle disorder primarily associated with aging (1). Although diagnostic thresholds vary across the major criteria, these frameworks converge on a common definition based on muscle strength, muscle mass, and physical performance (2,3). Its global prevalence is estimated at 10-16% and is rising across regions (4,5), with both prevalence and severity increasing with advancing age (6). As a geriatric syndrome, sarcopenia frequently coexists with obesity (7), frailty (8), and osteoporosis (9), thereby accelerating functional decline and clinical vulnerability. At the biological level, it is characterized by impaired muscle protein synthesis, diminished regenerative capacity, dysregulated anabolic signaling, and chronic low-grade inflammation (10,11). When it coexists with obesity, this inflammatory state has been proposed to link sarcopenia with metabolic syndrome, type 2 diabetes, dyslipidemia, and cardiovascular disease (12,13). These changes further raise the risks of falls, fractures, disability, cognitive impairment, and death, substantially compromising quality of life and imposing a considerable medical and socioeconomic burden (14)(15)(16).This burden may differ between the sexes. Epidemiological studies have reported sex-related differences in the prevalence and risk-factor profile of sarcopenia (17,18), although the direction and magnitude of these differences may depend on the diagnostic criteria, methods used to adjust muscle mass, age distribution, ethnicity, and population characteristics. Female sex has also been associated with sarcopenia risk in some cohort-based analyses (19). Older women generally have a lower baseline muscle reserve (20), while estrogen loss may reduce anabolic protection (21,22) and contribute to changes in inflammatory status (23,24). These factors may increase the risk of falls, fractures, and other adverse outcomes (25). Menopause may further contribute to an adverse cardiometabolic profile, including higher total and LDL cholesterol, lower HDL cholesterol, higher blood pressure, and insulin resistance (26,27). These physiological and metabolic differences may also influence anabolic responses to exercise, exercise tolerance, and recovery, and may therefore modify the exercise dose required to improve muscle mass. Thus, evidence derived from mixed-sex populations may not fully reflect the effects or dose-response relationships of exercise in older women with sarcopenia.Exercise and nutrition are the principal non-pharmacological strategies for managing sarcopenia (28). Evidence from systematic reviews and meta-analyses broadly indicates that resistancebased multicomponent programs, particularly when combined with nutritional supplementation, improve muscle strength, mass, physical function, and quality of life in older adults with sarcopenia (15,(29)(30)(31)(32)(33). These findings are nonetheless inconsistent, likely reflecting differences in population characteristics and intervention protocols.Given the sex-related differences in sarcopenia (18,34), specific evidence is needed, yet available female-focused meta-analyses have several limitations. First, the participants included across these reviews are potentially heterogeneous. Several reviews did not base their inclusion criteria on established diagnostic definitions of sarcopenia and enrolled predominantly postmenopausal women (35,36), which weakens the specificity of their conclusions. Second, a recent female-focused network meta-analysis, though it integrates direct and indirect evidence to rank interventions, remains susceptible to heterogeneity in the definition of interventions (37). All exercise interventions were pooled under a single "exercise" node, without distinguishing specific modalities or separating single-component from multi-component programs; such coarse classification may introduce within-node heterogeneity and obscure the contribution of specific exercise components. Third, these analyses overlooked the covariate effects of age, control-group condition, and exercise-prescription variables, thereby hindering the design of individualized rehabilitation. Prior reviews have chiefly assessed which interventions are effective and which rank highest (37,38), without examining the relationship between exercise dose and treatment effect (39). Muscle mass is both a core diagnostic indicator of sarcopenia and a principal target of non-pharmacological treatment (40). Addressing these gaps, the present study performed a network meta-analysis together with a dose-response analysis, to identify the most effective exercise components and estimate an effective weekly dose range for improving muscle mass in older women with sarcopenia.This systematic review and network meta-analysis followed the Cochrane Handbook and was reported according to the PRISMA 2020 statement. The protocol was registered in advance in PROSPERO (CRD420261449654).Eligibility followed the PICOS framework. Population: older women aged ≥60 years with sarcopenia diagnosed by EWGSOP/EWGSOP2, AWGS, IWGS, FNIH, SDOC, or comparable criteria. A trial was eligible if all participants were women or, in mixed-sex trials, if women constituted at least 70% of the enrolled sample; this ≥70% threshold defined a predominantly female trial. Participants who also presented with one related condition-obesity, osteoporosis, or frailty-remained eligible, provided that no more than one such condition was present. Intervention: structured exercise programs, including resistance, aerobic, balance, flexibility, and multi-component exercise, traditional Chinese exercises (CTX; Tai Chi, Baduanjin, Yijinjing, or comparable mind-body forms), and vibration training. These could be delivered in supervised, home-based, community-based, clinical, or mixed settings, alone or with non-pharmacological nutritional support. Comparator: usual or routine care, health education, waiting-list...
- 25 September 2026
- Published by Frontiers Media SA in Frontiers in Medicine
- Vol. 13, 1991448
Open AccessCorrection: Comparative effectiveness and dose–response relationship of exercise-based interventions for muscle mass in older women with sarcopenia: a systematic review and network meta-analysis
Sarcopenia is a progressive, generalized skeletal muscle disorder primarily associated with aging (1). Although diagnostic thresholds vary across the major criteria, these frameworks converge on a common definition based on muscle strength, muscle mass, and physical performance (2,3). Its global prevalence is estimated at 10-16% and is rising across regions (4,5), with both prevalence and severity increasing with advancing age (6). As a geriatric syndrome, sarcopenia frequently coexists with obesity (7), frailty (8), and osteoporosis (9), thereby accelerating functional decline and clinical vulnerability. At the biological level, it is characterized by impaired muscle protein synthesis, diminished regenerative capacity, dysregulated anabolic signaling, and chronic low-grade inflammation (10,11). When it coexists with obesity, this inflammatory state has been proposed to link sarcopenia with metabolic syndrome, type 2 diabetes, dyslipidemia, and cardiovascular disease (12,13). These changes further raise the risks of falls, fractures, disability, cognitive impairment, and death, substantially compromising quality of life and imposing a considerable medical and socioeconomic burden (14)(15)(16).This burden may differ between the sexes. Epidemiological studies have reported sex-related differences in the prevalence and risk-factor profile of sarcopenia (17,18), although the direction and magnitude of these differences may depend on the diagnostic criteria, methods used to adjust muscle mass, age distribution, ethnicity, and population characteristics. Female sex has also been associated with sarcopenia risk in some cohort-based analyses (19). Older women generally have a lower baseline muscle reserve (20), while estrogen loss may reduce anabolic protection (21,22) and contribute to changes in inflammatory status (23,24). These factors may increase the risk of falls, fractures, and other adverse outcomes (25). Menopause may further contribute to an adverse cardiometabolic profile, including higher total and LDL cholesterol, lower HDL cholesterol, higher blood pressure, and insulin resistance (26,27). These physiological and metabolic differences may also influence anabolic responses to exercise, exercise tolerance, and recovery, and may therefore modify the exercise dose required to improve muscle mass. Thus, evidence derived from mixed-sex populations may not fully reflect the effects or dose-response relationships of exercise in older women with sarcopenia.Exercise and nutrition are the principal non-pharmacological strategies for managing sarcopenia (28). Evidence from systematic reviews and meta-analyses broadly indicates that resistancebased multicomponent programs, particularly when combined with nutritional supplementation, improve muscle strength, mass, physical function, and quality of life in older adults with sarcopenia (15,(29)(30)(31)(32)(33). These findings are nonetheless inconsistent, likely reflecting differences in population characteristics and intervention protocols.Given the sex-related differences in sarcopenia (18,34), specific evidence is needed, yet available female-focused meta-analyses have several limitations. First, the participants included across these reviews are potentially heterogeneous. Several reviews did not base their inclusion criteria on established diagnostic definitions of sarcopenia and enrolled predominantly postmenopausal women (35,36), which weakens the specificity of their conclusions. Second, a recent female-focused network meta-analysis, though it integrates direct and indirect evidence to rank interventions, remains susceptible to heterogeneity in the definition of interventions (37). All exercise interventions were pooled under a single "exercise" node, without distinguishing specific modalities or separating single-component from multi-component programs; such coarse classification may introduce within-node heterogeneity and obscure the contribution of specific exercise components. Third, these analyses overlooked the covariate effects of age, control-group condition, and exercise-prescription variables, thereby hindering the design of individualized rehabilitation. Prior reviews have chiefly assessed which interventions are effective and which rank highest (37,38), without examining the relationship between exercise dose and treatment effect (39). Muscle mass is both a core diagnostic indicator of sarcopenia and a principal target of non-pharmacological treatment (40). Addressing these gaps, the present study performed a network meta-analysis together with a dose-response analysis, to identify the most effective exercise components and estimate an effective weekly dose range for improving muscle mass in older women with sarcopenia.This systematic review and network meta-analysis followed the Cochrane Handbook and was reported according to the PRISMA 2020 statement. The protocol was registered in advance in PROSPERO (CRD420261449654).Eligibility followed the PICOS framework. Population: older women aged ≥60 years with sarcopenia diagnosed by EWGSOP/EWGSOP2, AWGS, IWGS, FNIH, SDOC, or comparable criteria. A trial was eligible if all participants were women or, in mixed-sex trials, if women constituted at least 70% of the enrolled sample; this ≥70% threshold defined a predominantly female trial. Participants who also presented with one related condition-obesity, osteoporosis, or frailty-remained eligible, provided that no more than one such condition was present. Intervention: structured exercise programs, including resistance, aerobic, balance, flexibility, and multi-component exercise, traditional Chinese exercises (CTX; Tai Chi, Baduanjin, Yijinjing, or comparable mind-body forms), and vibration training. These could be delivered in supervised, home-based, community-based, clinical, or mixed settings, alone or with non-pharmacological nutritional support. Comparator: usual or routine care, health education, waiting-list...
- Open Access
- 25 August 2026
- Published by Frontiers Media SA in Frontiers in Medicine
- Vol. 13, 1938860
Comparative effectiveness and dose–response relationship of exercise-based interventions for muscle mass in older women with sarcopenia: a systematic review and network meta-analysis
Background: Exercise and nutrition are the principal non-pharmacological strategies for sarcopenia, but the relative effects of different exercise combinations and the dose required to improve muscle mass in older women remain uncertain. This study compared exercise-based interventions and examined the dose-response relationship. Methods: PubMed, Web of Science, Embase, the Cochrane Library, and EBSCO were searched from inception to 1 July 2026. Eligible trials enrolled women aged ≥ 60 years female with sarcopenia. Risk of bias was assessed using RoB 2. A frequentist random-effects network meta-analysis combined direct and indirect evidence, with effects expressed as standardized mean differences (SMDs) and 95% confidence intervals (CIs). Interventions were ranked using P-scores. Network meta-regression examined control condition, age, intervention duration, weekly frequency, weekly exercise dose, and female proportion. A Bayesian model-based network meta-analysis assessed the dose–response relationship using MET-min/week. Results: Twenty-seven controlled trials involving 1,845 participants were included. Heterogeneity was moderate, with no significant global inconsistency. Most interventions improved muscle mass compared with control, although no significant effect was detected for Chinese traditional exercise. Resistance, aerobic, and balance training combined with nutritional supplementation produced the largest effect (SMD = 1.41, 95% CI 0.81–2.01) and ranked first ( P -score = 0.882). Control condition, age, and weekly exercise dose were significant effect modifiers, but covariate adjustment did not materially alter the intervention rankings. The dose–response relationship was positive and nonlinear. Within the central evidence-supported range, the lower 95% credible limit first exceeded zero at 460 MET-min/week (SMD = 0.19, 95% CrI 0.01–0.37). At the upper boundary of this range (1,144 MET-min/week), the predicted SMD was 1.38 (95% CrI 1.08–1.61). Conclusion: Resistance-based multicomponent exercise may improve muscle mass in older women with sarcopenia, particularly when combined with nutritional supplementation. Exercise dose should be increased gradually according to individual tolerance. The reported dose values should be interpreted as evidence-supported estimates rather than fixed biological thresholds or universally optimal prescriptions. Systematic review registration: https://www.crd.york.ac.uk/PROSPERO/view/CRD420261449654 , identifier: PROSPERO (CRD420261449654).
- 25 August 2026
- Published by Frontiers Media SA in Frontiers in Medicine
- Vol. 13, 1938860
Open AccessComparative effectiveness and dose–response relationship of exercise-based interventions for muscle mass in older women with sarcopenia: a systematic review and network meta-analysis
Background: Exercise and nutrition are the principal non-pharmacological strategies for sarcopenia, but the relative effects of different exercise combinations and the dose required to improve muscle mass in older women remain uncertain. This study compared exercise-based interventions and examined the dose-response relationship. Methods: PubMed, Web of Science, Embase, the Cochrane Library, and EBSCO were searched from inception to 1 July 2026. Eligible trials enrolled women aged ≥ 60 years female with sarcopenia. Risk of bias was assessed using RoB 2. A frequentist random-effects network meta-analysis combined direct and indirect evidence, with effects expressed as standardized mean differences (SMDs) and 95% confidence intervals (CIs). Interventions were ranked using P-scores. Network meta-regression examined control condition, age, intervention duration, weekly frequency, weekly exercise dose, and female proportion. A Bayesian model-based network meta-analysis assessed the dose–response relationship using MET-min/week. Results: Twenty-seven controlled trials involving 1,845 participants were included. Heterogeneity was moderate, with no significant global inconsistency. Most interventions improved muscle mass compared with control, although no significant effect was detected for Chinese traditional exercise. Resistance, aerobic, and balance training combined with nutritional supplementation produced the largest effect (SMD = 1.41, 95% CI 0.81–2.01) and ranked first ( P -score = 0.882). Control condition, age, and weekly exercise dose were significant effect modifiers, but covariate adjustment did not materially alter the intervention rankings. The dose–response relationship was positive and nonlinear. Within the central evidence-supported range, the lower 95% credible limit first exceeded zero at 460 MET-min/week (SMD = 0.19, 95% CrI 0.01–0.37). At the upper boundary of this range (1,144 MET-min/week), the predicted SMD was 1.38 (95% CrI 1.08–1.61). Conclusion: Resistance-based multicomponent exercise may improve muscle mass in older women with sarcopenia, particularly when combined with nutritional supplementation. Exercise dose should be increased gradually according to individual tolerance. The reported dose values should be interpreted as evidence-supported estimates rather than fixed biological thresholds or universally optimal prescriptions. Systematic review registration: https://www.crd.york.ac.uk/PROSPERO/view/CRD420261449654 , identifier: PROSPERO (CRD420261449654).
- Top Cited PapersOpen Access
- 17 June 2025
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 27, e79217
Evaluation Strategies for LLM-Based Models in Exercise and Health Coaching: A Scoping Review (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: Large language model (LLM)-based AI coaches show promise for personalized exercise and health interventions. Their complex capabilities (communication, planning, movement analysis, monitoring) necessitate rigorous, multidimensional evaluation, but standardized frameworks are lacking. Objective: This scoping review systematically maps current evaluation strategies for LLM-based AI coaches in exercise and health, identifies strengths and limitations, and proposes future directions for robust, standardized validation. Methods: Following PRISMA-ScR guidelines, we systematically searched six databases using keywords for LLMs, exercise/health coaching, and evaluation. Studies describing LLM-based coaching systems with reported performance evaluation methods were included. Data on models, applications, evaluation strategies, and outcomes were charted. Results: Seventeen studies published between March 2023 and March 2025 met the inclusion criteria. Most utilized proprietary models (e.g., GPT-4), while some used open-source or custom models. Six studies incorporated multimodal inputs (video, sensor data). Evaluation strategies were highly heterogeneous, including quantitative metrics (Accuracy, F1, MAE), empirical methods (user studies, expert comparisons), and expert/user-centered feedback (expert scores [Kappa ≈ 0.79–0.82], user surveys [MITI, SASSI]). However, evaluations often lacked real-world testing, longitudinal assessment, and standardized benchmarks. Conclusions: Evaluating LLM-based exercise and health coaches requires multifaceted strategies: quantitative metrics for objective tasks, empirical validation for user interaction, and expert assessment for personalization and safety. Current evaluations are fragmented, lacking standardization, ecological validity, and longitudinal assessment. Future progress demands robust, multidimensional frameworks emphasizing real-world validation, integrating RAG for accuracy, and developing specialized, efficient multimodal models or agents for reliable and scalable AI coaching.
- 17 June 2025
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 27, e79217
Top Cited PapersOpen AccessEvaluation Strategies for LLM-Based Models in Exercise and Health Coaching: A Scoping Review (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: Large language model (LLM)-based AI coaches show promise for personalized exercise and health interventions. Their complex capabilities (communication, planning, movement analysis, monitoring) necessitate rigorous, multidimensional evaluation, but standardized frameworks are lacking. Objective: This scoping review systematically maps current evaluation strategies for LLM-based AI coaches in exercise and health, identifies strengths and limitations, and proposes future directions for robust, standardized validation. Methods: Following PRISMA-ScR guidelines, we systematically searched six databases using keywords for LLMs, exercise/health coaching, and evaluation. Studies describing LLM-based coaching systems with reported performance evaluation methods were included. Data on models, applications, evaluation strategies, and outcomes were charted. Results: Seventeen studies published between March 2023 and March 2025 met the inclusion criteria. Most utilized proprietary models (e.g., GPT-4), while some used open-source or custom models. Six studies incorporated multimodal inputs (video, sensor data). Evaluation strategies were highly heterogeneous, including quantitative metrics (Accuracy, F1, MAE), empirical methods (user studies, expert comparisons), and expert/user-centered feedback (expert scores [Kappa ≈ 0.79–0.82], user surveys [MITI, SASSI]). However, evaluations often lacked real-world testing, longitudinal assessment, and standardized benchmarks. Conclusions: Evaluating LLM-based exercise and health coaches requires multifaceted strategies: quantitative metrics for objective tasks, empirical validation for user interaction, and expert assessment for personalization and safety. Current evaluations are fragmented, lacking standardization, ecological validity, and longitudinal assessment. Future progress demands robust, multidimensional frameworks emphasizing real-world validation, integrating RAG for accuracy, and developing specialized, efficient multimodal models or agents for reliable and scalable AI coaching.
- Open Access
- 17 June 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Evaluation Strategies for LLM-Based Models in Exercise and Health Coaching: A Scoping Review (Preprint)
BACKGROUND Large language model (LLM)-based AI coaches show promise for personalized exercise and health interventions. Their complex capabilities (communication, planning, movement analysis, monitoring) necessitate rigorous, multidimensional evaluation, but standardized frameworks are lacking. OBJECTIVE This scoping review systematically maps current evaluation strategies for LLM-based AI coaches in exercise and health, identifies strengths and limitations, and proposes future directions for robust, standardized validation. METHODS Following PRISMA-ScR guidelines, we systematically searched six databases using keywords for LLMs, exercise/health coaching, and evaluation. Studies describing LLM-based coaching systems with reported performance evaluation methods were included. Data on models, applications, evaluation strategies, and outcomes were charted. RESULTS Seventeen studies published between March 2023 and March 2025 met the inclusion criteria. Most utilized proprietary models (e.g., GPT-4), while some used open-source or custom models. Six studies incorporated multimodal inputs (video, sensor data). Evaluation strategies were highly heterogeneous, including quantitative metrics (Accuracy, F1, MAE), empirical methods (user studies, expert comparisons), and expert/user-centered feedback (expert scores [Kappa ≈ 0.79–0.82], user surveys [MITI, SASSI]). However, evaluations often lacked real-world testing, longitudinal assessment, and standardized benchmarks. CONCLUSIONS Evaluating LLM-based exercise and health coaches requires multifaceted strategies: quantitative metrics for objective tasks, empirical validation for user interaction, and expert assessment for personalization and safety. Current evaluations are fragmented, lacking standardization, ecological validity, and longitudinal assessment. Future progress demands robust, multidimensional frameworks emphasizing real-world validation, integrating RAG for accuracy, and developing specialized, efficient multimodal models or agents for reliable and scalable AI coaching.
- 17 June 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessEvaluation Strategies for LLM-Based Models in Exercise and Health Coaching: A Scoping Review (Preprint)
BACKGROUND Large language model (LLM)-based AI coaches show promise for personalized exercise and health interventions. Their complex capabilities (communication, planning, movement analysis, monitoring) necessitate rigorous, multidimensional evaluation, but standardized frameworks are lacking. OBJECTIVE This scoping review systematically maps current evaluation strategies for LLM-based AI coaches in exercise and health, identifies strengths and limitations, and proposes future directions for robust, standardized validation. METHODS Following PRISMA-ScR guidelines, we systematically searched six databases using keywords for LLMs, exercise/health coaching, and evaluation. Studies describing LLM-based coaching systems with reported performance evaluation methods were included. Data on models, applications, evaluation strategies, and outcomes were charted. RESULTS Seventeen studies published between March 2023 and March 2025 met the inclusion criteria. Most utilized proprietary models (e.g., GPT-4), while some used open-source or custom models. Six studies incorporated multimodal inputs (video, sensor data). Evaluation strategies were highly heterogeneous, including quantitative metrics (Accuracy, F1, MAE), empirical methods (user studies, expert comparisons), and expert/user-centered feedback (expert scores [Kappa ≈ 0.79–0.82], user surveys [MITI, SASSI]). However, evaluations often lacked real-world testing, longitudinal assessment, and standardized benchmarks. CONCLUSIONS Evaluating LLM-based exercise and health coaches requires multifaceted strategies: quantitative metrics for objective tasks, empirical validation for user interaction, and expert assessment for personalization and safety. Current evaluations are fragmented, lacking standardization, ecological validity, and longitudinal assessment. Future progress demands robust, multidimensional frameworks emphasizing real-world validation, integrating RAG for accuracy, and developing specialized, efficient multimodal models or agents for reliable and scalable AI coaching.
- Top Cited PapersOpen Access
- 27 May 2025
- Published by JMIR Publications Inc. in JMIR Medical Informatics
- Vol. 13, e59309
Using Large Language Models to Enhance Exercise Recommendations and Physical Activity in Clinical and Healthy Populations: Scoping Review
Background: Regular exercise recommendations (ERs) and physical activity (PA) are crucial for the prevention and management of chronic diseases. However, creating effective exercise programs demand substantial time and specialized expertise from both medical and sports professionals. Large language models (LLMs), such as ChatGPT, offer a promising solution by helping create personalized ERs. While LLMs show potential, their use in exercise planning remains in its early stages and requires further exploration. Objectives: This study aims to systematically review and classify the applications of LLMs in ERs and PA. It also seeks to identify existing gaps and provide insights into future research directions for optimizing LLM integration in personalized health interventions. Methods: A scoping review methodology was used to identify studies related to LLM applications in ERs and PA. Literature searches were conducted in Web of Science, PubMed, IEEE, and arXiv for English language papers published up to March 21, 2024. Keywords included LLMs, chatbots, ERs, PA, fitness plan, and related terms. Two independent reviewers (XL and CH) screened and selected studies based on predefined inclusion criteria. Thematic analysis was used to synthesize findings, which were presented narratively. Results: An initial search identified 598 papers, of which 1.8% (11/598) of studies were included after screening and applying selection criteria. Of these, ChatGPT-based models were used in 55% (6/11) of the studies. In addition, 73% (8/11) of the studies used expert evaluations and user feedback to assess model usability, and 45% (5/11) of the studies used experimental designs to evaluate LLM interventions in ERs and PA. Key findings indicated that LLMs can generate tailored ERs, save time in clinical practice, and enhance safety by incorporating patient-specific data. They also increased engagement and supported behavior change. This made PA guidance more accessible, especially in remote or underserved communities. Conclusions: This review highlights the promising applications of LLMs in ERs and PA but emphasizes that they remain a supplement to human expertise. Expert validation is essential to ensure safety and mitigate risks. Future research should prioritize pilot testing, clinician training programs, and large-scale clinical trials to enhance feasibility, transparency, and ethical integration.
- 27 May 2025
- Published by JMIR Publications Inc. in JMIR Medical Informatics
- Vol. 13, e59309
Top Cited PapersOpen AccessUsing Large Language Models to Enhance Exercise Recommendations and Physical Activity in Clinical and Healthy Populations: Scoping Review
Background: Regular exercise recommendations (ERs) and physical activity (PA) are crucial for the prevention and management of chronic diseases. However, creating effective exercise programs demand substantial time and specialized expertise from both medical and sports professionals. Large language models (LLMs), such as ChatGPT, offer a promising solution by helping create personalized ERs. While LLMs show potential, their use in exercise planning remains in its early stages and requires further exploration. Objectives: This study aims to systematically review and classify the applications of LLMs in ERs and PA. It also seeks to identify existing gaps and provide insights into future research directions for optimizing LLM integration in personalized health interventions. Methods: A scoping review methodology was used to identify studies related to LLM applications in ERs and PA. Literature searches were conducted in Web of Science, PubMed, IEEE, and arXiv for English language papers published up to March 21, 2024. Keywords included LLMs, chatbots, ERs, PA, fitness plan, and related terms. Two independent reviewers (XL and CH) screened and selected studies based on predefined inclusion criteria. Thematic analysis was used to synthesize findings, which were presented narratively. Results: An initial search identified 598 papers, of which 1.8% (11/598) of studies were included after screening and applying selection criteria. Of these, ChatGPT-based models were used in 55% (6/11) of the studies. In addition, 73% (8/11) of the studies used expert evaluations and user feedback to assess model usability, and 45% (5/11) of the studies used experimental designs to evaluate LLM interventions in ERs and PA. Key findings indicated that LLMs can generate tailored ERs, save time in clinical practice, and enhance safety by incorporating patient-specific data. They also increased engagement and supported behavior change. This made PA guidance more accessible, especially in remote or underserved communities. Conclusions: This review highlights the promising applications of LLMs in ERs and PA but emphasizes that they remain a supplement to human expertise. Expert validation is essential to ensure safety and mitigate risks. Future research should prioritize pilot testing, clinician training programs, and large-scale clinical trials to enhance feasibility, transparency, and ethical integration.
- Open Access
- 12 April 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Mind the Gap: Evaluating Domain-Specific Language Models Beyond General Benchmarks in Exercise Health AI (Preprint)
UNSTRUCTURED Standard benchmarks used for general large language models (LLMs) are inadequate for evaluating specialized models in the high-stakes exercise health domain, where safety, personalization, and deep domain knowledge are paramount. Current evaluation practices, including expert scoring, knowledge-based Q&A, and user feedback, offer limited insights into real-world applicability and clinical robustness. This discussion highlights these critical evaluation gaps and argues for the necessity of a multi-faceted, domain-specific evaluation framework. We propose incorporating strategies such as contextualized scenario simulations, comparative benchmarking against professional tools, structured cross-disciplinary audits, iterative real-world feedback loops, and rigorous adversarial safety testing. Developing and adopting such tailored, comprehensive evaluation methods is crucial for ensuring the reliability, safety, and effectiveness of LLMs, thereby fostering trust and enabling their responsible integration into exercise health practice to benefit athletes and patients.
- 12 April 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessMind the Gap: Evaluating Domain-Specific Language Models Beyond General Benchmarks in Exercise Health AI (Preprint)
UNSTRUCTURED Standard benchmarks used for general large language models (LLMs) are inadequate for evaluating specialized models in the high-stakes exercise health domain, where safety, personalization, and deep domain knowledge are paramount. Current evaluation practices, including expert scoring, knowledge-based Q&A, and user feedback, offer limited insights into real-world applicability and clinical robustness. This discussion highlights these critical evaluation gaps and argues for the necessity of a multi-faceted, domain-specific evaluation framework. We propose incorporating strategies such as contextualized scenario simulations, comparative benchmarking against professional tools, structured cross-disciplinary audits, iterative real-world feedback loops, and rigorous adversarial safety testing. Developing and adopting such tailored, comprehensive evaluation methods is crucial for ensuring the reliability, safety, and effectiveness of LLMs, thereby fostering trust and enabling their responsible integration into exercise health practice to benefit athletes and patients.
- Open Access
- 12 April 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Breaking the Data Bottleneck in Exercise Health AI (Preprint)
UNSTRUCTURED Artificial intelligence (AI) holds significant potential to transform exercise health through applications like personalized training, injury prevention, and targeted rehabilitation. However, the scarcity of high-quality, real-world data—due to issues like data fragmentation, inconsistent formats, privacy constraints, and confidentiality concerns—severely restricts AI development and deployment in this domain. Learning from advancements in healthcare AI, strategies such as federated learning for privacy-preserving collaboration, synthetic data generation to augment limited datasets, transfer learning to leverage existing models, and structured data-sharing protocols offer viable solutions. Furthermore, an innovative closed-loop data generation process is proposed: leveraging large language model APIs to create extensive synthetic datasets from initial high-quality inputs, followed by rigorous human expert validation and refinement through real-world feedback. This paper argues that overcoming the data bottleneck necessitates an integrated strategy combining these methods. Implementing such comprehensive measures will enhance AI model robustness and generalizability, ultimately accelerating the delivery of tangible benefits to athletes and health practitioners.
- 12 April 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessBreaking the Data Bottleneck in Exercise Health AI (Preprint)
UNSTRUCTURED Artificial intelligence (AI) holds significant potential to transform exercise health through applications like personalized training, injury prevention, and targeted rehabilitation. However, the scarcity of high-quality, real-world data—due to issues like data fragmentation, inconsistent formats, privacy constraints, and confidentiality concerns—severely restricts AI development and deployment in this domain. Learning from advancements in healthcare AI, strategies such as federated learning for privacy-preserving collaboration, synthetic data generation to augment limited datasets, transfer learning to leverage existing models, and structured data-sharing protocols offer viable solutions. Furthermore, an innovative closed-loop data generation process is proposed: leveraging large language model APIs to create extensive synthetic datasets from initial high-quality inputs, followed by rigorous human expert validation and refinement through real-world feedback. This paper argues that overcoming the data bottleneck necessitates an integrated strategy combining these methods. Implementing such comprehensive measures will enhance AI model robustness and generalizability, ultimately accelerating the delivery of tangible benefits to athletes and health practitioners.
- Open Access
- 8 April 2024
- Published by JMIR Publications Inc. in JMIR Preprints
Large Language Models for Recommendations of Exercise and Physical Activity: A Scoping Review (Preprint)
BACKGROUND Despite a growing awareness among healthcare professionals regarding the benefits of exercise and physical activity, the integration of exercise recommendations into healthcare practices remains sub-optimal. The complexity of the recommendations, which necessitate interdisciplinary knowledge, poses challenges in formulating the structured exercise plans. The rapid development of Large Language Models (LLMs) and Chatbots, offers a potential solution by leveraging AI's capabilities to enhance the precision and personalization of fitness and health recommendations. OBJECTIVE This paper provides a systematic summary of existing research in the landscape of LLMs and Chatbot applications for exercise recommendations and physical activity, encompassing important topics, theoretical frameworks, methodologies, and potential avenues for further exploration. METHODS A comprehensive literature search was conducted across databases, including Web of Science, PubMed, IEEE, and arXiv. The focus was on studies that utilized LLMs or Chatbots in recommendations for exercise, physical activity or fitness. The search was limited to English-language articles, and the selection process involved two reviews and arbitration in case of disagreement among reviewers. RESULTS The review identified 598 articles initially, narrowing down to 11 reports after applying inclusion and exclusion criteria. The current research domain is remarkably active, playing a pivotal role in the field of exercise health, and studies predominantly focus on integrating LLMs and Chatbots with Just-In-Time Adaptive Interventions (JITAIs) to design personalized exercise programs and evaluate the quality of AI-generated exercise recommendations. The primary LLMs utilized in this field are predominantly from the ChatGPT series developed by OpenAI. CONCLUSIONS Notably, the review underscores the need for further advancements to match the expertise of exercise professionals and address individualized, evidence-based physical activity recommendations. While LLMs and Chatbots demonstrate promising utility in exercise recommendations and physical activity recommendations, there is a critical need for further development to enhance their accuracy, adaptability, and alignment with professional healthcare standards. The future direction emphasizes the fine-tuning of models, enhanced integration of wearable technology with AI and dynamic feedback loops in optimizing personalized health interventions.
- 8 April 2024
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessLarge Language Models for Recommendations of Exercise and Physical Activity: A Scoping Review (Preprint)
BACKGROUND Despite a growing awareness among healthcare professionals regarding the benefits of exercise and physical activity, the integration of exercise recommendations into healthcare practices remains sub-optimal. The complexity of the recommendations, which necessitate interdisciplinary knowledge, poses challenges in formulating the structured exercise plans. The rapid development of Large Language Models (LLMs) and Chatbots, offers a potential solution by leveraging AI's capabilities to enhance the precision and personalization of fitness and health recommendations. OBJECTIVE This paper provides a systematic summary of existing research in the landscape of LLMs and Chatbot applications for exercise recommendations and physical activity, encompassing important topics, theoretical frameworks, methodologies, and potential avenues for further exploration. METHODS A comprehensive literature search was conducted across databases, including Web of Science, PubMed, IEEE, and arXiv. The focus was on studies that utilized LLMs or Chatbots in recommendations for exercise, physical activity or fitness. The search was limited to English-language articles, and the selection process involved two reviews and arbitration in case of disagreement among reviewers. RESULTS The review identified 598 articles initially, narrowing down to 11 reports after applying inclusion and exclusion criteria. The current research domain is remarkably active, playing a pivotal role in the field of exercise health, and studies predominantly focus on integrating LLMs and Chatbots with Just-In-Time Adaptive Interventions (JITAIs) to design personalized exercise programs and evaluate the quality of AI-generated exercise recommendations. The primary LLMs utilized in this field are predominantly from the ChatGPT series developed by OpenAI. CONCLUSIONS Notably, the review underscores the need for further advancements to match the expertise of exercise professionals and address individualized, evidence-based physical activity recommendations. While LLMs and Chatbots demonstrate promising utility in exercise recommendations and physical activity recommendations, there is a critical need for further development to enhance their accuracy, adaptability, and alignment with professional healthcare standards. The future direction emphasizes the fine-tuning of models, enhanced integration of wearable technology with AI and dynamic feedback loops in optimizing personalized health interventions.
- Open Access
- 12 August 2022
- Published by Frontiers Media SA in Frontiers in Oncology
- Vol. 12, 935184
Fucosylated exosomal miRNAs as promising biomarkers for the diagnosis of early lung adenocarcinoma
Background: Considering the absence of apparent symptoms at the early stage, most patients with lung adenocarcinoma (LUAD) present at an advanced stage, leading to a dismal 5-year survival rate of <20%. Thus, finding perspective non-invasive biomarkers for early LUAD is very essential.Methods: We developed a fucose-captured strategy based on lentil lectin-magnetic beads to isolate fucosylated exosomes from serum. Then, a prospective study was conducted to define the diagnostic value of serum exosomal miRNAs for early LUAD. A total of 310 participants were enrolled, including 146 LUAD, 98 benign pulmonary nodules (BPNs), and 66 healthy controls (HCs). Firstly, exosome miRNAs in the discovery cohort (n = 24) were profiled by small RNA sequencing. Secondly, 12 differentially expressed miRNAs (DEmiRs) were selected for further screening in a screening cohort (n = 64) by qRT-PCR. Finally, four candidate miRNAs were selected for further validation in a validating cohort (n = 222).Results: This study demonstrated the feasibility of a fucose-captured strategy for the isolation of fucosylated exosomes from serum, evidenced with exosomal characteristics identified by transmission electron microscopy (TEM), nanoparticle tracking analysis (NTA), and Western blotting, as well as rapid and convenient operation of <10 min. Furthermore, a miRNA panel for early LUAD composed of miR4732-5p, miR451a, miR486-5p, and miR139-3p was defined with an AUC of 0.8554 at 91.07% sensitivity and 66.36% specificity.Conclusions: The fucose-captured strategy provides a reliable, as well as rapid and convenient, approach for the isolation of tumor-derived exosomes from serum. A four-fucosylated exosomal miRNA panel presents good performance for early LUAD diagnosis.
- 12 August 2022
- Published by Frontiers Media SA in Frontiers in Oncology
- Vol. 12, 935184
Open AccessFucosylated exosomal miRNAs as promising biomarkers for the diagnosis of early lung adenocarcinoma
Background: Considering the absence of apparent symptoms at the early stage, most patients with lung adenocarcinoma (LUAD) present at an advanced stage, leading to a dismal 5-year survival rate of <20%. Thus, finding perspective non-invasive biomarkers for early LUAD is very essential.Methods: We developed a fucose-captured strategy based on lentil lectin-magnetic beads to isolate fucosylated exosomes from serum. Then, a prospective study was conducted to define the diagnostic value of serum exosomal miRNAs for early LUAD. A total of 310 participants were enrolled, including 146 LUAD, 98 benign pulmonary nodules (BPNs), and 66 healthy controls (HCs). Firstly, exosome miRNAs in the discovery cohort (n = 24) were profiled by small RNA sequencing. Secondly, 12 differentially expressed miRNAs (DEmiRs) were selected for further screening in a screening cohort (n = 64) by qRT-PCR. Finally, four candidate miRNAs were selected for further validation in a validating cohort (n = 222).Results: This study demonstrated the feasibility of a fucose-captured strategy for the isolation of fucosylated exosomes from serum, evidenced with exosomal characteristics identified by transmission electron microscopy (TEM), nanoparticle tracking analysis (NTA), and Western blotting, as well as rapid and convenient operation of <10 min. Furthermore, a miRNA panel for early LUAD composed of miR4732-5p, miR451a, miR486-5p, and miR139-3p was defined with an AUC of 0.8554 at 91.07% sensitivity and 66.36% specificity.Conclusions: The fucose-captured strategy provides a reliable, as well as rapid and convenient, approach for the isolation of tumor-derived exosomes from serum. A four-fucosylated exosomal miRNA panel presents good performance for early LUAD diagnosis.