Background: The integration of artificial intelligence (AI) in health care has significant potential, yet its acceptance by health care professionals (HCPs) is essential for successful implementation. Understanding HCPs’ perspectives on the explainability and integrability of medical AI is crucial, as these factors influence their willingness to adopt and effectively use such technologies. Objective: This study aims to improve the acceptance and use of medical AI. From a user perspective, it explores HCPs’ understanding of the explainability and integrability of medical AI. Methods: We performed a mixed systematic review by conducting a comprehensive search in the PubMed, Web of Science, Scopus, IEEE Xplore, and ACM Digital Library and arXiv databases for studies published between 2014 and 2024. Studies concerning an explanation or the integrability of medical AI were included. Study quality was assessed using the Joanna Briggs Institute critical appraisal checklist and Mixed Methods Appraisal Tool, with only medium- or high-quality studies included. Qualitative data were analyzed via thematic analysis, while quantitative findings were synthesized narratively. Results: Out of 11,888 records initially retrieved, 22 (0.19%) studies met the inclusion criteria. All selected studies were published from 2020 onward, reflecting the recency and relevance of the topic. The majority (18/22, 82%) originated from high-income countries, and most (17/22, 77%) adopted qualitative methodologies, with the remainder (5/22, 23%) using quantitative or mixed method approaches. From the included studies, a conceptual framework was developed that delineates HCPs’ perceptions of explainability and integrability. Regarding explainability, HCPs predominantly emphasized postprocessing explanations, particularly aspects of local explainability such as feature relevance and case-specific outputs. Visual tools that enhance the explainability of AI decisions (eg, heat maps and feature attribution) were frequently mentioned as important enablers of trust and acceptance. For integrability, key concerns included workflow adaptation, system compatibility with electronic health records, and overall ease of use. These aspects were consistently identified as primary conditions for real-world adoption. Conclusions: To foster wider adoption of AI in clinical settings, future system designs must center on the needs of HCPs. Enhancing post hoc explainability and ensuring seamless integration into existing workflows are critical to building trust and promoting sustained use. The proposed conceptual framework can serve as a practical guide for developers, researchers, and policy makers in aligning AI solutions with frontline user expectations. Trial Registration: PROSPERO CRD420250652253; https://www.crd.york.ac.uk/PROSPERO/view/CRD420250652253

Joint Diseases and Related Surgery (JDRS), the official journal of the Turkish Joint Diseases Foundation, is an international, peer-reviewed, immediate open access journal. All published articles in JDRS has been free for everyone, without embargo, since 1990.

Objectives This study aimed to investigate the impact of interpregnancy weight changes (IPWC) on the gestational diabetes mellitus (GDM) in the second pregnancy. Design A single-centre retrospective cohort study was conducted in China. Setting Data were collected in Peking University Shenzhen Hospital from 2013 January to 2021 February. Participants Participants include women who had two consecutive singleton deliveries after 28 gestational weeks (n=2372). Outcomes The GDM in the second pregnancy (s-GDM) was set as the outcome. Methods IPWC was defined as the change in body mass index between the first trimester of the second pregnancy and that of the first pregnancy, categorised into four groups with −1 kg/m² to <1 kg/m² as the reference. Adjusted ORs (aORs) with 95% CIs attained from multivariable logistic regression were used to assess the association between IPWC and s-GDM, in both total subjects and stratified subgroups. Results In the overall analysis, s-GDM was found to be significantly associated with IPWC value (aOR 1.111; 95% CI 1.038 to 1.190) and an IPWC category of ≥3 kg/m² (aOR 1.821; 95% CI 1.197 to 2.772). In the stratified analysis, the significant association between IPWC ≥3 kg/m² and s-GDM was evident only in the subgroups of an interpregnancy interval (IPI) of less than 36 months (aOR 2.210, 95% CI 1.251 to 3.904), under the age of 35 (aOR 1.854, 95% CI 1.204 to 2.857), non-diabetic status in the first pregnancy (f-ND) (aOR 1.872, 95% CI 1.143 to 3.065) and those with normal weight in the first pregnancy (aOR 1.936, 95% CI 1.174 to 3.193). The significant association between IPWC value and s-GDM was also shown only in these subgroups (p<0.05). In f-DN subgroup, even an IPWC category of 1 kg/m² to <3 kg/m² was significantly associated with s-GDM (aOR 1.486, 95% CI 1.044 to 2.117). IPWC < −1 kg/m² was not significantly associated with s-GDM either in the overall analysis or in the stratified analysis (p>0.05). Conclusion An IPWC of 3 kg/m² or higher may increase the risk of s-GDM, particularly among women with an IPI of less than 36 months, those under 35 years old, individuals without diabetes, or those with normal weight during their first pregnancy. The potential influence of prior GDM on the relationship between IPWC and s-GDM warrants further investigation.

Background: DeepSeek is an open-source large language model (LLM), and it has greatly accelerated LLM adoption in health care. Its rapid deployment has sparked concerns regarding its impact on patient outcomes and safety. However, little is known about how DeepSeek is used and regulated in health care. Objective: This study aimed to (1) systematically review the characteristics of DeepSeek deployed in the top 100 hospitals in China, and (2) compare the performance and risks of DeepSeek between hospital disclosures and research evidence. Methods: We searched the official websites and WeChat accounts of the top 100 hospitals in China and the databases of Web of Science and PubMed, using the terms “DeepSeek” and “large language models.” Searches were limited to records after January 15, 2025, when DeepSeek was first released. All searches were conducted on May 20, 2025, with an update on June 28, 2025. We extracted the basic characteristics of DeepSeek; its aims, evaluation approach, performance, and risks; and hospital regulations. A coding framework was developed covering the application scenarios, evaluation dimensions, and risk sources of LLMs. The risk of bias was assessed using the Joanna Briggs Institute checklist. Results: We identified a total of 58 DeepSeek models in 48 out of the top 100 Chinese hospitals and found 27 studies in the literature. The first hospital deployment of DeepSeek was recorded on February 10, 2025, and deployment rapidly expanded to 37 hospitals within a month. Concurrently, most related research studies (20/27, 74%) were published after May 2025. Among deployments and studies that reported version information, DeepSeek-reasoner (R1) was the most frequently used model, and private deployment was the predominant approach. DeepSeek was mainly used to assist in clinical decision-making, including patient diagnosis and treatment recommendation. Among hospital disclosures, only 36% (21/58) clearly indicated a predeployment assessment, 22% (13/58) presented assessment results, and 9% (5/58) identified potential risks and countermeasures. We found poor transparency in hospital reporting, with none of the disclosures presenting evaluation details. Hospitals were more likely to report higher performance and fewer risks for DeepSeek. Conclusions: This is one of the first scoping reviews to reveal the rapid, widespread deployment of DeepSeek in China’s leading hospitals, primarily for clinical decision support. The deployment of DeepSeek in China’s leading hospitals poses potential risks to patient outcomes and safety. We highlight the urgent need for existing regulations to be expanded to downstream developers and users to promote the responsible use of LLMs in health care. Hospitals need to use a more rigorous validation process and adopt a more transparent reporting policy. The main limitations of this review include the restriction to top-tier hospitals and the inherent constraints of gray literature. These factors should be considered when interpreting the findings.

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: The integration of artificial intelligence (AI) in health care has significant potential, yet its acceptance by health care professionals (HCPs) is essential for successful implementation. Understanding HCPs’ perspectives on the explainability and integrability of medical AI is crucial, as these factors influence their willingness to adopt and effectively use such technologies. Objective: This study aims to improve the acceptance and use of medical AI. From a user perspective, it explores HCPs’ understanding of the explainability and integrability of medical AI. Methods: We performed a mixed systematic review by conducting a comprehensive search in the PubMed, Web of Science, Scopus, IEEE Xplore, and ACM Digital Library and arXiv databases for studies published between 2014 and 2024. Studies concerning an explanation or the integrability of medical AI were included. Study quality was assessed using the Joanna Briggs Institute critical appraisal checklist and Mixed Methods Appraisal Tool, with only medium- or high-quality studies included. Qualitative data were analyzed via thematic analysis, while quantitative findings were synthesized narratively. Results: Out of 11,888 records initially retrieved, 22 (0.19%) studies met the inclusion criteria. All selected studies were published from 2020 onward, reflecting the recency and relevance of the topic. The majority (18/22, 82%) originated from high-income countries, and most (17/22, 77%) adopted qualitative methodologies, with the remainder (5/22, 23%) using quantitative or mixed method approaches. From the included studies, a conceptual framework was developed that delineates HCPs’ perceptions of explainability and integrability. Regarding explainability, HCPs predominantly emphasized postprocessing explanations, particularly aspects of local explainability such as feature relevance and case-specific outputs. Visual tools that enhance the explainability of AI decisions (eg, heat maps and feature attribution) were frequently mentioned as important enablers of trust and acceptance. For integrability, key concerns included workflow adaptation, system compatibility with electronic health records, and overall ease of use. These aspects were consistently identified as primary conditions for real-world adoption. Conclusions: To foster wider adoption of AI in clinical settings, future system designs must center on the needs of HCPs. Enhancing post hoc explainability and ensuring seamless integration into existing workflows are critical to building trust and promoting sustained use. The proposed conceptual framework can serve as a practical guide for developers, researchers, and policy makers in aligning AI solutions with frontline user expectations. Trial Registration: PROSPERO CRD420250652253; https://www.crd.york.ac.uk/PROSPERO/view/CRD420250652253

Objective: This study aims to explore a streamlined risk-adjusted cesarean section rate (RCSR) model and to compare its practical application effects with the traditional RCSR models. Methods: Utilizing obstetric electronic health record (EHR) data from provincial multicenter hospitals, this study establishes a streamlined RCSR model alongside the traditional RCSR model and evaluates the efficacy of both models. Subsequently, the RCSRs of 56 hospitals within the province are calculated and ranked using both models. The consistency of these rankings is then quantified using Kendall's tau coefficient of concordance. Result: Comparison of model effectiveness evaluation of the traditional RCSR model versus the streamlined RCSR model is as follows: AUC (0.840 vs 0.839), accuracy (0.875 vs 0.872), sensitivity (0.690 vs 0.685), specificity (0.898 vs 0.892), positive predictive value (0.908 vs 0.903), negative predictive value (0.664 vs 0.660), and Brier score (0.069 vs 0.067). In the test of the consistency of hospital rankings based on two models, Kendall's tau coefficients were observed to be 0.979 (year 2017), 0.978 (year 2018), and 0.978 (year 2019) over a span of 3 years, with an aggregate coefficient of 0.974. Conclusion: In the realm of model performance evaluation as well as the pragmatic application within hospital settings, the streamlined model exhibits a substantial congruence with the traditional model. Therefore, the streamlined model can effectively serve as a viable surrogate for the traditional model, potentially establishing itself as a refined paradigm for the appraisal of quality in obstetric healthcare services.