JMIR Formative Research最新文献

{"title":"Explainable AI-driven analysis of radiology reports using text and image data: An experimental study.","authors":"Muhammad Tayyab Zamir, Safir Ullah Khan, Alexander Gelbukh, Edgardo Manuel Felipe Riverón, Irina Gelbukh","doi":"10.2196/77482","DOIUrl":"https://doi.org/10.2196/77482","url":null,"abstract":"<p><strong>Background: </strong>Artificial intelligence is increasingly being integrated into clinical diagnostics, yet its lack of transparency hinders trust and adoption among healthcare professionals. The explainable AI (XAI) has the potential to improve interpretability and reliability of AI-based decisions in clinical practice.</p><p><strong>Objective: </strong>This study evaluates the use of Explainable AI (XAI) for interpreting radiology reports to improve healthcare practitioners' confidence and comprehension of AI-assisted diagnostics.</p><p><strong>Methods: </strong>This study employed the Indiana University chest X-ray Dataset containing 3169 textual reports and 6471 images. Textual were being classified as either normal or abnormal by using a range of machine learning approaches. This includes traditional machine learning models and ensemble methods, deep learning models (LSTM), and advanced transformer-based language models (GPT-2, T5, LLaMA-2, LLaMA-3.1). For image-based classifications, convolution neural networks (CNNs) including DenseNet121, and DenseNet169 were used. Top performing models were interpreted using Explainable AI (XAI) methods SHAP and LIME to support clinical decision making by enhancing transparency and trust in model predictions.</p><p><strong>Results: </strong>LLaMA-3.1 model achieved highest accuracy of 98% in classifying the textual radiology reports. Statistical analysis confirmed the model robustness, with Cohen's kappa (k=0.981) indicating near perfect agreement beyond chance, both Chi-Square and Fisher's Exact test revealing a high significant association between actual and predicted labels (p<0.0001). Although McNemar's Test yielded a non-significant result (p=0.25) suggests balance class performance. While the highest accuracy of 84% was achieved in the analysis of imaging data using the DenseNet169 and DenseNet121 models. To assess explainability, LIME and SHAP were applied to best performing models. These models consistently highlighted the medical related terms such as \"opacity\", \"consolidation\" and \"pleural\" are clear indication for abnormal finding in textual reports.</p><p><strong>Conclusions: </strong>The research underscores that explainability is an essential component of any AI systems used in diagnostics and helpful in the design and implementation of AI in the healthcare sector. Such approach improves the accuracy of the diagnosis and builds confidence in health workers, who in the future will use explainable AI in clinical settings, particularly in the application of AI explainability for medical purposes.</p>","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":" ","pages":""},"PeriodicalIF":2.0,"publicationDate":"2025-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145149068","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Response to the Netflix Docuseries \"Big Vape: The Rise and Fall of JUUL\": Mixed Methods Analysis of YouTube Comments Using Qualitative Coding and Topic Modeling.","authors":"Beth Hoffman, Arpita Tripathi, Ariel Shensa, Julia Pengyue Dou, Piper Narendorf, Nishi Hundi, Jaime Sidani","doi":"10.2196/76737","DOIUrl":"10.2196/76737","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;On October 11, 2023, Netflix released the docuseries \"Big Vape: The Rise and Fall of JUUL,\" which chronicled the founding of JUUL, its rise in popularity among youth, and the subsequent public backlash. The official Netflix YouTube channel posted a trailer promoting the docuseries and an official clip from the docuseries. Recent studies have demonstrated the utility of using comments posted under YouTube videos to analyze reactions to the content and discourse around the health topics explored in the video.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;This study aimed to (1) systematically characterize nicotine and tobacco product (NTP)-related comments and replies posted in response to the docuseries trailer and video clip and (2) explore integration of automated topic modeling techniques with traditional human-generated qualitative coding.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;We extracted all comments and replies on the aforementioned YouTube clips 1 month after the docuseries' release (N=532). Research assistants manually double-coded the comments using a systematically developed codebook that assessed for NTP sentiment (pro-NTP, anti-NTP, complex sentiment, or no sentiment) and the presence or absence of specific electronic cigarette (e-cigarette)-related content. Given the substantial amount of comments coded as potential misinformation during the coding process, we conducted an in-depth qualitative content analysis of all comments coded as potential misinformation. Simultaneously, we used word clustering techniques including structural topic modeling to identify the overarching topics.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;Of the 73.8% ( 393/532) relevant comments, 63.6% (250/393) expressed NTP sentiment with 42.8% of these (107/250) expressing pro-NTP sentiment and 18.4% (46/250) expressing complex sentiment. The most frequent content category was potential misinformation (27.5%, 108/393). These 108 comments contained 152 individual pieces of misinformation that were broadly grouped within 6 themes with various numbers of subthemes; the most frequent misinformation theme was that e-cigarette use is completely safe or much safer than smoking (n=80). Other frequently occurring content categories included e-cigarette use is safer than smoking (17.6%, 69/393), and personal experience using e-cigarettes or JUUL (15.5%, 61/393). For topic modeling, we identified 9 topics that we qualitatively assigned into 4 thematic categories: comparisons with other drugs, mentions of government and pharma companies, role of media and parents, and harms associated with nicotine and tobacco products.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;To the best of our knowledge, this is the first study to examine viewer reactions to the docuseries about JUUL. Our analysis of YouTube comments offers insight into current sentiment and misinformation regarding NTPs and highlights the potential utility of using mixed methods to analyze NTP-related social media data, an","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"9 ","pages":"e76737"},"PeriodicalIF":2.0,"publicationDate":"2025-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12448255/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145091558","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Knowledge and Perceptions of AI Among Medical Students in Morocco: Cross-Sectional Study.","authors":"Imad Chakri, Otmane El Khayali, Laila Lahlou","doi":"10.2196/66156","DOIUrl":"10.2196/66156","url":null,"abstract":"<p><strong>Background: </strong>Artificial intelligence (AI) is rapidly transforming medical practice by enhancing diagnostic accuracy, streamlining workflows, and supporting clinical decision-making. However, the integration of AI into health care largely depends on the preparedness and acceptance of future physicians. Therefore, assessing their knowledge and perceptions of AI is crucial. Notably, no study has yet evaluated these factors among medical students in Morocco.</p><p><strong>Objective: </strong>The aim of this study was to describe Moroccan medical students' knowledge and perception of AI.</p><p><strong>Methods: </strong>A cross-sectional, observational study was conducted from February to May 2023 at the Faculty of Medicine and Pharmacy, Agadir, Morocco. All undergraduate medical students from the first to the seventh year were eligible, excluding graduate students. A snowball sampling method was used, with a calculated minimum sample size of 385. To account for potential missing data, and given the target population size of 1150, the sample size was increased by 50%. Data were collected through a validated online questionnaire and analyzed using JAMOVI 2.6.2, with significance set at P<.05.</p><p><strong>Results: </strong>A total of 580 medical students (female n=363, 62.6%; mean age 21.3, SD 2.13 years; response rate 50.4%) participated. While 96% (n=557) had heard of AI, 73.1% (n=424) were unfamiliar with key AI terminologies, only 11% (n=64) understood AI functioning, and 14.8% (n=86) were familiar with everyday AI applications. Objectively, 88.1% (n=511) correctly identified deep learning as a method for automated pattern recognition, with 71.5% (n=415) acknowledging its interpretability challenges. First-cycle students demonstrated significantly higher familiarity with AI terms (83/156, 53.2% vs 51/156, 32.7% vs 22/156, 14.1%; P<.001). In terms of perception, 83% (n=482) viewed AI as a collaborative tool, 84.1% (n=488) anticipated a transformative impact on medicine, 39% (n=227) expected noninterventional medicine to be replaced within a decade, and 57.1% (n=331) believed certain specialties could be supplanted by AI. Regarding AI in medical education, 90% (n=522) supported its integration into the curriculum and 94% (n=546) expected enhanced learning conditions, but only 48.1% (n=279) felt ready to use AI tools upon graduation. Additionally, gender and technology familiarity significantly influenced specific perceptions, with technology-savvy students reporting greater readiness (P<.001) and women more likely to view AI as revolutionary (315/488, 64.5% vs 173/488, 35.5%; P=.02).</p><p><strong>Conclusions: </strong>Medical students' knowledge of AI is still limited, but their awareness of the potential impact of this technology on future practice and their openness to its integration into the medical curriculum constitute a promising basis for the successful implementation of these new concepts in our health care system.</p>","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"9 ","pages":"e66156"},"PeriodicalIF":2.0,"publicationDate":"2025-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12456875/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145091449","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessing the User Experience of the EU Mobile App for Cancer Prevention: Mixed Methods Study.","authors":"Furqan Ahmed, Silvana Romero Saletti, Erica D'Souza, Carolina Espina, David Ritchie, Ana Molina Barceló, Marina Pinto Carbó, Paula Romeo Cervera, Teresa Seum, Hermann Brenner, Stephan Van den Broucke, Maria Krini, Cristiana Fonseca, Patricia Pinto, Diana Krivic, Helena Ros Comesana, Wendy Yared, Hajo Zeeb, Tilman Brand","doi":"10.2196/73844","DOIUrl":"10.2196/73844","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;In 2022, nearly 20 million new cancer cases and 9.7 million deaths occurred globally. Europe, comprising under 10% of the world's population, accounted for over 22% of cases and 20% of deaths, reflecting an aging population, lifestyle risk factors, and extensive screening. With 40% of cancers preventable through modifiable risk factor interventions, effective prevention is essential. The European Code Against Cancer provides evidence-based guidelines that drive health initiatives across Europe. Supported by Europe's Beating Cancer Plan and the EU4Health program, the EU Mobile App for Cancer Prevention was developed to disseminate these recommendations. However, its effectiveness depends on usability across populations with varying digital and health literacy; this study evaluates the app's usability among diverse European populations.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;This study aimed to identify enablers, barriers, and user requirements for the use and maintenance of the English version of the EU Mobile App for Cancer Prevention, focusing on how usability varied across individuals with different levels of digital health literacy and diverse sociodemographic backgrounds. In addition, user feedback on mock wireframes-visual representations of the app's interface and functionality-was gathered to evaluate usability and ease of use, providing insights for tailoring the app design to a broader population.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;We conducted a mixed methods study in 7 European countries with 76 adults aged 19-84 years recruited via purposive quota sampling. Participants completed quantitative usability testing using mock wireframes to perform 10 predefined tasks simulating core app functionalities (eg, profile setup and health goal tracking). We recorded task completion time, success rates, self-reported confidence, and perceived difficulty. Digital health literacy was assessed using the eHealth Literacy Scale (eHEALS) scale. Qualitative data were collected through focus group discussions guided by a semistructured interview guide, and transcripts were analyzed via thematic content analysis. Statistical analyses included descriptive statistics and 1-way ANOVA to explore group differences.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;Overall task completion rates ranged from 75% to 98%, with a median of 86%, indicating general usability. However, usability varied by age, education, and digital health literacy: younger participants and those with higher education and literacy levels reported greater confidence and lower difficulty, whereas older adults and lower-literacy users experienced more challenges. Qualitative analysis identified key themes affecting usability: the need for accessibility (multilingual support and simple language), user-centric design (age-friendly interfaces and intuitive navigation), ethical concerns (data privacy and security), and motivational features (gamification and personalized health goals).&lt;/","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"9 ","pages":"e73844"},"PeriodicalIF":2.0,"publicationDate":"2025-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12448259/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145091497","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Exploring the Role of Telemedicine in Duchenne Muscular Dystrophy: Benefits and Challenges.","authors":"Eliza Wasilewska, Andrzej Wasilewski, Alessandro Onofri, Jan Wasilewski, Dominika Sabiniewicz-Ziajka, Agnieszka Sobierajska-Rek, Jarosław Meyer-Szary, Karolina Śledzińska, Jolanta Wierzba, Marek Niedoszytko, Sylwia Małgorzewicz","doi":"10.2196/77698","DOIUrl":"10.2196/77698","url":null,"abstract":"<p><p>Duchenne muscular dystrophy (DMD) is the most frequent, progressive disease caused by a genetic defect that leads to the production of a nonfunctional form of dystrophin, thereby causing premature death. Ways to improve, adapt, and facilitate the care of people with DMD are still being explored. This viewpoint, developed by an accredited Duchenne center, aims to present current telemedicine options specifically tailored for patients with DMD and to discuss the advantages and limitations of these approaches across various health care domains. As one of the first centers in Poland to implement such an approach, the accredited Duchenne center provides targeted home-based care by using digital platforms and telemedicine tools. Additionally, we explore the potential of telemedicine to support different types of remote communication, including provider-to-provider, between patient/caregiver and provider, and between patient/caregiver and patient/caregiver interactions. This model has the potential to significantly enhance access to specialized care and improve the continuity and quality of life for those living with DMD.</p>","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"9 ","pages":"e77698"},"PeriodicalIF":2.0,"publicationDate":"2025-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12495364/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145091475","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Patient-Reported Experiences With Long-Term Lifestyle Self-Monitoring in Heart Disease: Mixed Methods Study.","authors":"Mayra Goevaerts, Nicole Tenbült-Van Limpt, Willem J Kop, Hareld Kemps, Yuan Lu","doi":"10.2196/76978","DOIUrl":"10.2196/76978","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Lifestyle behaviors strongly predict cardiovascular morbidity and mortality, emphasizing the need for strategies that support sustained lifestyle changes in patients with cardiac disease. Digital health solutions, including wearables, mobile apps, and chatbots, enable self-monitoring of lifestyle behaviors but often face challenges with engagement and usability. While self-monitoring systems can increase awareness and accountability, maintaining user engagement remains crucial for their effectiveness in promoting behavior change and long-term improvements.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;This study evaluated patient experiences with a lifestyle monitoring system combining a web application, health watch, and chatbot. We explored facilitators of and barriers to long-term adherence and assessed the impact of self-monitoring on lifestyle awareness and behavior change in patients with cardiac disease.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;We conducted a mixed methods study with patients who used an eHealth platform for self-monitoring lifestyle behaviors during 1 year following an invasive cardiac procedure. This study included 100 patients (mean age 61.6, SD 10.4 y; n=88, 88% male) comprising both completers (n=57, 57%) and dropouts (n=43, 43%). Patients engaged in quarterly phone interviews and questionnaires and completed an end-of-study questionnaire. Completers participated in a structured evaluation interview; dropouts provided a reason for discontinuation. Quantitative and qualitative data analyses focused on usability, long-term adherence facilitators and barriers, lifestyle awareness, and behavior change.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;Patients completed 157 quarterly questionnaires (n=145, 92.4% by completers and n=12, 7.6% by dropouts) and 217 phone interviews (n=171, 78.8% with completers and n=46, 21.2% with dropouts). In total, 77 patients (of whom n=54, 70% were completers and n=23, 30% were dropouts) completed end-of-study questionnaires, and 98% (56/57) of completers participated in the evaluation interviews. Completers reported higher perceptions of the platform's usefulness, ease of use, and satisfaction (P&lt;.001 in all cases) than dropouts. Dropout reasons linked to self-monitoring (34/43, 79%) included high self-report burden and dissatisfaction with the chatbot, poor overall usability experience, health watch technical challenges causing frustration, limited perceived usefulness, mental stress from self-monitoring, and low motivation. Key facilitators of long-term engagement included routine formation, structured reminders, and minimal effort associated with the wearable. Barriers included repetitive chatbot questions (causing cognitive burden) and technical issues with the health watch. Self-monitoring increased lifestyle awareness among completers, particularly regarding physical activity (25/56, 45%) and nutrition (29/56, 52%), with smaller effects for sleep quality (7/56, 13%) and mental stress (1","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"9 ","pages":"e76978"},"PeriodicalIF":2.0,"publicationDate":"2025-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12489404/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145080670","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparing Media and Law Enforcement Reports on Anti-Asian Hate Incidents During the COVID-19 Pandemic: Data Visualization Approach.","authors":"Young Ji Yoon, Su Hyun Shin, Dongwook Kim, Hee Yun Lee","doi":"10.2196/70881","DOIUrl":"10.2196/70881","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;During the COVID-19 pandemic, anti-Asian hate incidents (AAHIs) increased conspicuously. Literature reports discrepancies in how crimes are reported differently in media and law enforcement data, emphasizing potential biases and inconsistencies in AAHI reporting. Understanding the discrepancies in AAHI reporting between the two sources is crucial for improving documentation procedures and addressing systemic issues in reporting mechanisms.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;This study aimed to (1) present the monthly trends in AAHI counts reported by media and law enforcement data from 2020 to 2021, (2) investigate variations in AAHI counts across states and counties for each year, (3) examine discrepancies in AAHI reporting between the two sources at state and county levels during the same period, and (4) delineate differences in the types and geographic distribution of incidents as represented by the two sources.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;This study used two data sources for AAHIs, media data (n=1288) from The Asian American Foundation and law enforcement data (n=1086) from the Federal Bureau of Investigation, for the 2020-2021 period. Descriptive analyses were conducted to evaluate monthly trends, state and county-level variations, and differences in incident types and locations. Ratios of reported incidents between the two sources were calculated to assess discrepancies. Temporal trends were contextualized within key sociopolitical events to offer insights into reporting dynamics.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;First, both media and law enforcement data presented a sharp increase in reported AAHIs following the first confirmed COVID-19 case in the United States, peaking around March 2020, coinciding with controversial political rhetoric. A second peak occurred from March to April 2021, immediately following the pandemic's peak, and was followed by a decline as the situation improved. Second, in the two data sources, the state-level analysis indicated that California, Texas, New York, and Washington consistently reported the highest AAHI counts. In 2021, there were notable increases in reported incidents in states such as Wisconsin and Illinois. County-level data revealed persistent high counts in California, particularly in Los Angeles County. Ratios of AAHI counts between the two data sources presented significant discrepancies, with higher ratios in California and New York. Finally, the analysis of incident types revealed that media data reported a higher proportion of harassment (477/1288, 37%), while the law enforcement data reported more property-related incidents (239/1086, 22%). Regarding location types, media data frequently reported incidents in public areas (515/1288, 40%) and businesses (361/1288, 28%), whereas law enforcement data reported more incidents occurring in residential settings (201/1086, 18.5%).&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;This study highlighted significant trends and dispar","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"9 ","pages":"e70881"},"PeriodicalIF":2.0,"publicationDate":"2025-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12489402/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145080664","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Quality and Reliability of Transarterial Chemoembolization Videos on TikTok and Bilibili: Cross-Sectional Content Analysis Study.","authors":"Yushuo Niu, Guilan Song, Zheyu Niu, Sijian Xiao, Cuicui Li, Na Han, Hao Wan, Xiaohong Hou","doi":"10.2196/73855","DOIUrl":"10.2196/73855","url":null,"abstract":"<p><strong>Background: </strong>Transarterial chemoembolization (TACE) is a widely used treatment for advanced, unresectable hepatocellular carcinoma, often requiring multiple sessions for optimal efficacy. TikTok and Bilibili have gained widespread popularity as easily accessible sources of health information.</p><p><strong>Objective: </strong>This study aims to assess the quality of the information in Chinese short videos on TACE shared on TikTok and Bilibili.</p><p><strong>Methods: </strong>In November 2024, the top 100 TACE-related Chinese-language short videos on TikTok and Bilibili (a total of 200 videos) were assessed and reviewed. Initially, basic information about the videos was recorded and analyzed. Subsequently, Global Quality Score and the DISCERN tool were used to evaluate the information quality and reliability of the videos on both platforms. Finally, multifactorial analysis was used to identify potential factors influencing the quality of the videos.</p><p><strong>Results: </strong>TikTok is more popular than Bilibili, despite its videos being shorter in length (P<.001). The quality of short videos on TACE found on both platforms was of low quality, with average Global Quality Score scores of 2.31 (SD 0.81) on TikTok and 2.48 (SD 0.80) on Bilibili, as well as DISCERN scores of 1.86 (SD 0.40) on TikTok and 2.00 (SD 0.44) on Bilibili. The number of saves (β=.184, P=.008; β=.176, P=.01) and days (β=.214, P=.002; β=.168, P=.01) since publication were identified as closely related variables to video quality and reliability. Furthermore, the duration of the video was closely related to its reliability (β=.213, P=.002).</p><p><strong>Conclusions: </strong>This study indicates that the quality of TACE-related health information in the top 100 short videos on both Bilibili and TikTok platforms is suboptimal. Patients should exercise caution when relying on health-related information from these platforms. Social media companies should establish review teams with basic medical knowledge. It is essential for the platforms to enhance their recommendation algorithms and implement measures for video quality assessment. Health care professionals should be aware of the limitations of these videos and work to improve their quality.</p>","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"9 ","pages":"e73855"},"PeriodicalIF":2.0,"publicationDate":"2025-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12443354/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145080711","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Online Yoga Pilot Intervention for Black Women at High Cardiovascular Risk: Internet-Based Recruitment and Engagement.","authors":"Candace Crosby Johnson, Pascaline Ezouah","doi":"10.2196/41221","DOIUrl":"10.2196/41221","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Disproportionately adverse heart health outcomes in Black women, characterized by high metabolic syndrome prevalence, underscore the need for innovative, accessible interventions. Digital health strategies, particularly web-based yoga videos, show promise for engaging this high-risk group in health-promoting behaviors.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;This study aimed to evaluate the feasibility and acceptability of a web-based yoga intervention for community-dwelling Black women, providing preliminary data to inform a larger, mixed methods study on reducing cardiometabolic risks.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;In this 4-week pilot study, grounded in Pender's Health Promotion Model, 28 participants engaged in daily online health education and yoga activities through YouTube videos. Using Fitbit trackers, electronic blood pressure monitors, and web-based logs, the study measured metabolic syndrome risk factors and sedentary behavior. Participant experiences were further explored through postintervention focus groups aiming to contextualize the intervention's impact.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;We enrolled 28 women, with a completion rate of 79% (22/28), demonstrating successful recruitment and retention. Participants were an average age of 43.3 years with a mean BMI of 40.9 kg/m&lt;sup&gt;2&lt;/sup&gt;, indicating a high-risk group for metabolic syndrome. Engagement with 2 or more intervention components were significantly correlated with study completion (χ&lt;sup&gt;2&lt;/sup&gt;&lt;sub&gt;1&lt;/sub&gt;=7.14, P=.008). Specifically, viewing over one-half of the instructional videos (χ&lt;sup&gt;2&lt;/sup&gt;&lt;sub&gt;1&lt;/sub&gt;=4.39, P=.04) and daily blood pressure monitoring (χ&lt;sup&gt;2&lt;/sup&gt;&lt;sub&gt;1&lt;/sub&gt;=5.67, P=.02) were key to participant adherence. The intervention was well-received, with 95% (19/20) of survey respondents finding it satisfactory and suitable. Technology use was high, with all participants having access to the internet, 96% (27/28) owning smartphones, and 53% (15/28) having a YouTube account prior to the study. Recruitment was effectively conducted online, primarily via Facebook and a university newsletter, each accounting for 39.3% (11/28) of participants. The qualitative focus group data unveiled 4 major themes: (1) accountability, emphasizing the shift toward self-prioritization and collective health responsibility; (2) increased awareness, highlighting enhanced understanding of health behaviors and metabolic syndrome risks; (3) health benefits, noting observed improvements in blood pressure and stress levels; and (4) unanticipated stressors, identifying external factors that challenged engagement. These insights underscore the intervention's multifaceted impact, from fostering health awareness to navigating external stressors.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;This pilot study demonstrated the feasibility and acceptability of a culturally tailored, online yoga intervention among community-based, Black women at high risk for metabolic syndrom","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"9 ","pages":"e41221"},"PeriodicalIF":2.0,"publicationDate":"2025-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12489401/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145080705","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Stakeholder Perspectives on mHealth Technologies to Prevent Sitting-Acquired Pressure Injuries in Long-Term Care Facilities: Mixed Methods Study.","authors":"Tamara Vos-Draper, Erin Vinoski Thomas, Emily Graybill, Melissa Morrow, Kathleen A Jordan, Pamela R Manley, Sharon E Sonenblum","doi":"10.2196/59590","DOIUrl":"10.2196/59590","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Adults with Alzheimer disease (AD) or Alzheimer disease and related dementias (ADRD) who require a wheelchair to accommodate disease-associated decline in mobility are at elevated risk for pressure injuries. More than half of residents in long-term care (LTC) facilities in the United States experience AD or ADRD. In LTC facilities, bed-based technologies exist to facilitate pressure injury prevention efforts, but similar technologies have not yet been widely evaluated to address sitting-related pressure injuries.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;This study aimed to determine preliminary design inputs from care providers for technology to address sitting-related pressure injury prevention in LTC settings. Specifically, we sought to (1) understand the types and use of sitting-related equipment used in LTC for residents with AD or ADRD, (2) identify challenges faced by nurses and other caregivers when repositioning seated residents, and (3) understand care provider preferences for features of future sitting-related feedback technologies designed to facilitate effective and timely repositioning.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;Surveys (n=30) and semistructured interviews (n=9) of administrative and direct care providers in LTC facilities were administered. Survey results were summarized, and we used thematic qualitative analysis of interview responses to develop themes around challenges experienced by care providers and their perceptions about how technologies could facilitate the prevention of sitting-related pressure injuries.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;Survey respondents endorsed using many sitting surfaces for LTC residents with memory loss, such as padded reclining chairs, bedside or dining chairs, and wheelchairs with cushions. All indicated that shared equipment is provided by the facility, and 43% of respondents reported having access to a seating specialist at their facility. Sitting time was typically up to 12 hours per day. Themes related to pressure injury prevention in the LTC context, specific to those with memory loss, included (1) barriers to repositioning seated residents vary with the degree of memory loss, (2) care providers are aware of guidelines and policies around the 2-hour repositioning schedule, and (3) care providers are interested in technologies that have relative value over added burden. Care providers expressed interest in mobile health (mHealth) technologies that provide automatic repositioning in later stages of memory loss, delivery of cues for residents with mild memory loss to encourage independent repositioning, and tools to monitor resident sitting and pressure-related outcomes.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;These findings highlight the complexity of addressing the repositioning needs of seated LTC residents with AD or ADRD using mHealth technologies due to changes as the disease progresses. mHealth technologies should encourage more independence by residents experienci","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"9 ","pages":"e59590"},"PeriodicalIF":2.0,"publicationDate":"2025-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12489400/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145080708","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}