Frontiers in digital health最新文献

{"title":"Development of a novel artificial intelligence algorithm for interpreting fetal heart rate and uterine activity data in cardiotocography.","authors":"Rohit Pardasani, Renee Vitullo, Sara Harris, Halit O Yapici, John Beard","doi":"10.3389/fdgth.2025.1638424","DOIUrl":"10.3389/fdgth.2025.1638424","url":null,"abstract":"<p><strong>Introduction: </strong>Cardiotocography (CTG) assesses fetal well-being through measurements of fetal heart rate (FHR) and uterine activity (UA). Manual visual assessment of fetal tracings is variable due to the subjective nature of their interpretation. Artificial intelligence (AI) using automatic signal processing may be leveraged to support consistent, comprehensive interpretations. This study demonstrated the development and training of a novel AI algorithm that analyzes and interprets certain clinical events and parameters calculated during labor to assist with clinical decisions.</p><p><strong>Methods: </strong>Fetal tracings sourced from 19 birthing centers through a US-based healthcare delivery organization were clinically interpreted, labeled, quality checked, and ratified by clinicians to be included in the study. The algorithm using deep learning and rule-based techniques was developed to identify segments of interest (accelerations, decelerations, and contractions). A three parallel one-dimensional Unet design with two inputs (FHR and UA) and one channel output each (for accelerations, decelerations, and contractions) was selected as the final architecture. Algorithm performance was evaluated through recall (sensitivity), precision, <i>F</i>1 score, and duration and numerical ratios.</p><p><strong>Results: </strong>A total of 133,696 patient files were used to create fetal tracings. After the exclusion, labeling, and ratification processes, the final datasets included 1,600 tracings for training, 421 for validation, and 591 for testing. The model provided promising performance and achieved <i>F</i>1 scores of 0.803 for accelerations, 0.520 for decelerations, and 0.868 for contractions on the final test set, with a 91.5% predicted baseline accuracy (difference of ≤5 bpm) compared to clinician interpretation.</p><p><strong>Conclusion: </strong>This study demonstrates the successful development of a novel AI algorithm utilizing FHR and UA data to analyze and interpret fetal tracing events and parameters. The algorithm may have potential to enhance patient care by supporting bedside clinician CTG interpretation.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"7 ","pages":"1638424"},"PeriodicalIF":3.2,"publicationDate":"2025-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12479546/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145208437","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":"Beyond weight loss: digital therapeutic for behavioral change and psychological well-being for individuals with overweight and obesity in a primary healthcare setting-A randomized controlled pilot study.","authors":"Marthe Isaksen Aukan, Maria Arlèn Larsen, Tone Iren Melan, Øyvind Olav Salvesen","doi":"10.3389/fdgth.2025.1671649","DOIUrl":"10.3389/fdgth.2025.1671649","url":null,"abstract":"<p><strong>Objective: </strong>Mobile health (mHealth) through digital therapeutics (DTx) offer a promising approach to obesity management. This study evaluated the effectiveness of the Lifeness DTx for obesity care and its effect on anthropometrics, reward-related eating behaviors and quality of life in individuals with overweight and obesity within a community-based healthcare setting.</p><p><strong>Methods: </strong>A 12-week randomized controlled trial was conducted. Adults (BMI ≥ 27 kg/m², and central obesity) were recruited from municipal Healthy Life Centers in Norway. The intervention group (IG) received standard care plus full DTx app with program functionality and digital follow-up, whereas the control group (CG) received standard care with limited app functions and no DTx program. Outcome variables were measured at baseline and after 12 weeks.</p><p><strong>Results: </strong>No significant changes in body weight, or differences between groups were observed at W12. The IG showed reductions in waist circumference (-3.4 cm, <i>p</i> = 0.008, <i>d</i> = -0.926), waist-to-height ratio (-0.02, <i>p</i> = 0.008, <i>d</i> = -0.929), improvements on hedonic eating behavior, indicated by reduced disinhibition (-1.6, <i>p</i> = 0.013, <i>d</i> = -0.907), as well as increased quality of life (+5.0, <i>p</i> = 0.019, <i>d</i> = 0.899). Both groups increased self-esteem (IG +9.8, <i>p</i> = 0.018, <i>d</i> = 0.911, and CG +12, <i>p</i> = 0.050, <i>d</i> = 0.838).</p><p><strong>Conclusion: </strong>The DTx intervention was associated with improvements in central adiposity, reward-related eating behaviors, and psychological well-being beyond weight loss. These findings provide preliminary evidence that digital therapeutics may represent a feasible and scalable approach to support personalized obesity care in primary healthcare settings. Larger, adequately powered trials are needed to confirm these results.</p><p><strong>Clinical trial registration: </strong>clinicaltrials.gov, identifier NCT06667843 (Initial Release: 10/15/2024).</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"7 ","pages":"1671649"},"PeriodicalIF":3.2,"publicationDate":"2025-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12479526/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145208520","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":"Evaluating CFIR 2.0 in identifying digital twin implementation challenges in healthcare: bridging the dichotomy between engineering and healthcare communities.","authors":"Md Doulotuzzaman Xames, Taylan G Topcu, Sarah H Parker, Vivian Zagarese, John W Epling","doi":"10.3389/fdgth.2025.1611225","DOIUrl":"10.3389/fdgth.2025.1611225","url":null,"abstract":"<p><strong>Background: </strong>Digital twin (DT) technology holds significant promise for healthcare systems (HSs) due to real-time monitoring based on streaming operational data and <i>a priori</i> analysis capabilities without interrupting clinical workflows. However, the sociotechnical complexity of HSs presents challenges for effective DT implementation. A dichotomy also exists between the engineering and implementation science (IS) communities regarding DT implementation challenges. This study assesses the efficacy of the updated Consolidated Framework for Implementation Research (CFIR 2.0) in identifying DT implementation challenges, aiming to bridge the knowledge gap between IS and DT communities.</p><p><strong>Methods: </strong>This study presents findings from a DT implementation case study in a family medicine clinic, an operational healthcare microsystem. It adopts CFIR 2.0 to guide semi-structured interviews with four key stakeholder groups (e.g., family medicine specialists, engineers, organizational psychologists, and implementation scientists). Participants (<i>N</i> = 8) were purposively sampled based on their roles in DT implementation. Thematic coding categorized interview data into seven themes: technological, data-related, financial and economic, regulatory and ethical, organizational, operational, and personnel. Thematic data were then cross-analyzed with challenges documented in DT literature to assess how effectively CFIR 2.0 identifies DT implementation challenges.</p><p><strong>Results: </strong>Challenges were grouped into three categories: (i) shared challenges captured by both IS and DT communities, (ii) CFIR 2.0-identified challenges overlooked in DT literature, and (iii) challenges documented in DT research but not captured through CFIR 2.0-guided interviews. While there was strong overlap between the communities, a formidable gap also remains. CFIR 2.0 effectively identified a diverse set of issues-predominantly in organizational, financial, and operational themes-including many overlooked by the DT community. However, it was less effective in capturing technological and data-related barriers critical to DT performance, such as modeling, real-time synchronization, and sensor reliability.</p><p><strong>Conclusions: </strong>CFIR 2.0 effectively identifies organizational and operational barriers to DT implementation in healthcare but falls short in addressing technological and data-related complexities. This study highlights the need for interdisciplinary collaboration for the successful transition of emerging DT technologies into practice to maximize their impact on HS efficiency and patient outcomes.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"7 ","pages":"1611225"},"PeriodicalIF":3.2,"publicationDate":"2025-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12477195/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145201991","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":"What are the functionalities and features of mobile health record apps supporting persons experiencing social exclusion? A systematic literature review.","authors":"Felicien Izaturwanaho, Marie E Ward, Clíona Ní Cheallaigh, Maeve Moran, Geraldine Fitzgerald, David Mockler, Una Geary, Siobhán Corrigan","doi":"10.3389/fdgth.2025.1629289","DOIUrl":"10.3389/fdgth.2025.1629289","url":null,"abstract":"<p><strong>Background: </strong>Research into mobile health record apps has focused on narrow outcomes, such as medication adherence for persons experiencing chronic conditions. However, no review has examined their use in the context of social exclusion. Persons experiencing social exclusion (PESE) face complex health needs, limited healthcare access, and increased exposure to traumatic life experiences. It is imperative to consider a trauma-informed and integrated care approaches when developing an app for them, and they should be involved as key stakeholders to ensure equitable care. This review examined these apps' functionalities and features that support PESE in relation to their reported outcomes and the delivery of a trauma-informed and/or integrated care.</p><p><strong>Methods: </strong>A systematic search of ten databases: Web of Science Core Collection, Medline, PsycINFO, CINAHL, Cochrane, Embase, Scopus, ProQuest Dissertations and Theses A&I, Lenus and OpenGrey International were undertaken, and was supplemented with non-indexed and grey literature. Searches were undertaken in April 2024 in English with no date limit, and used the PRISMA 2020 guidelines. Studies were deemed eligible if they met the SPIDER framework criteria.</p><p><strong>Results: </strong>One thousand three hundred and thirty-two papers were found eligible for the review, of which eleven qualified for inclusion following screening and quality assessment using QATSDD and MMAT tools. Four themes were found (supporting integrated and connected care; enhancement of user engagement and care coordination; improving data accuracy and access to care; and provision of ongoing monitoring and feedback) related to apps' functionalities and features, which in turn were linked to reported outcomes. Although a few of these apps' functionalities and features were aligned with the six principles of trauma-informed care, none of them were implemented considering a trauma-informed care and/or integrated care.</p><p><strong>Conclusion: </strong>This review provided insights into the complexities of implementing a mobile health record app for PESE. However, limited available data restricted a comprehensive understanding of these apps' functionalities and features in their specific implementation settings in relation to their reported outcomes. Next steps include translating these findings into survey and interview questions to identify end-user requirements for developing an app for PESE from a trauma-informed perspective to promote integrated care.</p><p><strong>Systematic review registration: </strong>PROSPERO CRD42024535090.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"7 ","pages":"1629289"},"PeriodicalIF":3.2,"publicationDate":"2025-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12477232/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145202058","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":"How to evaluate user experience in digital health? A scoping review of questionnaires in virtual reality applications.","authors":"Francesca Bruni, Valentina Mancuso, Elisa Pedroli","doi":"10.3389/fdgth.2025.1561364","DOIUrl":"10.3389/fdgth.2025.1561364","url":null,"abstract":"<p><p>The exponential growth and integration of virtual reality technology in clinical environments necessitates a comprehensive user experience evaluation. This assessment is critical for clinical populations and geriatric cohorts presenting peculiar needs and expectations. Despite the longstanding conceptual framework of user experience, a consensus regarding its definition and optimal evaluation methodologies remains elusive, especially within healthcare contexts. This systematic scoping review examines state-of-the-art questionnaire-based instruments for assessing user experience in healthcare virtual reality applications, synthesizing current evaluation approaches and identifying key user experience dimensions. Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses for Scoping Review guidelines, we analyzed articles employing questionnaires to evaluate user experience in virtual reality-based health applications for adults. Following a meticulous screening process of 325 papers across PubMed, Web of Science, and Embase databases, 17 studies met our inclusion criteria. Studies predominantly used multiple and diverse questionnaires exploring several dimensions. Eight key user experience dimensions emerged: usability and functionality, aesthetics of design, engagement, emotional state, presence, realism of environments, side effects, and motivation and intention of use. Current evaluation lacks standardization and theoretical consistency. We propose a comprehensive eight-domain framework and recommend integrating multidisciplinary expertise, implementing longitudinal evaluation approaches, and developing psychometrically validated instruments. These findings provide essential guidance for improving patient outcomes and healthcare delivery efficacy through optimized virtual reality-based implementation.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"7 ","pages":"1561364"},"PeriodicalIF":3.2,"publicationDate":"2025-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12477235/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145202018","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":"A systematic review of ethical considerations of large language models in healthcare and medicine.","authors":"Muhammad Fareed, Madeeha Fatima, Jamal Uddin, Adeel Ahmed, Muhammad Awais Sattar","doi":"10.3389/fdgth.2025.1653631","DOIUrl":"10.3389/fdgth.2025.1653631","url":null,"abstract":"<p><p>The rapid integration of large language models (LLMs) into healthcare offers significant potential for improving diagnosis, treatment planning, and patient engagement. However, it also presents serious ethical challenges that remain incompletely addressed. In this review, we analyzed 27 peer-reviewed studies published between 2017 and 2025 across four major open-access databases using strict eligibility criteria, robust synthesis methods, and established guidelines to explicitly examine the ethical aspects of deploying LLMs in clinical settings. We explore four key aspects, including the main ethical issues arising from the use of LLMs in healthcare, the prevalent model architectures employed in ethical analyses, the healthcare application domains that are most frequently scrutinized, and the publication and bibliographic patterns characterizing this literature. Our synthesis reveals that bias and fairness ( <math><mi>n</mi> <mo>=</mo> <mn>7</mn></math> , 25.9%) are the most frequently discussed concerns, followed by safety, reliability, transparency, accountability, and privacy, and that the GPT family predominates ( <math><mi>n</mi> <mo>=</mo> <mn>14</mn></math> , 51.8%) among examined models. While privacy protection and bias mitigation received notable attention in the literature, no existing review has systematically addressed the comprehensive ethical issues surrounding LLMs. Most previous studies focus narrowly on specific clinical subdomains and lack a comprehensive methodology. As a systematic mapping of open-access literature, this synthesis identifies dominant ethical patterns, but it is not exhaustive of all ethical work on LLMs in healthcare. We also synthesize identified challenges, outline future research directions and include a provisional ethical integration framework to guide clinicians, developers, and policymakers in the responsible integration of LLMs into clinical workflows.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"7 ","pages":"1653631"},"PeriodicalIF":3.2,"publicationDate":"2025-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12460403/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145187461","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":"Synthetic patient and interview transcript creator: an essential tool for LLMs in mental health.","authors":"Aleyna Warner, Jeffrey LeDue, Yutong Cao, Joseph Tham, Timothy H Murphy","doi":"10.3389/fdgth.2025.1625444","DOIUrl":"10.3389/fdgth.2025.1625444","url":null,"abstract":"<p><p>Developing high-quality training data is essential for tailoring large language models (LLMs) to specialized applications like mental health. To address privacy and legal constraints associated with real patient data, we designed a synthetic patient and interview generation framework that can be tailored to regional patient demographics. This system employs two locally run instances of Llama 3.3:70B: one as the interviewer and the other as the patient. These models produce contextually rich interview transcripts, structured by a customizable question bank, with lexical diversity similar to normal human conversation. We calculate median Distinct-1 scores of 0.44 and 0.33 for the patient and interview assistant model outputs respectively compared to 0.50 ± 0.11 as the average for 10,000 episodes of a radio program dialog. Central to this approach is the patient generation process, which begins with a locally run Llama 3.3:70B model. Given the full question bank, the model generates a detailed profile template, combining predefined variables (e.g., demographic data or specific conditions) with LLM-generated content to fill in contextual details. This hybrid method ensures that each patient profile is both diverse and realistic, providing a strong foundation for generating dynamic interactions. Demographic distributions of generated patient profiles were not significantly different from real-world population data and exhibited expected variability. Additionally, for the patient profiles we assessed LLM metrics and found an average Distinct-1 score of 0.8 (max = 1) indicating diverse word usage. By integrating detailed patient generation with dynamic interviewing, the framework produces synthetic datasets that may aid the adoption and deployment of LLMs in mental health settings.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"7 ","pages":"1625444"},"PeriodicalIF":3.2,"publicationDate":"2025-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12460306/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145187534","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":"Antibiotic stewardship through clinical data digitization: perceived opportunities and obstructions by medical doctors from semi-urban setting in central India.","authors":"Mala Kanthali, Gautam Bhagwat, Ashish Pathak, Manju Purohit","doi":"10.3389/fdgth.2025.1652086","DOIUrl":"10.3389/fdgth.2025.1652086","url":null,"abstract":"<p><strong>Background: </strong>Antibiotic resistance is a global public health concern. Inadequate record-keeping and irrational antibiotic prescriptions are challenging factors for antibiotic stewardship. This study explores the perceptions of medical doctors in a semi-urban setting in India, regarding the role of clinical data digitization in mitigating antibiotic resistance.</p><p><strong>Methods: </strong>The study was conducted at R D Gardi Medical College located in a semi-urban district of Central India. Qualitative data from 20 medical doctors from government and private sector were collected through in-depth, semi-structured interviews of which 18 interviews were analyzed using thematic analysis following Braun and Clarke's framework.</p><p><strong>Results: </strong>Two major themes emerged from four overarching subthemes: (1) digitization enhances accountability and continuity of care, (2) potential for local antimicrobial surveillance, (3) infrastructural and technological barriers to adoption, and (4) the necessity of government support and capacity building. The participants believe that digitization could help in rational antibiotic prescription if there is a government mandate and infrastructure feasibility in resource constrained settings.</p><p><strong>Conclusion: </strong>Clinicians in semi-urban India perceive the digitization of clinical data as a promising tool to combat antibiotic resistance. However, systemic and infrastructural challenges must be addressed to utilize its full potential.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"7 ","pages":"1652086"},"PeriodicalIF":3.2,"publicationDate":"2025-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12459205/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145152059","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":"Examining the use of different message categories to communicate AMR: a content analysis of instagram posts.","authors":"Sana Parveen, Wei-Jan Chang, Patricia McHugh, Akke Vellinga","doi":"10.3389/fdgth.2025.1564584","DOIUrl":"10.3389/fdgth.2025.1564584","url":null,"abstract":"<p><strong>Background/objectives: </strong>Antimicrobial resistance (AMR) is a major worldwide concern with severe implications for public health, contributing to almost 5 million deaths in 2019. One of the main causes of AMR is overuse and misuse of antibiotics, which can be addressed by increasing awareness and educating the public on this issue. Studies have demonstrated the potential of social media to educate the public and influence behaviour. Instagram's unique features, such as its visual nature and user-friendly interface, make it well-suited for exploring health behaviours and disseminating information on various health topics. Research shows 65.4% of young adults (18-36 years old) use Instagram as their main source of information.</p><p><strong>Methods: </strong>This study examined AMR posts from Instagram from January 1, 2017 to July 15, 2023. In total, 4,606 images and videos were initially extracted which corresponded to 3,261 Instagram posts. After data cleaning, a final dataset of 574 posts were categorised into 6 message categories which were humour, shock/disgust/fear, personal stories/statements, educational/informative, opportunistic and advocacy.</p><p><strong>Results: </strong>The most common post category was educational/informative (78%) and humour was the least common (2%). We also looked at the average engagement (likes) with these posts, the educational/informative category received the most likes per post (mean of 30). The fear/shock/disgust category received 25 likes per post, humour and personal stories/statements 18 and 21 respectively.</p><p><strong>Conclusions: </strong>Our study shows Instagram has hardly been used for AMR interventions. An important population group of young adults who use Instagram as their main source of information, is missed in public health messaging on AMR.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"7 ","pages":"1564584"},"PeriodicalIF":3.2,"publicationDate":"2025-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12457341/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145151536","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":"Alter egos alter engagement: perspective-taking can improve disclosure quantity and depth to AI chatbots in promoting mental wellbeing.","authors":"Christopher You, Rashi Ghosh, Melissa Vilaro, Roshan Venkatakrishnan, Rohith Venkatakrishnan, Andrew Maxim, Xuening Peng, Danish Tamboli, Benjamin Lok","doi":"10.3389/fdgth.2025.1655860","DOIUrl":"10.3389/fdgth.2025.1655860","url":null,"abstract":"<p><strong>Introduction: </strong>Emotionally intelligent AI chatbots are increasingly used to support college students' mental wellbeing. Yet, adoption remains limited, as users often hesitate to open up due to emotional barriers and vulnerability. Improving chatbot design may reduce some barriers, but users still bear the emotional burden of opening up and overcoming vulnerability. This study explores whether <i>perspective-taking</i> can support user disclosure by addressing underlying psychological barriers.</p><p><strong>Methods: </strong>In this between-subjects study, 96 students engaged in a brief reflective conversation with an embodied AI chatbot. <b>Perspective-Taking</b> participants defined and imagined a designated other's perspective and responded from that viewpoint. <b>Control</b> participants provided self-information and responded from their own perspective. Disclosure was measured by quantity (word count) and depth (information, thoughts, and feelings). Additional immediate measures captured readiness, intentions for mental wellbeing, and attitudes toward the chatbot and intervention.</p><p><strong>Results: </strong><b>Perspective-Taking</b> participants disclosed significantly greater quantity, overall depth, thoughts depth, and frequencies of high disclosures of thoughts and information. Both groups showed significant improvements in readiness and intention to address mental wellbeing, with no difference in improvement magnitude. However, <b>Control</b> participants reported significantly lower (better) skepticism towards the intervention and greater increases in willingness to engage with AI chatbots comparatively.</p><p><strong>Discussion: </strong>This study highlights how perspective-taking and distancing may facilitate greater disclosure to AI chatbots supporting mental wellbeing. We explore the nature of these disclosures and how perspective-taking may drive readiness and enrich the substance of disclosures. These findings suggest a way for chatbots to evoke deeper reflection and effective support while potentially reducing the need to share sensitive personal self-information directly with generative AI systems.</p>","PeriodicalId":73078,"journal":{"name":"Frontiers in digital health","volume":"7 ","pages":"1655860"},"PeriodicalIF":3.2,"publicationDate":"2025-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12457298/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145152078","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}