Background: Smoking continues to be a leading cause of preventable morbidity and mortality, and more than 480,000 Americans die annually due to smoking-related illness attributable to smoking and secondhand smoke. More advanced, responsive, and tailored digital interventions using machine learning and artificial intelligence may be a valuable tool for successful smoking cessation referrals. Objective: This study used the dynamic systems development method to incorporate patient and consumer sources of conversational data to develop a technology-assisted motivational interviewing (TAMI) chatbot, a digital agent using machine learning models to deliver motivational interviewing (MI) for tobacco cessation. Methods: During the functional model iteration phase, user-centered design interviews with smokers (n=3) informed the creation of TAMI. The design and build phase involved the use of existing datasets to guide the incorporation of MI-consistent utterances, language recognition, and topic classification to guide a discussion about smoking, and providing a tailored quit plan if indicated. During the implementation phase, user experience interviews with randomly selected participants (n=9) in a pilot trial discussed their experiences with TAMI. Results: User-centered design interviews indicated a desire for a chatbot that was engaging and adaptable to personal interests in quitting smoking. Inductive analysis of user experience interviews revealed that anonymity, regular reminders, and a humanized experience facilitated engagement with TAMI, but technical glitches, chatbot misunderstandings, and issues with rapport were barriers to engagement. Conclusions: Informed by user input and patient and consumer datasets, TAMI can use MI skills to elicit change talk and/or accurately evaluate readiness for tobacco cessation. Further development will enhance TAMI’s ability to seamlessly engage with users when discussing behavior change and assist underserved populations achieve improvements in a variety of health behavior goals.
<img src="https://jmir-production.s3.us-east-2.amazonaws.com/thumbs/a42048e0fe380701c4eca3d0bc79f43b" />
Preliminary Evaluation of a Large Language Model–Powered Chatbot for Osteoporosis Self-Management Education: Formative Randomized Controlled Trial
<strong>Background:</strong> With the increasing burden of chronic diseases, self-management education (SME) is crucial. Traditional SME based on face-to-face delivery by clinicians is resource-intensive, and general digital tools such as web-based platforms often provide limited interactivity for patient learning. Although chatbots based on large language models (LLMs) show promise in interactivity, their real-world effectiveness lacks empirical evidence. <strong>Objective:</strong> This study aimed to explore the feasibility and preliminary effectiveness of an LLM-based chatbot specifically designed for osteoporosis SME. <strong>Methods:</strong> A formative randomized controlled trial was conducted in a tertiary hospital from February 2024 to March 2025. Adults aged ≥18 years with osteoporosis were recruited and randomly assigned (1:1) to either the intervention (OPBot) group or a control group receiving traditional health education. The chatbot provided interactive educational content and question-and-answer support, while the control group received face-to-face education and written materials. Osteoporosis knowledge was assessed using the Osteoporosis Knowledge Assessment Tool at baseline and discharge. Nurses’ time spent on health education was self-recorded during each intervention session and aggregated across sessions. Adherence to disease management was assessed at 1, 3, and 6 months after discharge via telephone using Likert-scale questionnaires. The reliability of OPBot responses was evaluated by 2 clinician assessors using a 5-point Likert scale, with interrater agreement calculated using Cohen <i>κ</i>. Group comparisons were conducted using 2-tailed independent <i>t</i> tests, Mann-Whitney <i>U</i> tests, and chi-square tests, and adherence outcomes were analyzed using mixed-effects models. <strong>Results:</strong> A total of 100 participants were randomized; 12% (12/100) were excluded due to loss to follow-up, refusal of the second knowledge assessment, or death, leaving 88% (88/100) participants for analysis (n=45, 51.1% in the OPBot group and n=43, 48.9% in the control group). The OPBot group showed significantly higher postintervention knowledge scores than the control group (median 80.0, IQR 70.0-89.0 vs median 75.0, IQR 65.5-80.0; <i>P</i>=.01). Nurses in the OPBot group spent lesser time on SME than those in the control group (median 5.0, IQR 2.0-17.0 vs median 23.0, IQR 20.0-25.0 minutes; <i>P</i><.001). For adherence outcomes, a significant group×time interaction was observed for calcium supplement intake (odds ratio 1.49, 95% CI 1.08-2.06; <i>P</i>=.02), indicating differing adherence trajectories over time. The OPBot group also showed higher odds of consuming calcium-rich foods across time points (odds ratio 2.87, 95% CI 1.04-7.89; nominal <i>P</i>=.04), although this association did not remain significant after Holm correction. No significant effects were observed for sun exposure (<i>P</i>=.56), exercise (<i>P</i>=.79), or total adherence scores (<i>P</i>=.33). In the question-and-answer module, most OPBot responses were rated as highly reliable 89.4% (76/85), with high interrater agreement (Cohen <i>κ</i>=0.83). <strong>Conclusions:</strong> LLM-based chatbots specifically designed for osteoporosis SME may improve patient knowledge, supporting adherence behaviors, and reducing healthcare workload. However, further large-scale studies are needed to confirm these findings.
Pre-Exposure Prophylaxis Adherence and HIV Self-Testing App Among Women in the South Bronx: 12-Month Usability, Acceptability, and Feasibility Study
Background: HIV pre-exposure prophylaxis (PrEP) is underused by cis- and transgender women despite a significant HIV burden. Smartphone technologies are promising tools to support HIV prevention but have yet to be assessed in women. Objective: We conducted a 12-month feasibility study to assess the use and acceptability of a mobile phone app, SmartPrEP, designed to support PrEP adherence and HIV self- and partner-testing among women living in an area of elevated HIV burden in New York City. Methods: Nonpregnant adult cisgender and transgender women who met US PrEP eligibility criteria and were PrEP naïve, reported PrEP use for <3 months, or had inconsistent PrEP use were eligible. Participants received oral PrEP and HIV self-testing kits and downloaded the SmartPrEP app, which sent daily reminders to take PrEP and record adherence through the app. PrEP adherence was assessed based on participants’ self-recorded average doses per week as entered in the app. Sexual behaviors and app acceptability were evaluated quarterly by interviewer-administered questionnaires. Results: From February 2022 to August 2023, 40 participants were enrolled in the study. Median age was 30 (IQR 24-35) years, 70% (28/40) identified as cisgender women, 30% (12/40) as transgender women, 48% (19/40) as Hispanic, and 35% (14/40) as Black. At baseline, the majority, 80% (32/40), had no history of PrEP use, and 65% (26/40) reported that they did not believe they were at risk of HIV. However, 90% (36/40) reported ≥1 and 25% (10/40) reported >4 HIV risk behaviors in the past 6 months, with 58% (23/40) reporting anal or vaginal sex with more than 1 partner. Over the course of the study, although 8 participants withdrew early, and 14 were lost to follow-up, there were 2 pregnancies and 1 HIV seroconversion. PrEP adherence was low, with 80% (32/40) recording <3 doses per week, 17% (7/40) recording 3‐5 doses, and 3% (1/40) recording ≥6 doses per week. PrEP adherence averaged over the second half of study participation was lower than adherence in the first half, with only 10% (4/40) recording >3 doses per week compared to 20% (8/40). In total, 4 participants conducted HIV self- or partner-testing using SmartPrEP during study follow-up. App acceptability assessed at month 12 was moderate to high (median score 3.71 of max 5, IQR 3.47‐4.16). Conclusions: Despite consistently rating the app as acceptable and receiving quarterly HIV testing and counseling, most participants did not achieve optimal PrEP adherence, demonstrating the limitations of this mobile health app among women at elevated risk of HIV. Active navigation and strategies to address gaps in risk perception among women will remain critical, as new, long-acting formulations of PrEP become available. Trial Registration: ClinicalTrials.gov NCT05111119; https://clinicaltrials.gov/study/NCT05111119
Recommendations for Research and Clinical Implementation of Ambulatory Assessment, Mood Monitoring, Digital Phenotyping, and Remote Measurement Technology in Mood Disorders: Synthesis of Systematic Review Findings
Background: Ambulatory assessment and active and passive monitoring all offer a real-time, flexible approach to assessing mood and behavior in mood disorders. Despite their potential, concerns remain regarding the performance, usability, adherence, and potential safety of these tools. Objective: This study synthesizes the findings from 7 systematic reviews, integrating quantitative and qualitative data from randomized trials, observational studies, and user experience research to evaluate the performance, feasibility, acceptability, and clinical impact of ambulatory assessment and mood monitoring in people with depression and bipolar disorder. We assessed studies over the medium or long term (3 months or more). Methods: A summary of a series of systematic reviews was carried out by the authors—including meta-analyses (for quantitative data) and meta-syntheses (for qualitative data). Eight electronic databases were searched, and mixed methods studies were included. Studies were assessed for risk of bias. The results were checked for coherence, and recommendations were made by individuals with lived experience, methodologists, and psychiatrists. GRADE (Grading of Recommendations Assessment, Development, and Evaluation) was used to assess the quality and strength of the evidence. Results: The 111 included studies included 19,945 participants and used 69 different ambulatory assessment protocols or mood-monitoring interventions. Key barriers to implementation were identified, including performance inconsistency, adverse effects, and user disengagement. Evidence-based recommendations are provided to guide future clinical and research applications. Conclusions: Ambulatory assessment and mood monitoring hold promise in research and clinical practice, yet their implementation requires more rigorous evaluation, greater personalization, and responsible, user-centered design. Crucially, these measures can add granularity and confirmation, but additional context is often required, and none of these measures are robust enough yet to replace current outcomes.
<img src="https://jmir-production.s3.us-east-2.amazonaws.com/thumbs/a83582c454b5b6028210d2cb063318df" />
AI-Powered Blood Test Detects Early Retinal Damage in Diabetes
Scientists have developed an AI-assisted prediction tool that can identify patients with type 2 diabetes at high risk of developing diabetic retinal neurodegeneration (DRN) before symptoms appear. Their findings were published today in the journal PLOS Medicine.
“Our study suggests that early retinal nerve damage in diabetes leaves measurable signals in the blood,” write the authors of the study, led by Wei Wang, MD, PhD, associate professor at the Guangdong Provincial Clinical Research Center for Ocular Diseases. “These findings suggest that a simple blood test analyzed with artificial intelligence may help identify people with diabetes who are at highest risk of early retinal nerve damage, well before visible damage appears on the retina.”
Type 2 diabetes affects more than half a billion people worldwide, carrying with it an increased risk of long-term complications including progressive neurodegeneration. Retinal nerves are among the earliest tissues to be damaged, which can eventually lead to severe visual impairment and vision loss. However, current diagnostic methods can only detect DRN once the retina has already suffered irreversible damage.
Wang and colleagues developed a machine learning algorithm called Pro-DRN using data from 1,218 participants in the Guangzhou Diabetic Eye Study, all of whom were diagnosed with type 2 diabetes but had not yet developed DRN at the time of enrollment. The AI model integrated proteomics data from blood samples with yearly retinal images collected over a six-year follow-up period.
This led to the identification of 71 proteins associated with the development of DRN. Among them, the proteins most consistently driving accurate predictions were ACTA2, COL6A3, and HSPG2, which are key structural components involved in maintaining the integrity of the nerve and muscle tissue in the eyes. These results were then validated in an independent cohort of 502 patients from UK Biobank, where the core effects and protein signals were reproduced.
Pro-DRN has been deployed as an interactive, web-based risk assessment tool that doctors can use to support early DRN screening and monitor patient evolution over time. Individuals identified as being at high risk of DRN could benefit from more frequent checkups and early interventions aimed at preventing or slowing down progressive neurodegeneration.
Because DRN is one of the first symptoms of nerve degeneration induced by diabetes, early detection could also signal the onset of nerve injury elsewhere in the body. Such damage can contribute to cognitive impairment, dementia, and peripheral neuropathy, which can cause loss of sensation and motor control in the hands, feet, and other extremities. A single eye test could therefore provide valuable insights into the overall health of the nervous system.
In addition, the proteins identified to be involved in DRN progression could be investigated as potential targets for the development of novel therapies. Furthermore, the AI-based tool could also prove valuable for the selection and stratification of participants in clinical trials evaluating neuroprotective strategies designed to prevent or delay nerve damage.
“Pro-DRN may help move diabetic eye care from detecting established damage toward earlier, molecularly informed risk stratification, so that closer monitoring and future neuroprotective interventions can be directed to the people most likely to benefit,” Wang and colleagues write.
The post AI-Powered Blood Test Detects Early Retinal Damage in Diabetes appeared first on Inside Precision Medicine.
New national action plan targets gaps at the intersection of mental health and criminal justice.
FOR IMMEDIATE RELEASE
OTTAWA, ON – The Mental Health Commission of Canada (the Commission) today released “Finding New Pathways: An action plan for criminal justice and mental health in Canada”. The plan provides an evidence-based roadmap to address systemic gaps in care, coordination, and community supports, recognizing the impact of mental health on the criminal justice system.
Many are working to improve outcomes but needs remain high and progress is uneven. This action plan is a practical pan-Canadian reference point that seeks to contribute to the mental health and wellbeing of all individuals who interact with the criminal justice and forensic mental health systems, including those who work within them.
Developed through a rigorous five-year process, the action plan was shaped by input from national subject matter experts, research, and the lived and living experiences of justice-involved individuals and system workers. The action plan comes at a critical time, as individuals experiencing mental health challenges currently comprise roughly three-quarters of all federally incarcerated people in Canada.
“Meaningful and sustainable transformation is within reach for Canada,” said Lili-Anna Pereša, President and CEO of the Mental Health Commission of Canada. “While these transformative changes cannot be implemented overnight or by one group alone, this action plan serves as a starting point. It centralizes evidence-based approaches designed to break the cycle of recidivism and prioritize prevention, diversion, end-to-end supports and continuity of care.” The action plan stands on three strategic pillars:
- Care, not criminalization: Ensuring all people in Canada have access to supports that help prevent involvement with the criminal justice system, and prioritizing diversion for those with mental illnesses.
- Care during criminal justice involvement: Providing access to high-quality, trauma-informed, and culturally safe health and social supports for those within the system.
- Care after criminal justice involvement: continuity of care and seamless integration into community-based mental health
and substance use services upon release.
“Finding New Pathways” identifies 68 specific recommendations across individual, community, institutional, systemic, and societal levels. It pays particular attention to priority populations, including people from First Nations, Inuit, and Métis, and African, Caribbean, and Black, and other equity-deserving groups who are currently overrepresented in the justice system and face distinct mental health needs.
The action plan further highlights the critical importance of supporting the psychological health and safety of workers within the criminal justice and forensic mental health systems, noting that public safety personnel are significantly more likely to experience symptoms consistent with mental disorders than the general population.
Howard Sapers, current executive director of the Canadian Civil Liberties Association, former Correctional Investigator of Canada, and project advisor for the action plan, emphasized the necessity of these reforms: “ or too long, Canada’s criminal justice system has been asked to shoulder responsibilities it was never designed to carry. The over representation of people with mental health needs in police encounters, courts, and correctional facilities is a predictable consequence of systemic gaps in care, coordination, and community supports. The Mental Health Commission of Canada’s National Action Plan offers something we have been missing for years: a coherent, evidence-based roadmap that prioritizes health, human rights, and dignity.”.
Media Contact:
For English requests, please contact Josie Sabatino at jsabatino@summa.ca; 250-649-6856.
For French requests, please contact Carlene Variyan, cvariyan@summa.ca; 613-601-7456.
The post New national action plan targets gaps at the intersection of mental health and criminal justice. appeared first on Mental Health Commission of Canada.
STAT+: Radiopharmaceutical shows promise in post-Pluvicto setting
Want to stay on top of the science and politics driving biotech today? Sign up to get our biotech newsletter in your inbox.
Good morning. Today, we’re looking at mixed reactions to a closely watched immunology trial and growing scrutiny of a type of telehealth business model.
The need-to-know this morning
- Vera Therapeutics and the FDA reached an agreement to allow Vera to accelerate the analysis of a confirmatory Phase 3 study involving atacicept, its treatment for the chronic kidney disease IgA nephropathy. The drug is already under review for accelerated approval, with a decision expected by July 7. The new FDA agreement will allow Vera to analyze its Phase 3 study, needed for full approval, in the third quarter, rather than wait for an additional year of data to accrue.
Abivax’s positive data weighed down by cancer concerns
Abivax said yesterday that its experimental treatment for ulcerative colitis showed significant efficacy in a closely watched maintenance trial.
Kinase Droplets Activate Growth Signals, Path for Cancer Therapy
A new study published in Cell Reports titled, “Kinase condensates enrich ATP and trigger autophosphorylation,” suggests that cellular phase separation, a mechanism that organizes biomolecules into dense, liquid-like condensates, may play a previously underappreciated role in regulating kinase activity. The findings suggest that aberrant condensate formation could contribute to oncogenic signaling while also offering new opportunities for drug targeting.
“Many biological molecules have this propensity to spontaneously separate,” said Lindsay Case, PhD, assistant professor of biology at Massachusetts Institute of Technology (MIT) and corresponding author of the study. “We were really interested in asking, if we have these kinases forming droplets, what is the consequence of that in the context of signaling?”
Phase separation occurs when proteins condense into highly concentrated liquid-like droplets within cells, analogous to oil droplets separating from vinegar. Although biomolecular condensates have emerged as important organizers of cellular processes, their impact on kinase signaling has remained incompletely understood.
The researchers examined three kinases: focal adhesion kinase (FAK), Mst2, and Abl. Across all three systems, condensate formation increased kinase activity by concentrating enzymes and substrates, thereby promoting phosphorylation reactions.
For FAK, the team found that elevated protein levels were sufficient to drive droplet formation and activate downstream growth signaling. The findings raise the possibility that FAK overexpression in tumors could promote constitutive signaling through condensate formation, potentially contributing to cancer progression and metastasis.
“It was surprising that just by condensing this protein into a droplet, you can actually turn on a signaling pathway that should be turned off,” said Case. “If FAK concentration is too high, you’re always getting these droplets and you’re always signaling, regardless of what the receptors that are supposed to be controlling this are doing.”
Mst2 and Abl also phase separated at high concentrations, which led to increased activity. For Mst2, phase separation is a strategy that healthy cells use to control the Hippo signaling pathway, which promotes cell growth and survival. Phase separation can also lead both enzymes to phosphorylate additional targets, and activate different signaling pathways.
“It’s not just that you’re getting faster phosphorylation, but in those cases, the patterns of what is actually getting phosphorylated were very different inside of the droplet compared to what might be happening in a non-droplet context,” Case says. “The kinase is able to phosphorylate amino acid residues beyond the set of canonical sites that have been described before.”
Mechanistically, the team found that kinase condensates selectively concentrate ATP, the phosphate donor required for kinase activity. Positively charged regions within kinases appear to recruit negatively charged ATP molecules to support phosphorylation.
Using machine-learning analysis, the investigators predicted that approximately 45% of the roughly 500 human kinases possess the molecular features needed to form similar condensates. The findings suggest that phase separation may represent a widespread regulatory mechanism that could influence both normal cellular signaling and disease-associated kinase activity.
In future work, Case hopes to explore designing drugs that could mimic ATP’s ability to be attracted into droplets within a cell, which could reduce side effects.
The post Kinase Droplets Activate Growth Signals, Path for Cancer Therapy appeared first on GEN – Genetic Engineering and Biotechnology News.
Circio and GenAssist Collaborate on Gene Therapy for Muscle Disease and In Vivo Cell Therapy
Oslo, Norway-based Circio and Suzhou, China-based GenAssist entered into a research collaboration to develop circVec-enhanced AAV vectors specifically engineered for in vivo cell therapy and targeted, low dose systemic gene therapy.
Genetic muscle disease is an area of major unmet medical need, where current gene therapy’s high dosing requirements are associated with severe toxicity. By integrating Circio’s and GenAssist’s complementary technologies, the parties aim to develop joint next generation of AAV gene therapy candidates, according to a Circio spokesperson. The focus is on addressing genetic muscle conditions where high and broad muscle-specific expression is required at substantially lower therapeutic AAV doses than can be achieved by conventional AAV gene therapy.
“Our second-generation AAV platform establishes a new benchmark for safety, utilizing highly tissue-specific, de-targeted capsids to dramatically lower systemic dosing while eliminating off-target toxicity,” said Chunyan He, PhD, CEO of GenAssist. “Through our collaboration with Circio, we integrate their unique circular RNA technology. This partnership directly addresses the core demands of next-generation genetic medicine, overcoming the traditional dose-expression trade-off to deliver safer and more effective therapies.”
In addition, Circio and GenAssist will explore the potential of generating joint in vivo CAR T candidates for oncology and autoimmune applications. The collaboration will involve production of novel AAVs combining GenAssist´s T-cell targeting with the circVec expression cassette from Circio. The combined AAVs will subsequently be tested in vitro and in vivo, and if successful, candidates for further development will be nominated for preclinical development.
“The targeted AAVs developed by GenAssist have the ability to specifically and efficiently transduce muscle or T-cells upon systemic delivery with near-complete liver de-targeting,” added Thomas Hansen, PhD, CTO of Circio. “The partnership between Circio and GenAssist will aim to evaluate whether the enhanced circVec expression acts synergistically with these targeted capsids and promoters.
“This fits perfectly into Circio’s strategy of testing circVec in multiple tissues using different AAV variants, both internally and externally. This will allow us to identify new therapeutic avenues where circVec delivers a benefit, and forge partnerships potentially enabling multiple future development opportunities. China is a particularly interesting geography, with cutting edge science and accelerated pathways to establish early clinical data.”
The post Circio and GenAssist Collaborate on Gene Therapy for Muscle Disease and <i>In Vivo</i> Cell Therapy appeared first on GEN – Genetic Engineering and Biotechnology News.
Detection of Self-Harm in Electronic Mental Health Records Using Privacy-Preserving Local Language Models: Methodological Study
Background: Self-harm is the strongest risk factor for suicide and an important outcome for mental health care. Although prevalent in clinical populations, it is often imprecisely captured in routinely collected clinical data, where it is often recorded and stored as unstructured free text. Contemporary language models, such as GPT (OpenAI) and Gemini (Google), can analyze free-text clinical notes, but such models may violate data governance of processing sensitive patient data. Objective: This study aimed to evaluate whether a privacy-preserving language model running entirely within an institution’s secure computing infrastructure (here, the UK National Health Service [NHS]) could accurately identify the presence and timing of self-harm using electronic health records from secondary mental health care. Methods: Clinical notes were drawn from Oxford Health NHS Foundation Trust using a multistage workflow: (1) a random sample of 1000 patients with a psychiatric diagnosis, defined according to the (; codes F00–F99); (2) candidate-note identification using a Gemma3-4b language model to flag notes containing self-harm content; and (3) from those candidates, 1352 randomly sampled notes were selected for expert annotation, resulting in gold-standard corpus enriched for self-harm content. Clinical notes were annotated for the presence of self-harm and its timing (≤90 days, >90 days, or unknown). A privacy-preserving locally served 27-billion-parameter Gemma 3 language model (“Gemma3-27b”) was used as the core model. Prompts were systematically developed and refined using a labeled development set to identify self-harm and generate a structured output per clinical record. Gemma3-27b performance was compared against a strong baseline multilabel text classification model based on robustly optimized BERT pretraining approach (RoBERTa), a transformer-based language model architecture. Model performance was evaluated using precision, recall, and the -score (harmonic mean of precision and recall), with 95% CIs estimated from 1000 bootstrap samples with replacement. Results: Gemma3-27b outperformed the RoBERTa classifier across all categories, achieving Precision=0.92, Recall=0.92 (sensitivity), and -score=0.92 for notes containing self-harm, and Precision=0.97, Recall=0.97 (specificity), and -score=0.97 for notes without self-harm. For the 51 notes labeled as recent self-harm in the held-out test set, Gemma3-27b achieved Precision=0.84, Recall=0.75, and -score=0.79. The global weighted -score of Gemma3-27b across all categories was 0.88, compared to 0.85 for RoBERTa. Conclusions: With systematic prompt development on a labeled development set, but no gradient-based fine-tuning, the current Gemma3-27b language model matched or exceeded a fine-tuned RoBERTa classifier for ascertaining self-harm events and their timing. Aggregate gains were modest, while improvements were largest in the most challenging, lower-frequency timing categories. On a simplified binary recent-versus-other task, RoBERTa performed marginally better, indicating that supervised classifiers remain highly effective when the task is simplified and sufficient labeled data exist. This work demonstrates the technical feasibility of privacy-preserving self-harm detection within a secure NHS research environment.
