Background: University students experience elevated psychological distress, with limited access to mental health services. While cognitive behavioral therapy (CBT) demonstrates efficacy for anxiety and depression, treatment gaps persist due to access barriers and insufficient between-session support. Large language model (LLM) chatbots could improve and scale CBT delivery. However, the scientific evaluation of chatbot-enhanced protocols is just emerging. Objective: This pilot study aimed to assess the feasibility, acceptability, and preliminary efficacy of an LLM-based ChatBot as an adjunct to group Unified Protocol (UP) therapy for between-session support in university students with subclinical anxiety and depression symptoms. Methods: A single-arm feasibility trial recruited university students aged 18 years and older with moderate subclinical symptoms (Social Phobia Inventory: 21‐40, Patient Health Questionnaire-9: 5‐14, or Generalized Anxiety Disorder-7: 5‐14), excluding those with current psychiatric disorders, suicidal ideation, or psychotropic medication use. The intervention comprised 4 weekly group UP counseling sessions complemented by an adjunctive Claude 3.7-Sonnet LLM ChatBot programmed with UP-based therapeutic prompts for between-session support rather than a stand-alone therapeutic agent. Primary feasibility outcomes included treatment adherence, chatbot engagement metrics, and system usability (System Usability Scale). Secondary outcomes assessed changes in generalized anxiety (Generalized Anxiety Disorder-7 Scale), social anxiety (Social Phobia Inventory), depression (Patient Health Questionnaire-9), and well-being (Short Warwick-Edinburgh Mental Wellbeing Scale) using paired tests. Qualitative feedback was collected through focus group interviews and analyzed using thematic analysis. Results: Of 72 screened participants, 37 met eligibility criteria and 19 initiated treatment (mean age 22.06, SD 1.78 years; 70.6% female). Retention was high with 17 completers (10.5% dropout rate). Among completers, 94.1% (16/17) attended ≥3 group sessions. The engagement with the CBT ChatBot was substantial: participants were active on a median of 23 days during the 34-day study period and exchanged a median of 15 messages in total. System usability was rated as excellent (mean 84.94, SD 10.98 out of 100). Pre-to-post comparisons revealed significant improvements in generalized anxiety (mean change −3.00, SD 3.46; =3.01, =.004; Cohen =0.71) and mental well-being (mean change +2.29, SD 3.65; =−2.17, =.02; Cohen =0.69). Social anxiety and depression showed nonsignificant trends toward improvement. Qualitative feedback highlighted the CBT ChatBot’s accessibility and nonjudgmental support while noting limitations in personalization. No adverse events or inappropriate chatbot interactions occurred. Conclusions: Augmenting a group UP therapy with an LLM ChatBot demonstrated high feasibility, acceptability, and preliminary efficacy signals for university students with subclinical symptoms. The hybrid intervention package achieved strong retention and engagement while maintaining safety. These findings support progression to a randomized controlled trial to definitively evaluate this technology-enhanced approach for expanding access to evidence-based mental health interventions.
<img src="https://jmir-production.s3.us-east-2.amazonaws.com/thumbs/ceb992b5356218740d97471bbda48397" />
Using a Virtual Reality CAVE–Based Mindfulness Intervention to Promote Mental Well-Being in Adolescents With Anxiety Symptoms: Pre-Post Mixed Methods Pilot Study
Background: Adolescent anxiety is a growing public health concern associated with significant social and emotional impairment. Mindfulness-based interventions (MBIs) have shown promise in reducing anxiety and improving well-being; however, engagement remains challenging. Virtual reality (VR)–based delivery may enhance immersion and attention, potentially addressing barriers of traditional mindfulness formats. Evidence on VR-based mindfulness interventions for adolescents, particularly in Hong Kong, remains limited. Objective: This study aimed to evaluate the feasibility and acceptability of a VR-MBI delivered via a CAVE, an enclosed VR environment with three projected walls displaying immersive natural scenes and ambient sounds, for adolescents with mild-to-moderate anxiety symptoms in Hong Kong. Secondary aims were to explore preliminary effects on psychological outcomes and physiological stress regulation and to identify facilitators and barriers to engagement. Methods: A mixed methods, single-group pre-post study was conducted with adolescents experiencing mild-to-moderate anxiety symptoms, recruited from secondary schools and youth service organizations in Hong Kong. Participants completed an 8-week group-based VR-MBI. Feasibility and acceptability were assessed using recruitment, attendance, retention, homework practice frequency, dropouts, and adverse events. Psychological outcomes were measured using the Depression Anxiety Stress Scale–21 and the Mindful Attention Awareness Scale. Heart rate variability indices, including the standard deviation of normal-to-normal intervals and root-mean-square of successive differences, were collected at baseline and postintervention using a wearable device. Focus group interviews explored participants’ experiences. Paired-sample tests and Wilcoxon signed rank tests examined pre-post changes, and qualitative data were analyzed using thematic analysis, with findings integrated through triangulation. Results: A total of 42 participants (mean age 14.88, SD 1.90 years; 20/42, 47.6% female; 22/42, 52.4% male) enrolled and completed both assessments. Attendance was high, with 73.8% (31/42) of participants attending at least 80% (8/10) sessions, and participants engaged in regular homework practice. No dropouts or adverse events were reported. No significant pre-post changes were observed in self-reported distress, anxiety, depression, stress, or trait mindfulness (all >.05). However, significant improvements were observed in both heart rate variability indices, standard deviation of normal-to-normal intervals (mean difference 17.6 ms, 95% CI −33.88 to −1.32; =.04; Cohen =0.38) and root-mean-square of successive differences (mean difference 20.20 ms, 95% CI −38.76 to −1.65; =.03; Cohen =0.39), which may suggest preliminary enhancements in physiological stress regulation. Qualitative findings suggested perceived benefits in emotional regulation, stress reduction, focus, and sleep, with the immersive environment and group-based format identified as key facilitators. Conclusions: The CAVE-based VR-MBI was feasible and acceptable for adolescents with mild-to-moderate anxiety symptoms in Hong Kong. Despite no significant changes in self-reported outcomes, physiological improvements and positive qualitative feedback suggest early benefits not captured by self-report measures. These findings support further investigation of using controlled designs and longer follow-up periods.
Prevalence and Predictors of Self-Reported Adverse Experiences in Digital Meditation Training: 2 Randomized Controlled Trials
Background: Digital meditation-based interventions (MBIs) reach vast global audiences with millions of active users, yet concerns persist about the frequency and nature of adverse experiences (ie, AExs) occurring during meditation training. Some researchers have argued that AExs are substantially underdetected and reflect iatrogenic harm caused by meditation (ie, adverse effects [AEfs]). Others contend that these experiences largely reflect common stressors that would be experienced without meditation. These competing perspectives underscore the need for further research, particularly in the context of digital MBIs, the most widely used form of meditation training. Objective: This study examined the prevalence, predictors, and subjective evaluations of AExs during a digital MBI and tested whether reported experiences may be caused by meditation practice via comparisons between meditation-exposed and nonexposed participants. Methods: Data were drawn from 2 trials of the Healthy Minds Program. Exploratory study 1 (n=315) consisted of a sample of distressed US undergraduate students to estimate the prevalence of AExs and identify baseline predictors. Preregistered confirmatory study 2 (n=594) sampled distressed US adults from all 50 states to replicate findings from study 1 and to examine participants’ subjective evaluations of AExs. Study 2 additionally compared AEx rates between participants who did and did not complete guided meditations to assess whether AExs could be caused by meditation exposure. Study 3 (n=87) used qualitative methods to analyze study 1 participants’ responses to an open-ended question regarding their strategies for coping with AExs. Results: In studies 1 and 2, 27.9% (88/315) and 10.1% (40/396) of participants, respectively, reported at least one AEx during the study period, with 6.7% (21/315) and 3% (12/396) reporting functional impairment, largely aligning with previous research. Critically, in study 2, rates of AExs did not significantly differ between participants who did and did not complete guided meditations, suggesting that these experiences were not caused by meditation practice. Higher baseline depression, anxiety, loneliness, experiential avoidance, and perceived barriers to meditation predicted more frequent AExs. In studies 1 and 2, 89.8% (79/88) and 90% (36/40) of participants who reported AExs, respectively, indicated that they were glad to have learned to meditate. Qualitative analyses showed that participants used diverse coping strategies, often using skills learned through the Healthy Minds Program. Conclusions: AExs were relatively common but occurred at comparable rates among participants who did and did not meditate, challenging claims that such experiences were caused by meditation practice in distressed individuals. Although a small subset of participants reported some degree of functional impairment, most evaluated their AExs as tolerable and described their overall MBI experience as positive. Together, these findings highlight the importance of distinguishing AExs that likely reflect epiphenomena of preexisting distress or symptoms from iatrogenic harm attributable to MBIs. Trial Registration: Study 1: ClinicalTrials.gov NCT04741529; https://clinicaltrials.gov/study/NCT04741529; Study 2: ClinicalTrials.gov NCT06282523; https://clinicaltrials.gov/study/NCT06282523
<img src="https://jmir-production.s3.us-east-2.amazonaws.com/thumbs/09106e6bad7f3f15798304e0c00626ec" />
STAT+: Trump administration revisits policy to close Medicare drug price negotiation loophole
WASHINGTON — The Trump administration on Friday proposed to change a policy that is designed to prevent drugmakers from avoiding Medicare price negotiation by adding active ingredients to drugs.
The policy is part of an annual proposed rule that establishes the process that the Centers for Medicare and Medicaid Services uses to choose the next 20 drugs and biologics for price negotiation. Those drugs will be announced by Feb. 1, 2027, and their negotiated prices will take effect in 2029. The administration also considered a similar policy last year but put off a decision to study it further.
Medicare must wait seven to 11 years after a product is approved by the Food and Drug Administration before it can negotiate its price, depending on the type of medicine. Biologics that are typically administered in doctor offices get more time than drugs taken orally.
Passive Smart Home Monitoring for Delirium-Relevant Anomaly Detection in People Living With Dementia: Proof-of-Concept Study
<strong>Background:</strong> Delirium superimposed on dementia is associated with poor outcomes yet remains underdetected in home settings. Current detection relies on face-to-face clinical assessment (eg, the Confusion Assessment Method criteria), which is rarely applied outside hospitals. <strong>Objective:</strong> This proof-of-concept study developed a theory-driven framework for detecting delirium-consistent anomalous patterns in home-dwelling people with dementia, using passive smart home sensor data. <strong>Methods:</strong> The Technology Integrated Health Management dataset, an open access resource comprising a clinically derived cohort of older adults (aged 50 years) with a confirmed diagnosis of dementia or mild cognitive impairment, was used. The analysis included 13 patients who had at least 50% valid data for at least one 10-day analysis window, with data collected between April 1, 2019, and June 30, 2019. Individualized anomaly detection algorithms, including Isolation Forest and Long Short-Term Memory models, were applied to identify delirium-related anomalies within each participant. Predictor features consisted of theory-driven digital markers approximating key Confusion Assessment Method criteria, including agitation, disrupted sleep-wake cycles, and disorientation (indexed by activity entropy), along with clinically relevant indicators, such as physiological instability (early warning scores) and urinary tract infections. <strong>Results:</strong> Using matched thresholds, the Isolation Forest identified 77 anomalies (anomaly rate: 15.65%), and the Long Short-Term Memory model identified 78 anomalies (anomaly rate: 15.85%), with anomalies typically occurring in short temporal clusters; agreement between methods ranged from 0% to 40% across individuals. Feature importance analyses indicated that activity entropy, sleep quality, and early warning scores were the most influential features, with stronger interfeature correlations observed during anomaly periods than during nonanomaly periods. <strong>Conclusions:</strong> This study demonstrates the technical feasibility of detecting delirium-related anomalies through passive smart home monitoring. While lacking ground truth validation, the approach shows promise for early intervention in community settings. Future validation studies with clinically confirmed delirium labels are essential. <strong>Trial Registration:</strong>
Correction: Barriers and Facilitators in the Implementation of the Systematic Medical Appraisal, Referral, and Treatment (SMART) Mental Health Digital Intervention in Rural India: Mixed Methods Process Evaluation Study
The Download: “reprogramming” aging, and the hidden sense of interoception
This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology.
Why “reprogramming” is the buzziest approach to reversing aging right now
Earlier this week, Life Biosciences, a biotech company focused on reversing age-related diseases, announced that it had dosed its first volunteer. A person with glaucoma has had an experimental treatment injected straight into their eyeball.
The idea is to treat the disease by regenerating healthy nerves in the eye—but the company already hopes to go further. If the treatment can reverse glaucoma, similar treatments could reverse other diseases of aging. Maybe, just maybe, they could reverse aging altogether.
The approach relies on “reprogramming” cells to a younger state. It’s one of many strategies being explored by biotech companies looking to slow and reverse aging. But of all of them, it seems to be the one that is truly taking off.
Read the full story on the pursuit of reprogramming for rejuvenation.
—Jessica Hamzelou
This story is from The Checkup, our weekly newsletter giving you the inside track on all things biotech. Sign up to receive it in your inbox every Thursday.
Inside Interoception: The hidden sense of how you feel inside
Scientists have a word for how we sense ourselves from the inside: interoception. Today, thanks to a 2021 Nobel Prize and new tools that can map internal signaling across the body, research into interoception is taking off.
As researchers decode how signals move between body and brain, a clearer picture is starting to take shape—with implications for how we understand and treat conditions from obesity to chronic pain to anxiety.
Find out how it’s leading to a “new continent of awareness.”
—Katherine W. Isaacs
This story is part of MIT Technology Review Explains, our series untangling the complex, messy world of technology to help you understand what’s coming next. You can read more from the series here.
The must-reads
I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology.
1 SpaceX has officially delivered the largest IPO in history
It’s raised a record $75 billion at a $1.77 trillion valuation. (Axios)
+ Making Elon Musk the world’s first trillionaire (on paper). (Reuters $)
+ The IPO will now put his “extreme ownership” to the test. (Wired $)
+ While China attempts to build a Starlink rival. (Rest of World)
+ And other challenges to SpaceX emerge. (MIT Technology Review)
2 Jeff Bezos wants to build an “artificial general engineer”
Through his new industrial AI startup, Prometheus. (NYT $)
+ Which just raised $12 billion, valuing it at $41 billion. (TechCrunch)
+ Meanwhile, OpenAI is building a fully automated researcher. (MIT Technology Review)
3 Chinese regulators are dramatically intensifying tech enforcement
A spell of relative restraint has ended. (SCMP)
+ Regulators have admonished e-commerce giants Alibaba and JD.com. (FT $)
+ And blocked Meta’s acquisition of Chinese AI startup Manus. (BBC)
4 Google says Chinese cybercriminals used Gemini to scam Americans
It’s suing the network over the alleged AI-powered scams.(NYT $)
+ “Supercharged scams” are one of our 10 Things That Matter in AI Right Now. (MIT Technology Review)
5 Ukraine’s defense AI chief predicts a “new paradigm” of warfare
He expects AI systems to unify into a single battlefield network. (Reuters $)
+ AI chatbots could be used for targeting decisions. (MIT Technology Review)
6 Anthropic has rankled users with its safety-first Fable model
Stringent safety rules and refusals to help have sparked a backlash. (NBC)
+ Anthropic has backtracked on some policies. (Wired $)
7 Pokémon Go data trained AI that could assist military drones
It could help them locate themselves in war zones. (Guardian)
+ Pokémon Go data is also training delivery robots. (MIT Technology Review)
8 Orbital data centers are harder than Silicon Valley thinks
Shedding heat in space requires ingenious new designs. (IEEE Spectrum)
+ We need a few things to put data centers in space. (MIT Technology Review)
9 A toy universe shows time could be a quantum illusion
It could emerge from quantum interactions, rather than just existing by default. (New Scientist $)
10 Chatbots keep telling stories about a lighthouse keeper called Ella
And now we may finally know why. (404 Media)
Quote of the day
“People are paying a trillion dollars for Elon.”
—Ross Gerber, the CEO of Gerber Kawasaki, which owns SpaceX stock, tells the New York Times why he believes the company’s IPO is overvalued.
One More Thing
GEORGE WYLESOL
How generative AI could reinvent what it means to play
I was immediately attracted to open-world games, in which you’re free to explore a vast simulated world and choose what challenges to accept. To make them feel alive, these games are inhabited by crowds of “nonplayer characters” (NPCs). But the illusion starts to weaken when you spend enough time with them.
It may not always be like that. Just as it’s upending other industries, generative AI is opening the door to entirely new kinds of in-game interactions that are open-ended, creative, and unexpected. The game may not always have to end.
Discover how generative AI could make games—and other worlds—deeply immersive.
—Niall Firth
We can still have nice things
A place for comfort, fun, and distraction to brighten up your day. (Got any ideas? Drop me a line.)
+ My feet have fallen for the Crocs x Super Mario collection.
+ Denmark’s 2026 Mullet Championship is the hottest hairdo contest of the year.
+ Hungry at half-time? Here are seven mouth-watering international recipes inspired by the World Cup.
+ Feast your eyes on a helicopter bound for Mars and a flowery Milky Way frame in Nature’s top images from last month.
Artificial intelligence for autism spectrum disorder: advances in diagnosis, behavior analysis and educational support
IntroductionArtificial intelligence has become an increasingly relevant field of research in the study of Autism Spectrum Disorder (ASD), offering novel technological approaches for the analysis, detection, and support of individuals on the autism spectrum. The aim of this study was to systematically review recent scientific literature examining the application of artificial intelligence in ASD.MethodsThe review was conducted following the PRISMA 2020 guidelines. Searches were performed in PubMed, Scopus, Dialnet, and Google Scholar, including studies published between 2019 and 2025. After applying predefined inclusion and exclusion criteria, 18 empirical studies were included in the final analysis. Methodological quality and risk of bias were assessed using Joanna Briggs Institute critical appraisal tools adapted to the methodological design of each study.ResultsCurrent research focuses primarily on four areas: early detection and diagnostic support, automated analysis of behavioral and social patterns, AI-based educational technologies, and communication support systems. Although the reviewed studies demonstrate promising advances in machine learning, computer vision, and natural language processing, important methodological limitations remain, particularly regarding external validation, dataset representativeness, and heterogeneity of performance indicators.DiscussionOverall, artificial intelligence shows considerable potential for supporting diagnosis, education, and communication in ASD; however, greater methodological robustness, transparency, and ethical safeguards remain necessary before broader implementation in real clinical and educational settings.
Distribution of bladder afferent activity across the sacral roots in sheep shows marked individual variation: implications for neuroprosthesis design
ObjectiveImplantable sacral anterior root stimulators enable bladder emptying after spinal cord injury but do not prevent reflex incontinence. A closed-loop neuroprosthesis that detects and inhibits reflex bladder contractions could address this, but first, reliable detection of bladder fullness from the sacral roots. Further, the distribution of afferent bladder activity between sacral roots, and the relationship between efferent and afferent activity within each root, remains unclear and must be clarified to guide implant design.MethodsElectrode books were implanted on the S1–S3 extra-dural sacral roots bilaterally in six terminally anesthetized sheep. Afferent electroneurogram (ENG) was recorded concurrently from all implanted roots during filling cystometries and correlated with bladder pressure. Each root was individually electrically stimulated and the bladder pressure response recorded. Post-mortem morphometric analysis determined fiber size distribution in each root.ResultsOverall, S2 ENG activity showed the highest correlation with bladder pressure, and electrical stimulation of S2 and S3 produced the greatest increases in bladder pressure. Fiber size distribution did not correlate with either ENG activity or bladder pressure response. Significant variation was identified between individual sheep, but notably, in four of six sheep, a single sacral root had both the highest ENG correlation to bladder pressure and the greatest bladder response to stimulation.SignificanceThis study demonstrates reliable recording of bladder afferents from sacral roots using clinically applicable electrodes. It provides the first systematic recording of bladder ENG concurrently across three pairs of sacral roots in multiple animals, and the first characterization of signal distribution between roots. Significant individual variation is identified, impacting the design of future implantable sacral neuroprostheses for bladder control.
Head circumference assessment in pediatric MRI: a pilot study of manual measurement methods and automated segmentation-based alternatives
PurposeHead circumference (HC) is an important clinical parameter in neuropediatrics, but it is often missing or outdated in referral information. This can lead to subjective, reader-dependent estimation during MRI interpretation. We first aimed to compare magnetic resonance imaging (MRI)-based methods for HC measurement against the tape measure (ground truth), and second to establish an automated alternative.MethodsIn 23 children (mean age 4.5 years, range 0.5–17 years), HC was prospectively measured with a tape measure (ground truth) on the day of MRI. MRI-based HC measurements were derived from 3D T1-weighted MPRAGE and followed a two-step workflow: measurement plane selection and circumference measurement within that plane. Plane selection was performed using visual-based, rule-based, atlas-based [(infant) FreeSurfer], or neural network (nn)-based methods. Circumference measurement was performed using manual ellipsoid, manual contour, automated ellipsoid, or automated contour methods. The relative technical error of measurement (r-TEM; acceptable < 1.5%) and intraclass correlation coefficient (ICC; two-way mixed ANOVA model) were used to assess accuracy and consistency with the tape measure.ResultsVisual-based with manual ellipsoid/contour and rule-based with manual ellipsoid/contour showed acceptable accuracy (r-TEM 0.73%–1.12%). Visual-based with automated ellipsoid and rule-based with automated ellipsoid also demonstrated acceptable accuracy (r-TEM 0.77% and 0.68%). Atlas-based with automated ellipsoid achieved the lowest r-TEM (0.55%), followed by nn-based with automated ellipsoid (r-TEM 0.75%). In contrast, automated contour approaches showed unacceptable accuracy (r-TEM 3.42%–4.21%). Seven nn-based measurements with automated ellipsoid/contour were spurious. ICCs were high across all methods (0.993–0.997); however, manual contour and automated ellipsoid were associated with overfitting issues.ConclusionThe developed, fully automated algorithm based on (infant) FreeSurfer provides precise and reliable head circumference measurements from pediatric MRI scans with acceptable overall accuracy and excellent consistency with manual measurements using a tape (gold standard). Our algorithm simplifies the head circumference measurement process and provides a reproducible, reader-independent value that enhances the interpretation of neuroradiological findings. Further studies should be conducted to validate with larger sample sizes and to develop deep neural network algorithms for segmentation.