Tag: Sleep and arousal regulation

Posted on

Tailoring AI solutions for health care needs

The AI market is full of big promises of grand transformation. Health care is a prime target for those promises, beset as it is by financial pressures, labor shortages, and the growing burden of caring for an aging population. AI developers are targeting functions that vary widely, from curing cancer and performing surgery to streamlining routine administrative tasks.

The opportunity is genuine, but execution can be difficult. Numerous software vendors have tried to “fix” health care challenges but failed because they misunderstood the environment. “Health care is very complex,” says Steve Bethke, vice president of the solution developer market for Mayo Clinic Platform, which supports the buildout and deployment of digital solutions for health care companies through data-based insights and expert validation. “Solution developers must have a deep focus on clinical and technical capabilities, and then align their solutions to the relevant business impacts. If they miss any dimension, the solution will not be adopted or drive value.”

DOWNLOAD THE REPORT

AI applications for health care are proliferating rapidly. The U.S. Food and Drug Administration has approved more than 1,300 AI-enabled medical devices, mostly for interpreting diagnostic images. More than half of these were approved in the past three years, with the earliest dating as far back as 1995. Non-radiological applications carry out tasks as diverse as tracking sleep apnea, analyzing heart rhythms, and planning orthopedic surgeries.

AI applications that do not count as medical devices— for example, those that handle scheduling and administrative tasks—are more difficult to track but are also rapidly increasing. AI can help coordinate complex tasks and workflows that are often conventionally managed by whiteboards and sticky notes. Such functions may well outstrip clinical uses in their impact on health systems. A recent survey of technology leaders found that 72% said their top priority for AI was reducing caregiver burden and improving caregiver satisfaction, while over half (53%) cited workflow efficiency and productivity.

Any health care-related application can potentially impact patient care, whether directly or indirectly, and AI apps that are poorly designed or inadequately trained and validated can put patients at risk. Providers recognize that risk: In the same survey, 77% said immature AI tools are a significant barrier to adoption. Regulators and lawmakers are also keeping an eye on the risks as development and adoption burgeon, though the U.S. regulatory picture is still in flux, as a 2024 report to Congress on AI in health care observes.

To tackle some of the technical challenges, many health care providers are partnering with application developers to build AI solutions. In a recent study, McKinsey found that 61% of health care organizations intend to pursue partnerships with third-party vendors to develop customized generative AI solutions as a primary strategy as opposed to building them in-house or buying off-the-shelf products.

But health care-specific AI applications must also be tailored to the nuanced clinical needs of medical providers as well as the complex business and regulatory considerations of the wider sector. This is where developers can benefit from working with a partner with a deep understanding of the health care environment to tailor applications to what providers want and need most. Doing so helps to position AI products for maximum impact and value, avoiding the pitfalls unique to the health care environment.

Download the report.

This content was produced by Insights, the custom content arm of MIT Technology Review. It was not written by MIT Technology Review’s editorial staff. It was researched, designed, and written by human writers, editors, analysts, and illustrators. This includes the writing of surveys and collection of data for surveys. AI tools that may have been used were limited to secondary production processes that passed thorough human review.

Posted on

Excessive Internet use and depressive symptom levels in adolescents with depressive disorders: chain mediation of social anxiety and sleep quality

BackgroundAdolescents with depressive disorders are at elevated risk for adverse mental health outcomes, and excessive Internet use has been increasingly linked to greater symptom severity. Therefore, this study aimed to examine the chain mediating roles of social anxiety and sleep quality in the association between excessive Internet use and depressive symptoms among adolescents with depressive disorders.MethodsA cross-sectional design was used. A total of 266 Chinese adolescents with clinically diagnosed depressive disorders (M = 15.79 years, SD = 1.85; 71.4% female) were assessed using the Internet Addiction Test, Zung Self-Rating Depression Scale, Social Anxiety Scale for Children, and Pittsburgh Sleep Quality Index. Correlation analyses and bootstrapping methods were conducted using SPSS and the PROCESS macro to examine the chain mediating effects of social anxiety and sleep quality.ResultsThe total indirect effect of excessive Internet use on depressive symptoms accounted for 65.66% of the total effect. Specifically, the indirect effects via social anxiety and sleep quality accounted for 24.10% and 26.51% of the total effect, respectively. In addition, the chain mediating effect of social anxiety and sleep quality was significant, accounting for 14.76% of the total effect.ConclusionExcessive Internet use was positively associated with more severe depressive symptoms among adolescents with depressive disorders, both directly and indirectly through the chain mediating effects of social anxiety and sleep quality. These findings highlight potential targets for preventing and intervening in excessive Internet use among this population.
Posted on

Distinct sleep-disordered breathing phenotypes in elderly patients with depressive disorder: links to hypoxemia severity and inflammatory burden

ObjectiveTo identify sleep-disordered breathing phenotypes in older adults with depressive disorder and obstructive sleep apnea-hypopnea syndrome (OSAHS) and to evaluate their associations with systemic inflammation.MethodsElderly patients with depressive disorder and OSAHS were consecutively enrolled from January to December 2025. A Gower distance matrix was constructed and phenotypes were derived using partitioning around medoids (PAM; k-medoids), with k selected based on silhouette, elbow criteria, and clinical interpretability. Blood samples were collected the morning after PSG to measure serum high-sensitivity C-reactive protein (hs-CRP), interleukin-6 (IL-6), interleukin-1β (IL-1β), and tumor necrosis factor-α (TNF-α).ResultsAmong 198 participants, k = 2 was selected based on internal validity metrics (silhouette and elbow) and clinical interpretability. Compared with the lower-hypoxia/less-severe OSAHS phenotype (Cluster 1, n = 92), the high-hypoxia/severe OSAHS phenotype (Cluster 2, n = 106) had higher BMI, HAMD-17, and ESS, and more severe AHI/ODI/TS90 with a lower LSaO2. The high-hypoxia/severe OSAHS phenotype also showed higher hs-CRP, IL-6, IL-1β, TNF-α, WBC, neutrophils, and NLR. The inflammatory burden score was higher in the high-hypoxia/severe OSAHS phenotype (β = 1.10 SD unadjusted; β = 1.67 SD adjusted for age, sex, BMI, comorbidity, smoking, drinking, education, and MoCA; β = 1.45 SD further adjusted for HAMD-17 and ESS; all P < 0.001). In men (n = 135), PAM clustering similarly identified two phenotypes differentiated mainly by AHI/ODI, with selective elevations in IL-1β and neutrophil counts.ConclusionsThe high-hypoxia/severe OSAHS phenotype in older adults with depressive disorder is independently associated with a higher systemic inflammatory burden.
Posted on

Data-Driven Tool Identifies Individuals at Highest Risk of Obesity-Related Disease

A new clinical risk model may transform how obesity is managed, by identifying which individuals are most likely to develop serious complications, regardless of their body mass index (BMI).

Developed by researchers at Queen Mary University of London and the Berlin Institute of Health, the tool, called OBSCORE, uses just 20 routinely collected clinical variables to predict the future risk of 18 obesity-related conditions, ranging from type 2 diabetes to cardiovascular disease.

Published in Nature Medicine, the study challenges the long-standing reliance on BMI as the primary metric for assessing obesity-related health risk.

Moving beyond BMI

BMI has long served as a simple proxy for obesity, but it fails to capture the biological heterogeneity of patients. Two individuals with similar BMI can have vastly different risks of developing complications.

The new model addresses this limitation directly. As described in the study, it “provides information beyond BMI” by integrating multiple dimensions of health into a unified risk score.

These include demographic data, clinical biomarkers, disease history, and lifestyle factors, variables already commonly available in healthcare settings.

The findings show that BMI alone is a poor discriminator of risk. The model consistently outperformed BMI-based approaches across all tested outcomes.

Large-scale data enables precision risk prediction

To build the model, researchers analyzed health data from nearly 200,000 individuals with overweight or obesity from the UK Biobank.

Using an interpretable machine learning framework, they screened more than 2,000 potential predictors and distilled them into a core set of 20 features that best predicted long-term health outcomes.

The resulting OBSCORE model estimates the 10-year risk of developing 18 conditions, including cardiovascular disease, kidney disease, sleep apnea, and metabolic disorders.

The model demonstrated strong predictive performance, with median concordance indices around 0.75 across outcomes, indicating robust discrimination between high- and low-risk individuals.

Hidden high-risk individuals

One of the most striking findings is that high-risk individuals are not always those with the highest BMI.

A substantial proportion of individuals classified as high risk fell into the “overweight” category (BMI 27–30 kg/m²), rather than obesity. In some outcomes, up to ~40% of those in the highest risk group had BMI below the obesity threshold.

This reveals a critical gap in current clinical practice: individuals who may benefit from intervention could be overlooked simply because they do not meet BMI-based criteria.

On the other hand, some individuals with obesity may have relatively low risk and may not require intensive intervention.

Strong risk stratification across diseases

Beyond prediction, the scientists believe that OBSCORE enables meaningful risk stratification. Individuals in the highest risk group showed dramatically higher rates of disease compared to those in the lowest group.

For example, the study reports:

  • Up to 89-fold higher risk for chronic kidney disease
  • 42-fold higher risk for type 2 diabetes
  • 47-fold higher risk for cardiovascular mortality

These differences exceed those observed when comparing individuals based solely on BMI categories, underscoring the added value of multidimensional risk assessment.

Clinical and healthcare implications

The implications of these findings are significant, particularly in the context of emerging obesity therapies.

Highly effective drugs such as GLP-1 receptor agonists and dual incretin therapies have transformed treatment options, but their high cost and limited availability make patient prioritization essential.

As the authors note, current systems lack robust frameworks to identify which patients should receive treatment.

OBSCORE offers a potential solution by enabling risk-based allocation of interventions, ensuring that treatment is directed toward those most likely to benefit.

This could improve clinical outcomes while optimizing healthcare resource use.

Toward implementation in clinical practice

One of the key strengths of OBSCORE is its practicality. Unlike many predictive models, it relies on a small number of variables that are already routinely collected, making it suitable for integration into electronic health records.

The researchers envision the model being used as a decision-support tool in clinical settings, complementing rather than replacing existing frameworks.

External validation in independent cohorts—including populations of different ancestry, demonstrated strong generalizability, further supporting its potential for real-world deployment.

Limitations and next steps

Despite its promise, the model requires further validation in broader populations, including younger individuals and more diverse healthcare settings.

Additionally, while OBSCORE effectively stratifies risk, translating these predictions into actionable treatment thresholds will require clinical consensus and cost-effectiveness analyses.

The authors also emphasize that the model identifies predictive, not necessarily causal, factors, and should be interpreted accordingly.

Taken together, the findings mark a shift toward precision medicine in obesity, moving from simplistic metrics like BMI to data-driven, individualized risk assessment.

By capturing the complex interplay of metabolic, clinical, and behavioral factors, OBSCORE could enable earlier intervention, better targeting of therapies, and improved long-term outcomes for patients living with overweight and obesity.

The post Data-Driven Tool Identifies Individuals at Highest Risk of Obesity-Related Disease appeared first on Inside Precision Medicine.

Posted on

Examining circadian rhythm dysregulation using actigraphy among treatment-seeking individuals with alcohol use disorder

BackgroundIdentifying factors predictive of relapse in patients with alcohol use disorder (AUD) following a period of abstinence and/or treatment is essential to discover effective treatment plans for this disease. Previous evidence found that individuals with AUD who relapsed had lower sleep regularity scores than those who did not relapse. This analysis aimed to extend previous work to explore the relationship between circadian rhythms and relapse.MethodsTreatment-seeking individuals with AUD (n = 126) were admitted to an inpatient treatment program for approximately 28 days and, upon discharge, wore Philips Respironics Actiwatches® for the subsequent 4 weeks during an unprotected environment. A subset of these participants wore the devices prior to discharge for up to 7 days inpatient (n = 36). Relapse status was assigned if a participant consumed any alcohol during the outpatient period of data collection. Inpatient and outpatient circadian rhythm nonparametric statistics were calculated for all participants including weekly interdaily stability (IS) and intradaily variability (IV), and daily most active 10 h (M10), least active 5 h of the day (L5), relative amplitude (RA), and the wake time. Linear and logistic generalized mixed models were fitted to estimate the effect of discharge on circadian rhythms, the effect of preceding circadian rhythms on the probability of relapse, and the effect of relapse on circadian rhythms. All analyses accounted for within-patient repeated measurements.ResultsThe final cohort size was n = 103 for the outpatient subset and n = 36 for the inpatient subset after actigraphy data filtering. Participants were 48.6 ± 11.3 years of age, and 32% were female. A total of 26 (25.2%) participants relapsed. There were significant decreases in IV and L5 and increases in M10, RA, and wake time between inpatient and outpatient settings, whereas IS did not substantially differ between the two settings. Following relapse, there was a moderate decrease in hourly IV and an increase in (later) wake time of ~1 h; no other circadian variables were significantly predictive of relapse.ConclusionOverall, circadian rhythms shifted after discharge but were not predictive of relapse. Instead, relapse was followed by a delayed average wake time and a moderate reduction of daily activity patterns. These results highlight the potential value of monitoring circadian changes as indicators of relapse occurrence rather than relapse risk.
Posted on

Tryptophan modulates the impact of prolactin on insomnia in perimenopausal women: a cross-sectional study

BackgroundInsomnia is highly prevalent among perimenopausal women and exerts detrimental effects on physical health, psychological well-being, and overall quality of life. However, its underlying mechanisms remain incompletely understood. This cross-sectional study aimed to identify factors associated with insomnia in perimenopausal women.MethodsA total of 187 perimenopausal women aged 45–55 years were enrolled. Insomnia, anxiety, and depression severity were assessed using the Insomnia Severity Index (ISI), Generalized Anxiety Disorder-7 (GAD-7), and Patient Health Questionnaire-9 (PHQ-9), respectively. Serum levels of relevant amino acids and hormones were measured. Spearman correlation and linear regression analyses were performed to examine the associations among prolactin levels, tryptophan levels, insomnia, anxiety, and depression. Moderation analysis was further conducted to evaluate the potential moderating role of tryptophan in these relationships.ResultsSerum prolactin levels were positively associated with scores of ISI, GAD-7, and PHQ-9. Furthermore, prolactin levels were positively correlated with the severity of sleep-onset difficulties, sleep maintenance problems, noticeability of impairment, and sleep-related distress. Of note, serum tryptophan levels significantly moderated the association between prolactin levels and ISI scores (β = 0.227, 95% CI = 0.04–0.41, p = 0.0148). To wit, he positive relationship between prolactin levels and insomnia severity was stronger in perimenopausal women with higher serum tryptophan levels compared with those with lower levels.ConclusionsThe moderating effect of serum tryptophan on the relationship between prolactin levels and insomnia in perimenopausal women helps us understand the neuroendocrine mechanisms underlying perimenopausal insomnia and may inform future research on targeted preventive and therapeutic strategies.
Posted on

Advanced Neural Probes Reveal Predictable Patterns in Epileptic Brain Activity

In addition to suffering seizures, many people with epilepsy also experience bursts of abnormal brain activity called interictal epileptiform discharges (IEDs). These can happen thousands of times a day and interfere with attention, memory, language, and sleep. New data from a study led by scientists at University of California, San Francisco (UCSF) shows that these brain blips are not random events as once thought. The data shows that they unfold in a predictable pattern that can be detected before they occur, suggesting it may be possible to prevent them. 

Details of their work are published in Nature Neuroscience in a paper titled “Laminar organization of cellular microcircuits modulating human interictal epileptiform discharges.” In it, the scientists explain that they used a high-resolution technology recently adapted for humans that records individual neuron activity to track more than 1000 neurons in four patients undergoing surgery for epilepsy. The so-called Neuropixel probes provide “a view into new ways we might address a debilitating aspect of epilepsy that we haven’t been able to tackle,” said Jon Kleen, MD, PhD, an associate professor of neurology at UCSF and co-senior author of the study. 

Preventing brain blips would be a boon for patients’ quality of life because over time, the effects of these mental disruptions can be significant and may account for some of the cognitive impairment experienced by about half of people with epilepsy. 

Neuropixels probes, which are thin devices lined with hundreds of sensors, are designed to record activity throughout the human cortex. This means that unlike current sensors which are limited to brain signals on the surface of the brain, Neuropixels can provide a three-dimensional view of brain activity. For the study, the scientists implanted the probes seven millimeters deep into the part of the brain where patients’ seizures originate—this is the tissue that surgeons typically remove to reduce epilepsy symptoms. 

Inserting the probes here made it possible to observe what happened in the neurons before, during, and after each IED. While seizures appear as a burst of neurons firing in synchrony, when IEDs occur, they unfold sequentially. Specifically, one set of neurons was active about a second before the IED started followed by another set that generated the sharp electrical spike at its peak, and then a third set became active as the IED faded. “We could see individual neurons that were just microns apart from each other playing different roles in the process,” said Alex Silva, the study’s first author and a medical student and doctoral candidate in the UCSF-UC Berkeley Joint PhD program in bioengineering. “It was really striking.”

Previous studies have demonstrated that most neurons involved in IEDs are used in normal cognitive processing. According to this study, nearly 80% of the neurons involved in IEDs were also involved in language and perception. Current implantable devices for epilepsy may be able to help. They include closed loop neurostimulators that can detect abnormal brain activity and deliver electrical pulses that interrupt it. So in the case of IEDs, devices that monitor single neurons could use the activity of the first set of neurons announcing the arrival of the abnormal pattern as a warning signal. “That would be a major step forward, changing treatment from reactively responding to abnormal brain bursts to proactively preventing them in the first place,” Kleen said.

The post Advanced Neural Probes Reveal Predictable Patterns in Epileptic Brain Activity appeared first on GEN – Genetic Engineering and Biotechnology News.

Posted on

STAT+: New obesity tool aims to predict risk of 18 serious complications

Body mass index has its limitations, but for now it’s the metric medicine often defaults to when predicting weight-related health problems. A new tool promises to better define who’s at risk for obesity complications, based on measures that include BMI but also family history, diet, current illness, and socioeconomic factors culled from medical records.

One aim of the research is to better understand who’s a candidate for an obesity drug, often prescribed based on BMI alone or BMI in combination with another disease. Over time, GLP-1 medications, whose initial target was type 2 diabetes, have revealed their power to ease cardiovascular disease, kidney disease, liver disease, sleep apnea, and osteoarthritis, in addition to promoting significant weight loss. But discerning who’s the best fit for the costly, lifelong treatment has been uncertain. 

“We really wanted to have an integrated model that enables us to look at not one, but 18 different obesity-relevant complications,” Claudia Langenberg, co-author of a study about the new model published Thursday in Nature Medicine, said in a media briefing Tuesday. She is director and professor of medicine and population health at Precision Healthcare University Research Institute of Queen Mary University of London.

Continue to STAT+ to read the full story…

Posted on

Effects of bifrontal-transcranial direct current stimulation combined with music listening on sleep quality, cortical activation and functional connectivity in patients with insomnia: a randomised controlled trial by fNIRS

BackgroundAlthough music listening and transcranial direct current stimulation (tDCS) alone have certain effects in the treatment of insomnia, the sleep regulatory effects and neural mechanisms of the combined treatment in patients with insomnia disorder (ID) are unclear. This study aimed to investigate the efficacy of combined bifrontal-tDCS (F3: anode, F4: cathode) with music listening in patients with ID using functional near-infrared spectroscopy (fNIRS).Methods76 ID patients were randomly divided into an intervention group (n=38) and a control group (n=38), and received 4 weeks of a total of 20 sessions of music + tDCS therapy and music + sham tDCS therapy (30-second stimulation with fade-in/fade-out to mimic somatic sensations), respectively. The Pittsburgh Sleep Quality Index Scale (PSQI), Self-rating Depression Scale (SDS), Self-rating Anxiety Scale (SAS), and Perceived Stress Scale (PSS-14) were compared between the two groups before and after treatment. Oxy-haemoglobin (HbO2) concentration and functional connectivity (FC) were assessed during the verbal fluency task using fNIRS.ResultsCompared with the control group, the PSQI total score (mean difference: -2.57 points, 95% CI: -4.43 to -0.71, p = 0.001), PSQI sub-scores except “sleep disturbance and daytime dysfunction”, SDS and SAS scores of the intervention group improved significantly after treatment. It was observed by fNIRS that the HbO2 concentration in the medial prefrontal cortex (mPFC), left dorsolateral prefrontal cortex (DLPFC), right ventrolateral prefrontal cortex, and right superior frontal cortex (SFC) increased significantly after treatment in the intervention group but was not superior to the control group. In addition, the FC enhancement of left SFC-left DLPFC and left SFC-mPFC after treatment was significantly better in the intervention group than in the control group, and the PSQI improvement was positively correlated with the FC enhancement of channel-averaged and left SFC-right DLPFC.ConclusionsCombining bifrontal-tDCS with music listening is more helpful in improving sleep quality and prefrontal functional connectivity in ID patients compared with music listening alone. For ID patients, music electrical stimulation headphones may be a safe, effective, and convenient new treatment strategy.Clinical trial registrationhttps://www.chictr.org.cn/, identifier ChiCTR2400086233.
Posted on

PAD-S/CSA as a candidate shared representation layer for computational psychotherapy: minimal architecture and a staged validation roadmap

Psychotherapy schools often describe overlapping process phenomena in non-interoperable vocabularies. This pluralism is clinically valuable but computationally costly: datasets become difficult to compare, clinically load-bearing distinctions are collapsed into convenience labels, and artificial intelligence (AI) systems inherit annotation schemes rather than a clinically interpretable intermediate representation. Building on the Perceive–Assess–Dose–Safeguard (PAD-S) framework and the Conflict-Square Algorithm (CSA), this theory article asks a narrower question than the prior PAD-S and CSA papers: can the same variables be formulated as a candidate shared representation layer between heterogeneous observation models and school-specific intervention policies? The proposed layer projects a high-dimensional biopsychosocial state into four clinically observable process coordinates—defensive/avoidant organization (DEF), anxiety/arousal and tolerance (ANX), progression toward direct experience and action (PRO), and self-attack/shame processes (SUP)—plus a safety threshold that constrains admissible intervention intensity. The contribution is architectural rather than empirical: it isolates the representational role from earlier decision-grammar and transcript-coding roles; clarifies the distinction between observations, representation, and policy; specifies a minimal falsifiable family of state-transition models; illustrates translation across four pragmatic therapy families; and defines a staged validation order from reliability and function linkage to transcript-level predictive operationalization and only then sparse equation discovery. The framework should therefore be read as a candidate shared representation layer for computational psychotherapy and computational psychiatry rather than as a therapy protocol, a fitted predictive model, a complete generative theory, or an autonomous decision system. No new dataset, fitted classifier, transcript-level predictive result, or discovered equation is reported here. The article aims instead to state what would count for or against PAD-S/CSA as a clinically interpretable interface for later empirical modeling.