SEARCH · Search PubMed

Results for “Therapy, Computer-Assisted”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

768 records · Page 10Linked to original sources

PubMed2026

Imaging-based surgical stratification of parasagittal meningiomas involving the superior sagittal sinus: a case analysis of 62 patients.

OBJECTIVE: The objective was to evaluate the Superior Sagittal Sinus Involvement Grading (SSIG) system as an imaging-based surgical stratification framework for parasagittal meningiomas adjacent to the superior sagittal sinus (SSS) and to assess its relationship with established sinus invasion grading, venous sinus patency, and operative strategy. METHODS: In this single-center retrospective cohort study, the authors included 62 consecutive parasagittal meningioma resections performed by a single surgeon. SSIG grade was assigned primarily on contrast-enhanced coronal MRI, with CT/MR venography used when available to evaluate sinus patency and collateral venous drainage. Operative variables, resection strategy, and clinicopathological factors were compared across SSIG and Sindou grades, and postoperative complications were compared between low- and high-involvement SSIG groups. RESULTS: SSIG correlated significantly with Sindou grade (rs = 0.790, τb = 0.702, both p < 0.001), and among patients with available venous imaging, it also correlated with the venous sinus involvement grade (rs = 0.742, τb = 0.665, both p < 0.001). With increasing SSIG grade, operative time, intraoperative blood loss, and intraoperative fluid administration increased (p = 0.012, p = 0.008, and p = 0.007, respectively). Compared with the low-involvement group (SSIG grades 1, 2, and 4a), the high-involvement group (SSIG grades 3, 4b, and 5) was less likely to achieve Simpson grade I resection and more likely to adopt Simpson grades II-III strategies (66.7% vs 13.6%, p < 0.001; OR 12.667). Surgery-related complication rates did not differ significantly between groups. The mean follow-up was 13.3 ± 7.9 months, with no radiographic recurrence or progression at last follow-up. CONCLUSIONS: SSIG characterizes parasagittal meningiomas by integrating sinus invasion, venous patency, falcine extension, and parasagittal convexity involvement on preoperative imaging. This surgically oriented framework may help anticipate operative exposure, sinus handling, and resection strategy. Its predictive value for complications and long-term oncological outcomes requires validation in larger cohorts with longer follow-up.

Humans↗

PubMed2026

Machine learning vs. traditional methods for predicting postoperative cardiac complications after non-cardiac surgery: a systematic review and Bayesian network meta-analysis.

INTRODUCTION: Accurate prediction of peri-operative cardiac complications is critical to optimise pre-operative decision-making. Traditional risk prediction scores, such as the Revised Cardiac Risk Index, show only modest discrimination. Machine learning can model complex, non-linear relationships but their predictive performance compared with traditional scores remains unclear. METHODS: We performed a systematic review and Bayesian network meta-analysis. The primary outcome was postoperative adverse cardiac events following non-cardiac surgery. Prediction models were assessed relative to the Revised Cardiac Risk Index. As many studies evaluated multiple versions of each model type, the highest performing ('best version') and lowest performing ('worst version') results were analysed. Models were ranked using the surface under the cumulative ranking curve (SUCRA). RESULTS: Thirteen studies evaluating 54 models and 927,113 patients were included. Machine learning approaches generally outperformed traditional risk scores. Automated machine learning ranked highest (SUCRA 96.6) showed the greatest improvement in the best version analysis (mean difference (MD) 0.28 (95%CrI 0.16-0.40)) and remained superior in the sensitivity analysis (MD 0.30 (95%CrI 0.14-0.45)). Gradient boosting models showed superior performance over the Revised Cardiac Risk Index across analysis (best version: MD 0.20 (95%CrI 0.14-0.26), worst version: MD 0.18 (95%CrI 0.12-0.25), SUCRA 82.4). The Gupta Perioperative Risk for Myocardial Infarction or Cardiac Arrest score outperformed the Revised Cardiac Risk Index in the best version analysis (MD 0.16 (95%CrI 0.01-0.32)). Between-study heterogeneity was low. None of the included studies externally validated their machine learning models and only six were judged to be at low risk of bias. DISCUSSION: Most machine learning models showed better discrimination than traditional risk scores, with automated machine learning and gradient boosting models ranking highest. However, study quality, calibration reporting and absence of external validation limit immediate clinical adoption. Prospective, multicentre evaluation is required before integration of these models into peri-operative practice.

Humans↗

PubMed2026

Application of SPI-guided analgesia in laparoscopic gynecologic surgery: a randomized controlled trial evaluating the remifentanil-sparing effect and predictive value of time-weighted SPI.

This study aimed to achieve two primary objectives: (1) to evaluate the opioid-sparing effect of Surgical Pleth Index (SPI)-directed analgesia during surgery via a randomized controlled trial (RCT), and (2) to propose and preliminarily assess a novel dynamic metric, Threshold-based Time-Weighted SPI (Tb-TW-SPI), which integrates stimulus intensity and duration, for its predictive efficacy regarding postoperative moderate-to-severe pain. Employing an RCT combined with exploratory analysis, 61 patients undergoing elective laparoscopic gynecologic surgery were randomized into an SPI-directed analgesia group or a conventional analgesia group. The primary outcome was total intraoperative remifentanil consumption. Postoperatively, an exploratory analysis of the control group data evaluated the correlation between Tb-TW-SPI and Numeric Rating Scale (NRS) pain scores in the post-anesthesia care unit (PACU), calculating its predictive value for moderate-to-severe pain (NRS ≥ 4). Results: The SPI-directed group required significantly less intraoperative remifentanil than the conventional group [median (IQR): 5.84(5.02,6.62)vs. 6.96(5.81,8.19)µg/kg/h; P = 0.016]. Postoperative pain scores did not differ significantly between groups (P > 0.05). Exploratory analysis of the conventional analgesia group revealed that Tb-TW-SPI values were significantly higher in patients with moderate-to-severe postoperative pain (NRS ≥ 4) compared to those without (P = 0.0417).The area under the ROC curve for Tb-TW-SPI predicting this pain was 0.74 (95% CI: 0.52-0.96), with 67% sensitivity and 76% specificity at an optimal cutoff of 1210. This RCT suggests that SPI-directed analgesia can safely and moderately reduce intraoperative remifentanil consumption. Furthermore, the proposed Tb-TW-SPI metric, in this exploratory analysis, suggests potential for predicting postoperative pain, though this finding requires validation in larger cohorts with higher-frequency SPI sampling, offering a new direction for SPI interpretation. Large-scale, multicenter trials are warranted to validate the predictive utility of Tb-TW-SPI. Clinical Trial Registration, China Clinical Trial Registry: ChiCTR2400088444.

Humans↗

PubMed2026

Closed-loop vasopressor systems for hemodynamic control in perioperative and critical care settings: a systematic review and meta-analysis.

Maintaining mean arterial pressure (MAP) within a predefined target is central to haemodynamic management in surgical and critically ill adults receiving vasopressors. Closed-loop vasopressor (CLV) systems automate titration to optimise blood pressure control, but their clinical effectiveness remains uncertain. We performed a systematic review and meta-analysis comparing CLV with manual titration. This PRISMA 2020-compliant review was prospectively registered in PROSPERO (CRD420250655697). MEDLINE, Embase, Scopus, Web of Science, CENTRAL, and the Cochrane Library were searched (January 2000-June 2025). Randomised controlled trials enrolling adults receiving vasopressors in perioperative or intensive care settings were included. Primary outcomes were time within the MAP target range and time spent in hypotension or hypertension. Risk of bias was assessed using RoB 2.0 and certainty of evidence using GRADE. Random- or fixed-effects models were selected according to heterogeneity. Six randomized controlled trials (215 patients) were included in the systematic review, whereas five perioperative trials contributed to the meta-analysis of haemodynamic control outcomes, and one ICU-based study was summarized narratively because it did not report comparable MAP control endpoints. CLV increased time within the MAP target range (mean difference [MD] 33.94%, 95% CI 20.41-47.46; I2 = 77%) and reduced time in hypotension (MD - 18.24%, 95% CI - 28.95 to - 7.53; I2 = 73%). There was no significant difference in time in hypertension, cumulative norepinephrine dose, or major/minor adverse events. ICU length of stay was not pooled because of clinical and methodological heterogeneity. Certainty of evidence ranged from low to high (moderate for haemodynamic control outcomes). CLV systems improved haemodynamic control, primarily in perioperative settings, but heterogeneity and small samples limit confidence in effect size and generalisability. Evidence in critically ill populations remains limited, and larger trials are needed to determine whether improvements in these physiological surrogate endpoints translate into meaningful patient-centred outcomes.

Humans↗

PubMed2026

Effectiveness of Mobile-Delivered Exercise and Yoga Programs on Depressive Symptom Reduction in Employees: Randomized Controlled Trial.

BACKGROUND: Mental health challenges such as stress and depression are prevalent among employees. Mobile health platforms that deliver exercise or yoga interventions offer a promising approach to improve mental health outcomes in this population. OBJECTIVE: This study aimed to assess the effectiveness of 12-session adaptive moderate-intensity exercise and yoga programs delivered via a motion-detecting digital platform in reducing stress and depressive symptoms among employees. METHODS: This was an unblinded, 3-arm, parallel-group, randomized controlled trial conducted at Seoul National University Bundang Hospital and Boramae Medical Center between November 2023 and January 2024. Eligible participants were full-time employees. Seventy-five participants were randomly assigned to an exercise, a yoga, or a cognitive behavioral therapy-based self-care control group using computer-generated randomization. The exercise and yoga groups engaged in motion-detecting, adaptive physical activity training, whereas the control group accessed mobile-based, self-directed stress management educational materials. The intervention was largely automated, with no individualized therapeutic guidance provided. Allocation was concealed until trial entry. All recruitment and outcome assessments were conducted in person at the hospitals. The primary outcomes were perceived stress and depressive symptoms, whereas the secondary outcomes included posttraumatic stress, insomnia severity, cognitive stress response, occupational stress, and burnout. Physiological outcomes were assessed using heart rate variability and electroencephalography. Measurements were collected at baseline, immediately after the intervention, and at 4-week follow-up. Data were analyzed using a multivariate linear model to evaluate the main effects of time, group, and time×group interactions. RESULTS: Of the 75 randomized participants (exercise: n=24, 32%; yoga: n=25, 33.3%; and control: n=26, 34.7%), 71 (94.7%) who completed at least 9 of the 12 sessions (≥40 min each) were included in the outcome analysis (exercise: n=21, 29.5%; yoga: n=24, 33.8%; and control: n=26, 36.6%). For the coprimary outcomes, the group×time interaction for depressive symptoms (Patient Health Questionnaire-9) approached but did not reach the Bonferroni-corrected threshold (F4,136=2.71; P=.03; adjusted α=.025); however, planned pairwise comparisons revealed significantly greater improvement in the yoga group compared to the control group at 4-week follow-up (β=-3.67; adjusted P<.001). For the Perceived Stress Scale, the interaction was not significant (P=.29), although a significant main effect of time (P<.001) indicated overall stress reduction across all groups. For secondary outcomes, a significant group×time interaction was found for the Cognitive Stress Responses Scale (P=.003), indicating differential trajectories of improvement. The yoga group showed a consistent linear decrease, whereas the exercise group showed immediate but less sustained gains. CONCLUSIONS: Digitally delivered adaptive yoga programs demonstrated superior and sustained improvements in depressive symptoms and Cognitive Stress Responses Scale scores compared with the active cognitive behavioral therapy-based self-care control group. However, the exercise program showed more modest and less sustained effects, warranting further investigation using larger samples.

Adult↗

PubMed2026

Physician-Modified Fenestrated Stent-Grafts Planned Using Three-Dimensional Techniques for Complex Aortic Pathology: A Systematic Review and Meta-Analysis.

BACKGROUND: Complex aortic pathology involving the visceral arteries remains a significant therapeutic challenge. Open repair is associated with considerable perioperative risk, particularly in patients with multiple comorbidities, while standard endovascular aneurysm repair (EVAR) is often not feasible because of inadequate proximal sealing zones. Fenestrated and branched endovascular repair (F/BEVAR) represents an established treatment strategy; however, the use of custom-made devices is limited by manufacturing time and availability. Physician-modified stent grafts (PMSGs) have therefore emerged as a pragmatic alternative. Three-dimensional planning techniques have been increasingly used to facilitate accurate graft modification. The aim of this systematic review and meta-analysis was to evaluate the effectiveness and safety of PMSG procedures planned with three-dimensional techniques. Technical success, target vessel patency, early mortality, endoleak occurrence, and reintervention rates were analyzed. METHODS: A systematic search was conducted in the PubMed/MEDLINE and Embase databases. Studies describing the use of physician-modified fenestrated stent grafts planned with three-dimensional tools were included. Meta-analyses were performed using a random-effects model with restricted maximum likelihood estimation. A logit transformation was used for the analysis of proportions. RESULTS: The analysis included five studies involving 172 patients. The estimated weighted mean follow-up duration was 14.9 months. The overall technical success rate was 92.9% (95% confidence interval [CI]: 84.5-96.9%), with low-to-moderate heterogeneity. Target vessel patency was 96.9% (95% CI: 93.6-98.5%). Early mortality was 5.5% (95% CI: 2.1-13.3%). The incidence of endoleaks was 13.3% (95% CI: 5.8-27.4%), with significant heterogeneity among studies. Reinterventions were reported in 6.6% of patients (95% CI: 2.3-17.5%). CONCLUSION: The results indicate that PMSG procedures planned with three-dimensional techniques are associated with a high rate of technical success and preserved patency of target vessels in patients with complex aortic pathology. The observed variability in endoleak and reintervention rates likely reflects differences in anatomical complexity and patient selection among studies. Further prospective studies are needed to confirm long-term outcomes.

Humans↗

PubMed2026

Effects of an actuated ankle exoskeleton on walking stability in healthy adults: a controlled laboratory study.

BACKGROUND: Ankle exoskeletons are widely used to reduce the metabolic cost of walking, yet their effects on walking stability during unperturbed gait remain insufficiently understood. Walking stability can be characterized using complementary measures that capture stride-to-stride variability, global temporal organization, and local dynamic stability. Understanding how walking with an actuated ankle exoskeleton system influences these different aspects of gait stability is essential for the safe design and control of wearable robotic devices. METHODS: Eighteen healthy adults walked on a treadmill at a constant speed (1.1 m/s) with and without an actuated bilateral ankle exoskeleton in a randomized crossover design. Spatiotemporal variability was quantified using coefficients of variation (CoV) of stride length, step width, and stance ratio. Global gait stability was assessed using detrended fluctuation analysis of stride time. Local dynamic stability was evaluated using maximum Lyapunov exponent calculated for the trunk, hip, upper leg, lower leg, and foot. Paired-samples two-sided t-tests were used to compare conditions. RESULTS: Walking with the ankle exoskeleton resulted in increased stride-to-stride spatiotemporal variability, reflected by higher CoV values for stride length (p < 0.001) and stance ratio (p = 0.005), while mean stride length and step width remained unchanged. Mean stance ratio was reduced in the exoskeleton condition (p < 0.001). Global gait stability did not differ between conditions, indicating preserved long-range temporal gait organization. Local dynamic stability increased at the lower leg (p < 0.001) and foot (p = 0.019) when walking with the exoskeleton. CONCLUSIONS: Walking with the actuated ankle exoskeleton alters gait control across multiple levels during steady walking. While stride-to-stride variability in stride length and stance ratio increased, global gait stability remained unchanged. Local dynamic stability was increased at the lower leg and foot, suggesting segment-specific effects of ankle-level assistance close to the assisted joint. However, these findings should be interpreted as the combined effect of wearing the exoskeleton and receiving active assistance, rather than the isolated effect of plantarflexion assistance. These results provide insight for the design and control of ankle exoskeletons with respect to stability-related effects during walking.

Humans↗

PubMed2026

Respiratory effects of recruitment maneuvers according to lung recruitability assessed by electrical impedance tomography in patients with acute respiratory distress syndrome.

Recruitment maneuvers (RM) can improve oxygenation in patients with acute respiratory distress syndrome (ARDS), but their physiological effects depend on lung recruitability. This secondary analysis of a randomized controlled trial (RCT) evaluated oxygenation, respiratory mechanics, regional ventilation, and cardiorespiratory adverse events responses to a RM followed by electrical impedance tomography (EIT)-guided PEEP titration, using EIT to assess lung recruitability. In this study, fifty patients with moderate-to-severe ARDS underwent a stepwise RM followed by individualized PEEP titration guided by EIT. Lung recruitability was determined using the collapse index at PEEP 6 cmH₂O (CLPEEP6), defined as the proportion of collapsed lung at this PEEP level. Patients were classified into high- and low-recruitability groups based on median CLPEEP6 values. Oxygenation (PaO₂/FiO₂), static compliance (Cstat), driving pressure (Pdriv), regional ventilation distribution, and cardiorespiratory adverse events were compared before and after RM, during subsequent individualized EIT-guided PEEP titration. In patients with high recruitability (CLPEEP6 > 12.5), the PaO₂/FiO₂ ratio and Cstat increased significantly after RM (PaO₂/FiO₂: 100.8 ± 30.3 vs. 125.4 ± 38.3 mmHg, p < 0.05; Cstat: 21.5 ± 5.8 vs. 28.0 ± 7.0 mL/cmH₂O, p < 0.001), Pdriv decreased (19.1 ± 3.5 vs. 15.6 ± 3.2 cmH₂O, p < 0.001). EIT demonstrated a posterior redistribution of ventilation after RM. In contrast, patients with low recruitability (CLPEEP6 ≤ 12.5) showed no significant mechanical or oxygenation improvement and transient hypotension, arrhythmia, and desaturation appeared numerically more common in this group. No barotrauma or cardiac arrest occurred, and ICU mortality was similar between groups. A strategy combining a RM with subsequent individualized EIT-guided PEEP titration was associated with improved oxygenation and lung mechanics in patients with high lung recruitability, whereas patients with low recruitability showed limited physiological benefit, with cardiorespiratory adverse events appearing numerically more frequent. The EIT-derived CLPEEP6 index represents a feasible and clinically applicable https://clinicaltrials.gov/study/NCT06733168.

Humans↗

PubMed2026

Once-weekly IcoSema versus once-daily insulin glargine U100 in type 2 diabetes management (COMBINE 4): an open-label, multicentre, treat-to-target, randomised, phase 3b trial.

BACKGROUND: Stepwise treatment intensification is recommended for managing type 2 diabetes, including insulin initiation when non-insulin glucose-lowering medications are insufficient. Guidelines recommend combining a GLP-1 receptor agonist with basal insulin to improve glycaemic efficacy while reducing weight gain and hypoglycaemia risk. COMBINE 4 evaluated the efficacy and safety of IcoSema, a once-weekly combination therapy of basal insulin icodec and semaglutide (a GLP-1-receptor agonist) versus insulin glargine U100 (glargine U100) in people with type 2 diabetes on oral glucose-lowering medications. METHODS: COMBINE 4 was a 40-week, randomised, open-label, treat-to-target, phase 3b trial conducted across 97 sites in nine countries. Adults (aged ≥18 years) with type 2 diabetes (HbA1c ≥8·0%) receiving oral glucose-lowering medications were randomly allocated in a 1:1 ratio without stratification to IcoSema or once-daily glargine U100. The titration target was 3·9-5·0 mmol/L (70-90 mg/dL). The primary endpoint was change in HbA1c and the secondary confirmatory endpoint was change in bodyweight, both from baseline to week 40, evaluated in all randomly allocated participants. Adverse events were recorded during weeks 0-45. This trial is registered with ClinicalTrials.gov (NCT06269107) and is complete. FINDINGS: Of 653 individuals screened between Feb 15 and Aug 6, 2024, 151 did not meet screening criteria and 17 withdrew before initiating treatment; 243 were randomised to IcoSema and 242 to glargine U100. Of the 485 randomly allocated participants, 286 (59%) were male and 199 (41%) were female, and median age was 58 years (range 26-82). For HbA1c, from baseline (9·57% for IcoSema and 9·50% for glargine U100), mean change to week 40 was greater with IcoSema versus glargine U100 (-3·32 vs -2·44 percentage points; estimated treatment difference [ETD] -0·88 percentage points [95% CI -1·12 to -0·63]), confirming superiority of IcoSema (p<0·001). From baseline to week 40, mean bodyweight decreased with IcoSema and increased with glargine U100 (-0·79 vs 3·81 kg; ETD -4·61 kg [95% CI -5·46 to -3·75]), confirming superiority of IcoSema (p<0·001). Rate of combined clinically significant (blood glucose <3·0 mmol/L [<54 mg/dL], confirmed with a blood glucose meter) or severe hypoglycaemia (severe cognitive impairment requiring external assistance for recovery) was statistically significantly lower with IcoSema versus glargine U100 (0·29 vs 0·59 episodes per person-year of exposure; estimated rate ratio 0·56 [95% CI 0·32 to 0·97]; p=0·04). Gastrointestinal disorders were the most frequently reported adverse events with IcoSema. INTERPRETATION: Once-weekly IcoSema demonstrated superior HbA1c reduction and bodyweight change, with lower rates of clinically significant or severe hypoglycaemia, versus glargine U100, suggesting that IcoSema might be an effective once-weekly treatment option for insulin-naive individuals with type 2 diabetes inadequately controlled on oral glucose-lowering medications. FUNDING: Novo Nordisk.

Humans↗

PubMed2026

Webcam-Based Real-Time Visual Feedback During Baduanjin Practice in Older Adults: 6-Week Pilot Randomized Study.

BACKGROUND: Baduanjin qigong is a traditional mind-body exercise used to support balance and physical health in older adults. Age-related changes in proprioception may make accurate self-directed performance difficult without external guidance. OBJECTIVE: The aim of this study is to explore whether webcam-based real-time visual feedback delivered during supervised laboratory sessions was associated with differences in webcam-derived 2D pose discrepancy and movement consistency during Baduanjin practice in older adults. METHODS: A total of 31 older adults were enrolled, and 28 participants with complete analyzable records were included in this complete-case dataset (feedback group, n=14; nonfeedback group, n=14). All sessions were conducted face-to-face in a supervised motion-analysis laboratory. Weekly 2D pose-discrepancy values were analyzed using a linear mixed-effects model with fixed effects for group, categorical week, and the group-by-week interaction and a participant-specific random intercept. Joint- and movement-specific participant-level 6-week means were analyzed exploratorily using Welch independent-samples t tests. Holm correction was applied across 24 exploratory contrasts (6 week-specific, 8 joint-specific, and 10 movement-specific comparisons), and Hedges g and 95% CIs were reported. Participant-specific weekly slopes and within-participant variability were additionally examined to directly assess longitudinal error drift. RESULTS: The linear mixed-effects model showed no significant group-by-week interaction (Wald χ25=1.09; P=.96) and no significant overall week effect (Wald χ25=6.40; P=.27). Averaged across 6 weeks, the feedback group had an estimated mean 2D pose discrepancy 1.20° lower than the nonfeedback group (95% CI -2.38° to -0.02°; P=.046), although this marginal pilot finding was sensitive to an analytic approach. No week-specific contrast remained significant after Holm adjustment. Nominal right elbow, right shoulder, and right knee differences did not survive global Holm correction. Form 3 showed a lower mean discrepancy in the feedback group (mean difference -3.70°, 95% CI -5.86° to -1.54°; Hedges g=-1.30; unadjusted P=.002; Holm-adjusted P=.04). Direct analyses of participant-specific slopes and within-participant SDs did not support a significant between-group difference in longitudinal error drift. CONCLUSIONS: In this small exploratory pilot study conducted under supervised laboratory conditions, the 6-week trajectories did not differ significantly between groups. A marginally lower average 2D pose discrepancy was observed in the feedback group across the 6 weeks, but no individual week- or joint-specific comparison remained significant after multiplicity adjustment. Form 3 was the only exploratory contrast that remained significant after global Holm correction. Direct longitudinal analyses did not demonstrate prevention of error drift. Larger studies using validated reference measurements, prespecified outcomes, and adequately powered longitudinal designs are required.

Humans↗

PubMed2026

Evaluation of a cornea-specialized large language model for diagnostic and management accuracy in complex corneal cases.

PURPOSE: To evaluate whether a cornea-specialized large language model (LLM) enhanced with retrieval-augmented generation (RAG) improves clinicians' diagnostic and management accuracy in complex corneal cases compared to a general-purpose GPT-4o model and unaided clinician performance. METHODS: This prospective, randomized, masked evaluation study involved three cornea trainees who each independently reviewed 39 real-world corneal cases under three experimental conditions: unaided, GPT-4o-assisted, and assisted by a cornea-specialized GPT-4o model. The cornea-specialized model was constructed by embedding over 200 publicly available Wikipedia articles into GPT-4o's RAG framework. Participants provided open-ended diagnoses and selected the next-step management options (multiple choice). They were allowed up to three GPT-4o queries per case, and the AI-assisted arms were randomized to minimize bias. Accuracy for both tasks was compared against expert reference standards using McNemar's test. RESULTS: Diagnostic accuracy was 48.7%, 20.5%, and 38.5% unaided, improving to 69.2%, 46.2%, and 59.0% with general GPT-4o (p<0.04). The cornea-specialized GPT-4o further improved accuracy to 71.8%, 48.7%, and 74.4%, with improvements over unaided performance for all clinicians (p<0.01). For next-step decisions, unaided accuracy was 76.9%, 87.2%, and 59.0%. With the specialized model, Ophthalmologist 3 improved to 71.8% (p<0.05), Ophthalmologist 1 remained high at 82.1%, and Ophthalmologist 2 declined to 64.1% (p<0.05). CONCLUSIONS: A cornea-specialized LLM enhanced with RAG improved diagnostic accuracy in complex corneal cases, particularly among clinicians with lower baseline performance. Effects on management accuracy were inconsistent. Future studies should explore the use of open-ended management tasks and examine whether smaller, curated retrieval corpora yield better model performance.

Humans↗

PubMed2026

Multiscale Modeling Primer: Focus on Chromatin and Epigenetics.

A central challenge in modern biology is to understand how molecular interactions produce cellular and organismal functions across vast spatiotemporal scales. Nowhere is this challenge more apparent than in the study of chromatin, where meters of DNA compact into a micron-sized nucleus. How this polymer folds is a dynamic process, regulated by epigenetic modifications-chemical changes to DNA and histones that involve only a handful of atoms. These small changes cooperate to produce emergent, higher-order structures that define cellular identity and function. To explain this system, we must integrate static, high-resolution snapshots from techniques like cryo-EM with dynamic, lower-resolution data from microscopy and genomics. Multiscale computational models are essential tools that bridge these experimental gaps and reveal the mechanisms of emergent behavior. However, the communication divide between experimental biologists and quantitative modelers often hampers progress. This primer addresses that gap. It first introduces the fundamental biology of chromatin and epigenetics at an introductory level for non-biologists audiences. We then survey the landscape of computational approaches, from atomistic to systems-level models, and connect them to the experimental data that inform and validate them at an introductory level for non-computationalists. We argue that the next frontier will require us to build integrative models that can predict how molecular perturbations mechanistically alter cellular phenotypes, which will open a new era of chromatin-targeted therapeutics.

Chromatin Dynamics↗

PubMed2026

A systematic review of sleep hygiene interventions in physically and operationally demanding occupations.

Workers in physically and operationally demanding (POD) occupations (e.g. military personnel, emergency responders, law enforcement) experience disrupted sleep due to irregular schedules and high-stress work environments. Given the importance of sleep for physical and cognitive performance, interventions to improve sleep in POD occupations are crucial yet remain under-researched. To investigate the effectiveness of sleep hygiene interventions in improving sleep quality and quantity outcomes for POD occupations, a systematic review was conducted across five databases (PubMed, CINAHL, SPORTDiscus, PsycINFO, Web of Science) in September 2024. Eligible studies involved workers in POD occupations where a sleep hygiene intervention was compared against a waitlist or passive control. Ten studies were included, eight randomised controlled trials (RCTs) and two non-randomised controlled trials (NRCTs). Narrative synthesis identified Structured Training Programs (n = 7), Phone Assistance Programs (n = 2), and Single Lecture Interventions (n = 1). Evidence certainty ranged from Very Low to Moderate, converging towards Low quality due to imprecision, risk of bias, and indirectness. Structured Training Programs showed moderate improvements in sleep quality and small-to-moderate effects on sleep duration. Phone Assistance Programs had minimal effects, and Single Lecture Interventions had negligible impact. Higher-quality research is urgently needed to assess sleep hygiene interventions in POD occupations.

Humans↗

PubMed2026

Treatment of Established Groin Lymphatic Complications After Arterial Surgery: A Systematic Review.

BACKGROUND: To systematically review the effectiveness and safety of interventions for established groin lymphatic complications after arterial vascular surgery. METHODS: PubMed, Embase, and Cochrane CENTRAL were searched from inception to May 28, 2025. This review was registered in PROSPERO (CRD420251061708) and conducted according to PRISMA, PRISMA-S, and Synthesis Without Meta-analysis (SWiM) guidance. Eligible studies included adults with established groin lymphorrhea/lymphocutaneous fistula or seroma/lymphocele after arterial exposure in the common femoral or iliofemoral region. Prophylactic studies were excluded. Two reviewers independently screened studies and assessed full texts. Prespecified outcomes were resolution, time to resolution, length of stay, surgical site infection, wound complications, reintervention, and recurrence. Owing to substantial heterogeneity, meta-analysis was not performed. RESULTS: Eighteen retrospective single-center studies were included; no randomized trials were identified. Reported strategy-family totals were operative control (n = 240, including 55 dye-assisted cases), intranodal lymphangiography with glue embolization (n = 39), intracavitary or chemical instillation (n = 52), negative-pressure wound therapy (n = 20), radiotherapy (n = 206), and drainage or aspiration alone (n = 7). One additional cohort compared nonsurgical with surgical management. Definitions, follow-up windows, and statistical reporting were inconsistent, and arm-level denominators were frequently incomplete. Risk of bias was serious to critical across the evidence base, mainly because of retrospective design, confounding by indication, treatment-selection bias, inconsistent outcome definitions, and incomplete reporting. Using GRADE principles, certainty of evidence was very low across strategies and phenotypes. Across study arms, conduit-directed approaches for external leak and cavity-directed approaches for encapsulated collections often achieved control, but valid comparative inference was not possible. CONCLUSION: Evidence is limited to small retrospective series with heterogeneous definitions, serious to critical risk of bias, and very low certainty. Reliable comparisons between interventions cannot currently be made. Standardized phenotype-specific definitions and outcome reporting are needed to support prospective studies and future trials.

Humans↗

PubMed2025

Artificial Intelligence in Diagnosing Depression Through Behavioural Cues: A Diagnostic Accuracy Systematic Review and Meta-Analysis.

AIM: To synthesise existing evidence concerning the application of AI methods in detecting depression through behavioural cues among adults in healthcare and community settings. DESIGN: This is a diagnostic accuracy systematic review. METHODS: This review included studies examining different AI methods in detecting depression among adults. Two independent reviewers screened, appraised and extracted data. Data were analysed by meta-analysis, narrative synthesis and subgroup analysis. DATA SOURCES: Published studies and grey literature were sought in 11 electronic databases. Hand search was conducted on reference lists and two journals. RESULTS: In total, 30 studies were included in this review. Twenty of which demonstrated that AI models had the potential to detect depression. Speech and facial expression showed better sensitivity, reflecting the ability to detect people with depression. Text and movement had better specificity, indicating the ability to rule out non-depressed individuals. Heterogeneity was initially high. Less heterogeneity was observed within each modality subgroup. CONCLUSIONS: This is the first systematic review examining AI models in detecting depression using all four behavioural cues: speech, texts, movement and facial expressions. IMPLICATIONS: A collaborative effort among healthcare professionals can be initiated to develop an AI-assisted depression detection system in general healthcare or community settings. IMPACT: It is challenging for general healthcare professionals to detect depressive symptoms among people in non-psychiatric settings. Our findings suggested the need for objective screening tools, such as an AI-assisted system, for screening depression. Therefore, people could receive accurate diagnosis and proper treatments for depression. REPORTING METHOD: This review followed the PRISMA checklist. PATIENTS OR PUBLIC CONTRIBUTION: No patients or public contribution.

Humans↗

PubMed2026

Integrated photoelectrocatalytic reduction and oxidation processes to achieve efficient degradation of fluoxetine in pharmaceutical wastewater.

Fluorinated organic compounds have been frequently detected in aquatic environments, with the widespread use of fluorinated drugs. The existing processes of urban sewage treatment plants are difficult to completely remove these pollutants containing the persistent C-F bonds. In this work, an integrated system of UV-activated sulfite and UV-assisted electrochemical oxidation was innovatively constructed for efficient degradation of fluoxetine. For the UV-activated sulfite unit system, when the sulfite dosage was 0.5 mmol/L and the initial pH was about 10, the defluorination efficiency of 5 mg/L fluoxetine wastewater under nitrogen atmosphere was about 98 %. Subsequently, the UV-assisted electrochemical oxidation unit system was employed to treat the reduced wastewater mentioned above. When the sodium chloride dosage was 25 mmol/L, the initial pH was about 5, and the current density was 30 mA/cm2, the total organic carbon (TOC) removal of the wastewater arrived at 65 %. Active species capture experiments and ESR tests confirmed that hydrated electrons, hydroxyl, and chlorine radicals were the main components for the efficient degradation of fluoxetine. According to the analysis of Fukui function and HPLC-MS, the degradation pathway of pollutants was proposed including defluorination and mineralization. Meanwhile, the toxicity of intermediates was predicted using the ECOSAR program. In addition, the verification test of actual wastewater treatment indicated that the defluorination and TOC removal efficiency of fluorouracil by the integrated system were similar to those for fluoxetine. This work provided a new approach for the efficient degradation of fluorinated organic pollutants in pharmaceutical wastewater.

Fluoxetine↗

PubMed2026

The Role of Artificial Intelligence Combined With Digital Cholangioscopy for Indeterminant and Malignant Biliary Strictures: A Systematic Review and Meta-analysis.

BACKGROUND: Current endoscopic retrograde cholangiopancreatography (ERCP) and cholangioscopic-based diagnostic sampling for indeterminant biliary strictures remain suboptimal. Artificial intelligence (AI)-based algorithms by means of computer vision in machine learning have been applied to cholangioscopy in an effort to improve diagnostic yield. The aim of this study was to perform a systematic review and meta-analysis to evaluate the diagnostic performance of AI-based diagnostic performance of AI-associated cholangioscopic diagnosis of indeterminant or malignant biliary strictures. METHODS: Individualized searches were developed in accordance with PRISMA and MOOSE guidelines, and meta-analysis according to Cochrane Diagnostic Test Accuracy working group methodology. A bivariate model was used to compute pooled sensitivity and specificity, likelihood ratio, diagnostic odds ratio, and summary receiver operating characteristics curve (SROC). RESULTS: Five studies (n=675 lesions; a total of 2,685,674 cholangioscopic images) were included. All but one study analyzed a deep learning AI-based system using a convoluted neural network (CNN) with an average image processing speed of 30 to 60 frames per second. The pooled sensitivity and specificity were 95% (95% CI: 85-98) and 88% (95% CI: 76-94), with a diagnostic accuracy (SROC) of 97% (95% CI: 95-98). Sensitivity analysis of CNN studies (4 studies, 538 patients) demonstrated a pooled sensitivity, specificity, and accuracy (SROC) of 95% (95% CI: 82-99), 88% (95% CI: 72-95), and 97% (95% CI: 95-98), respectively. CONCLUSIONS: Artificial intelligence-based machine learning of cholangioscopy images appears to be a promising modality for the diagnosis of indeterminant and malignant biliary strictures.

Humans↗

PubMed2026

Long-term safety of oral orforglipron in Japanese participants with type 2 diabetes (ACHIEVE-J): a multicentre, randomised, open-label, parallel-group phase 3 trial.

BACKGROUND: Orforglipron, an oral GLP-1 receptor agonist, requires further evaluation in east Asian populations with type 2 diabetes, given this group's distinct pathophysiological characteristics. This study aimed to assess orforglipron as add-on treatment to diet and exercise alone or to oral antihyperglycaemic medications in Japanese participants with type 2 diabetes. METHODS: This multicentre, randomised, open-label phase 3 study was conducted in 40 medical research centres and hospitals in Japan. Adults with type 2 diabetes and elevated glucose levels managing their condition with diet and exercise alone or with one or two oral antihyperglycaemic medications were assigned (1:1:1) via computer-generated random sequence to receive once-daily oral orforglipron (3 mg, 12 mg, or 36 mg). Randomisation was stratified by background therapy, baseline HbA1c (≤8·5% or >8·5%), and metformin use (yes vs no; applied only to α-glucosidase inhibitors, thiazolidinedione, and glinides). Investigators, participants, and site staff were not masked to treatment. The primary endpoint was safety for 52 weeks, assessed in all randomly assigned participants who received at least one dose of orforglipron. This study is registered with ClinicalTrials.gov, NCT06010004 (ACHIEVE-J). FINDINGS: Between Sept 28, 2023, and June 5, 2025, 450 participants were screened and 401 were randomly assigned to three groups (3 mg, n=132; 12 mg, n=135; and 36 mg, n=134). 352 (88%) completed study treatment. 339 participants (85%, 95% CI 80·7-87·8) had at least one treatment-emergent adverse event (TEAE), more frequently in the 36-mg group (118 [88%, 95% CI 81·5-92·5]) than in the 3-mg (107 [81%, 73·5-86·8]) and 12-mg (114 [84%, 77·4-89·6]) groups. Most TEAEs were of mild (267 [67%, 61·8-71·0]) or moderate (63 [16%, 12·5-19·6]) severity. Discontinuations due to an adverse event occurred in 19 of 134 participants (14%, 95% CI 9·3-21·1) in the 36-mg group compared with seven of 132 (5%, 2·6-10·5) in the 3-mg group and 11 of 135 (8%, 4·6-14·0) in the 12-mg group. Across treatment groups, gastrointestinal symptoms were the most common TEAEs leading to study treatment discontinuation (3 mg: 5 [3·8%, 1·6-8·6]; 12 mg: 8 [5·9%, 3·0-11·3]; and 36 mg: 11 [8·2%, 4·7-14·1]). Level 2 (blood glucose <54 mg/dL) hypoglycaemia events occurred in three of 135 participants in the 12-mg group (2%, 0·8-6·3) and in three of 134 in the 36-mg group (2%, 0·8-6·4) groups. No level 3 (severe) hypoglycaemia events occurred. Outcomes were generally similar across background therapies. INTERPRETATION: Treatment with orforglipron in combination with diet and exercise alone or one or two oral antihyperglycaemic medications for 52 weeks demonstrated an acceptable safety profile in Japanese adults with type 2 diabetes. FUNDING: Eli Lilly. TRANSLATION: For the Japanese translation of the abstract see Supplementary Materials section.

Aged↗