Tag: Academic research

Posted on

An In-Hospital Mortality Risk Model for Patients Undergoing Coronary Artery Bypass Grafting Based on Machine Learning: Cohort Study

<strong>Background:</strong> Ischemic heart disease remains the leading cause of death worldwide. Coronary artery bypass grafting (CABG) remains the primary surgical treatment for ischemic heart disease. There is currently a lack of highly accurate and widely applicable models for assessing the risk of postoperative mortality following CABG. <strong>Objective:</strong> This study aimed to develop and validate an in-hospital mortality risk prediction system for patients undergoing coronary artery bypass grafting (CABG) by using machine learning algorithms and to compare its performance with the European System for Cardiac Operative Risk Evaluation II (EuroSCORE II) and Sino System for Coronary Operative Risk Evaluation (SinoSCORE). <strong>Methods:</strong> Between January 2017 and December 2020, 21,443 patients undergoing CABG in the Chinese Cardiac Surgery Registry were included. Patients were randomly divided into training (n=17,753) and test (n=3690) cohorts. We addressed class imbalance using the synthetic minority oversampling technique (SMOTE) and optimized hyperparameters via grid search. Fifteen machine learning algorithms were developed to predict in-hospital mortality. Performance was evaluated using the area under the receiver operating characteristic curve (AUC), calibration metrics (Brier score), and decision curve analysis, and was compared against EuroSCORE II and SinoSCORE. <strong>Results:</strong> A total of 21,443 patients were included. Overall, in-hospital mortality was 2.1% (n=450). The Extreme Gradient Boosting (XGBoost) model achieved the best performance with an AUC of 0.850 in the training cohort and 0.782 in the independent test cohort (this cohort was independent and not involved in model construction). While EuroSCORE II showed an AUC of 0.722 and SinoSCORE showed an AUC of 0.726 in the test cohort, the XGBoost model demonstrated superior discrimination and calibration (<i>P</i>&lt;.05). <strong>Conclusions:</strong> Our study developed and validated a machine learning–based risk prediction model for in-hospital mortality after CABG by using a large-scale Chinese multicenter registry. Among the algorithms tested, the XGBoost model demonstrated superior discrimination and calibration compared with the traditional EuroSCORE II and SinoSCORE, suggesting that locally calibrated models may better capture the risk profile of Chinese patients. The derived 7-variable web calculator may serve as an exploratory auxiliary tool to provide a preliminary reference for bedside risk stratification, though its direct impact on surgical decision-making requires further prospective validation. Future research should focus on independent test cohorts across diverse hospital tiers to ensure broad generalizability.
Posted on

Cell Therapy Shows Promise for Treating Advanced Liver Disease

A type of cell therapy made up of macrophage cells can lower risk for death or need for a liver transplant in people with cirrhosis due to advanced liver disease.

The research, published in Cell Stem Cell, was a long term follow-up study of patients recruited into an earlier Phase I/II trial and showed a 27.5% drop in risk for death or liver transplant in those who received the therapy versus usual care over up to four years of follow up.

“Cirrhosis represents a major global health burden, with substantial and rising morbidity and mortality,” write lead author Stuart Forbes, PhD, a professor at the University of Edinburgh and scientific founder of Resolution Therapeutics—the biotech developing the therapy, and colleagues.

“Apart from liver transplantation, therapeutic options for end-stage disease remain limited to supportive care and management of complications, underscoring the urgent need for innovative approaches to halt or reverse disease progression.”

The treatment used in the study involves extracting monocyte cells from the blood of patients and converting them to macrophages in the lab before injecting the macrophages into the liver to trigger repair.

The therapy is designed to overcome the tissue damage and scarring seen in people with advanced cirrhosis where the liver loses its normal ability to regenerate itself. Macrophages are able to break down scar tissue, reduce inflammation and encourage the growth of new and healthy cells in the liver.

In the initial MATCH Phase I/II study, 26 patients had macrophage therapy and 24 received standard care. After the follow up period, 30.8% of those in the treatment group had died or needed a transplant versus 58.3% of the standard care group. No serious adverse events were linked to the cell therapy.

“Liver disease is a major cause of death… Although we can use liver transplantation as a rescue treatment for a proportion of people who have advanced liver disease, this is restricted by a lack of suitable donor organs… There is therefore a desperate need for alternative treatments for patients with advanced liver disease,” commented Forbes, in a press statement.

“We hope this type of approach could one day add to our treatment choices for patients with advanced liver disease, reducing the need for liver transplants.”

Resolution has another similar Phase I/II trial to Match underway called Emerald that is building on the initial results but using macrophages modified to be more effective at tackling liver damage.

The company is also running an observational trial called Opal, a study in patients with cirrhosis and hepatic decompensation, to characterize disease trajectories and help design endpoints and inclusion criteria for future interventional trials.

The post Cell Therapy Shows Promise for Treating Advanced Liver Disease appeared first on Inside Precision Medicine.

Posted on

STAT+: Eli Lilly says Verve’s gene editor lowers cholesterol levels in early study

Eli Lilly said Monday that a high dose of its gene-editing therapy reduced cholesterol levels by 62% in participants in a clinical trial, an early but encouraging test of whether a one-time treatment may one day help people seeking to lower their LDL, or “bad,” cholesterol.

Lilly acquired the therapy, VERVE-102, in its $1 billion buyout of Verve Therapeutics last year. Executives tout it as a potential treatment to broadly prevent heart disease, the world’s leading killer, as many patients struggle to stay on existing, more conventional medicines for reducing cholesterol levels.

There were no treatment-related serious adverse events in the Phase 1 study — a notable finding, given that Verve had to shelve its first candidate due to safety concerns. 

Continue to STAT+ to read the full story…

Posted on

Machine learning for Alzheimer’s disease progression under extreme class imbalance

BackgroundTimely identification of individuals at risk for Alzheimer’s disease (AD) progression remains a major clinical challenge. Traditional cognitive assessments provide limited prognostic insight, while many machine learning (ML) models rely on costly biomarkers or poorly interpretable algorithms that limit clinical scalability. This study evaluated whether widely available baseline demographic, clinical, and cognitive measures could support short-term progression prediction using interpretable ML methods under extreme class imbalance.MethodsWe analyzed 3,240 participants from the Alzheimer’s Disease Neuroimaging Initiative (ADNI), of whom 2,423 had valid 24-month follow-up data. The primary outcome was strict unidirectional diagnostic worsening within 24 months (13 events; 0.5%). Baseline demographic, clinical, and cognitive variables were used to train XGBoost and logistic regression models under natural class imbalance using stratified k-fold cross-validation with out-of-fold predictions. Model performance was evaluated using AUROC, area under the precision-recall curve (AUPRC), calibration analyses, and bootstrap confidence intervals. Sensitivity analyses evaluated cost-sensitive learning, threshold optimization, and alternative imputation strategies (KNN and MICE). Longitudinal mixed-effects modeling was conducted separately to characterize cognitive decline and was not used as input to the predictive models. SHAP (Shapley Additive Explanations) quantified feature contributions.ResultsUnder natural class imbalance, XGBoost achieved AUROC = 0.912 and AUPRC = 0.051, while logistic regression achieved AUROC = 0.787 and AUPRC = 0.038. Although discrimination exceeded baseline prevalence, precision remained low and threshold optimization produced substantial false-positive burdens, limiting immediate clinical applicability. Cost-sensitive learning did not materially improve performance. MICE imputation produced results comparable to median imputation, whereas KNN imputation reduced performance. SHAP analyses identified baseline cognitive severity, functional measures, and diagnostic status as dominant predictors. Mixed-effects modeling confirmed significant cognitive decline over time (β = −0.027 points/month, p < 0.001).ConclusionAccessible baseline clinical and cognitive variables contain measurable but limited predictive signal for short-term AD progression under extreme event scarcity. These findings should be interpreted as an early-stage proof-of-concept rather than a clinically deployable decision-support tool. External validation remains necessary before clinical translation.
Posted on

The impact of face-to-face social exclusion on university students’ interpersonal cooperation behavior: a hyperscanning study based on fNIRS

BackgroundMost studies on social exclusion adopt virtual paradigms focusing on unilateral responses, while neglecting real-world face-to-face interaction and its neural basis. Functional near-infrared spectroscopy (fNIRS) hyperscanning allows recording of interpersonal neural synchronization (INS) during dyadic interaction, providing a novel approach for investigating interpersonal cooperation.MethodsThis study recruited 24 dyads of college students randomly assigned to social exclusion and inclusion groups. Using fNIRS hyperscanning combined with a face-to-face rejection paradigm and the Prisoner’s Dilemma task, we examined subjective experience, behavior, and INS.Results(1) The exclusion group reported lower intimacy, trust, belonging need and state self-esteem than the inclusion group. (2) Only defection decision reaction time was faster in the exclusion group, with no group differences in overall cooperation and defection rates, indicating exclusion primarily accelerates defection. (3) INS showed channel-specific differences: the exclusion group had weaker right orbitofrontal INS during cooperation and stronger left dorsolateral prefrontal INS during defection. (4) Cooperation reaction time negatively correlated with trust, while defection efficiency positively correlated with left frontopolar INS. (5) State self-esteem partially mediated the link between social exclusion and defection reaction time.ConclusionFrom an integrated psychological–behavioral–neural perspective, this study confirms that face-to-face social exclusion accelerates defection decisions by impairing subjective interpersonal experience, altering prefrontal INS, and through the mediating effect of subjective feelings. These findings provide empirical evidence for understanding the mechanisms underlying campus social exclusion.
Posted on

Long-term psychological and functional outcomes after hepatitis C eradication with direct-acting antivirals: an 80-month follow-up study

IntroductionDirect-acting antivirals (DAAs) have dramatically changed hepatitis C virus (HCV) treatment, by achieving high virological cure rates and a reduced percentage of adverse events compared with previous treatments. Despite this, long-term psychiatric and quality-of-life (QoL) outcomes after viral eradication remain insufficiently understood. This study provides an 80-month follow-up of a previously evaluated cohort.MethodsOf the original 62 patients, 24 (38.7%) were reassessed approximately 80 months after DAA initiation. Psychopathological symptoms (HAM-D, HAM-A, SCL-90-R), coping strategies (COPE), and QoL (SF-36) were evaluated and compared with baseline (T0). Patients were stratified by psychiatric history (Group P vs. Group NP). Non-parametric tests and exploratory correlations and regressions were performed.ResultsIn Group P, depressive and anxiety symptoms significantly improved from T0 to follow-up (HAM-D: 16 vs. 3, p < 0.01; HAM-A: 15 vs. 4, p < 0.01), with additional reductions in SCL-90-R Interpersonal Sensitivity, Paranoid Ideation, and Psychoticism (all p < 0.05). Group NP showed stable psychological profiles, reflecting the maintenance of improvements already observed at T1, with significant reductions in HAM-D (7 vs. 4, p < 0.01) and HAM-A (8 vs. 4, p < 0.001). Both groups demonstrated a significant decline in SF-36 Physical Component Summary (Group P: p < 0.05; Group NP: p < 0.01). No between-group differences were detected at follow-up. Avoidant coping and psychiatric history were significant negative predictors of long-term anxiety change.ConclusionsSeven years after HCV eradication, psychological well-being remains stable or improved, while physical QoL declines. DAAs demonstrate sustained long-term psychiatric safety, underscoring the need for integrated medical–psychological follow-up in HCV survivors.
Posted on

Role of psychological resilience and psychological distress in linking fear of disease progression to quality of life in chronic heart failure: a cross-sectional serial mediation analysis

ObjectiveTo examine whether psychological resilience and psychological distress serially mediate the association between fear of disease progression and quality of life (QoL) in patients with chronic heart failure (CHF).MethodsThis cross-sectional study enrolled 212 patients with CHF admitted between June 2023 and June 2025. Assessment tools included a demographic questionnaire, the Fear of Progression Questionnaire (FoP-Q), the Connor–Davidson Resilience Scale (CD-RISC), the Depression Anxiety Stress Scales-21 (DASS-21), and the Minnesota Living with Heart Failure Questionnaire (MLHFQ). Correlation and serial mediation analyses were performed using IBM SPSS Statistics for Windows, version 22.0, and the PROCESS macro, with the bootstrap method (5,000 resamples) used to test the mediation effects.ResultsThe mean scores were 43.60 ± 8.32 for FoP-Q, 52.71 ± 14.28 for CD-RISC, 44.29 ± 10.68 for DASS-21, and 48.63 ± 10.85 for MLHFQ. Correlation analysis indicated that FoP was negatively correlated with psychological resilience (r = −0.775) and positively correlated with psychological distress and MLHFQ scores (r = 0.868 and 0.773, respectively; all P < 0.05). Psychological resilience was negatively correlated with both psychological distress and MLHFQ scores (r = −0.728 and −0.744, respectively), while psychological distress was positively correlated with MLHFQ scores (r = 0.745; all P < 0.05). The mediation model revealed a direct effect of FoP on QoL (effect = 0.629, 41.14%), along with three indirect pathways: via psychological resilience alone (effect = 0.508, 33.22%), via psychological distress alone (effect = 0.344, 22.50%), and via the serial pathway from psychological resilience to psychological distress (effect = 0.048, 3.14%).ConclusionPatients with CHF exhibited elevated levels of FoP and generally reduced QoL. Psychological resilience and psychological distress served as significant serial mediators in the relationship between FoP and QoL. FoP could directly reduce QoL in patients with CHF and indirectly affect it by decreasing psychological resilience and exacerbating psychological distress. Clinical attention should be directed toward assessing the psychological status of patients with CHF, improving psychological resilience, alleviating negative emotions, reducing the adverse impact of FoP, and enhancing patients’ QoL.
Posted on

Factors influencing the sensory profile in patients with autism spectrum disorder from 16 months to 14 years: results of an observational study

IntroductionSensory processing abnormalities represent a high-impact clinical feature in individuals with Autism Spectrum Disorder (ASD). Sensory alterations can be classified by modality and behavioral responses to stimuli. The literature presents conflicting results regarding the association between the sensory profile and other clinical elements. Therefore, this study aimed to characterize the clinical variables that influence sensory anomalies in subjects with ASD.MethodsThis single-center retrospective observational study selected 159 subjects with a confirmed ASD diagnosis, aged 16 months to 14 years. Data collected included clinical assessments (developmental profile, intelligence quotient, adaptive profile and autistic symptom severity) and sensory evaluations using the Sensory Profile-2. A Bayesian cumulative logit model was implemented to assess the impact of clinical variables on sensory profiles.ResultsThe analysis identifies adaptive functioning as the most consistent predictor of sensory outcomes, where higher functioning correlates with a reduction in atypical patterns, particularly sensory sensitivity. Conversely, autism severity and sex generally showed no significant impact, though higher cognitive functioning was specifically linked to increased sensory avoiding behaviors. Furthermore, the Bayesian analysis revealed a modifying effect of age on the sensory profile, characterized by a non-monotonic trend clustered around two distinct age peaks. The preschool period is characterized by a predominance of sensory seeking behaviors and auditory processing challenges, whereas the 8-to-12-year developmental window marks a significant shift toward hypersensitivity and avoidant patterns, particularly within the visual and oral modalities.DiscussionOverall, the analysis revealed that the adaptive profile has a significant impact on the sensory profile, consistent with existing literature. Age emerged as a key modifying factor with a non-monotonic trajectory, contrasting with previous studies that typically highlight either improvement or stability over time. Study main limitations include its retrospective and non-longitudinal nature. Therefore, further prospective longitudinal studies are necessary to explore these dynamic associations between the sensory profile and other clinical variables over time.
Posted on

Prediction of 30-Day All-Cause Hospital Readmissions Using Limited Structured Electronic Health Record Data: Retrospective Comparative Study

Background: Unplanned hospital readmissions represent a critical operational and financial challenge for health care systems in the United States, with 3.8 million 30-day all-cause readmissions in 2018 at an average cost of US $15,200 each, totaling US $58 billion in costs. Many published prediction models rely on comprehensive information (eg, full billing abstractions, discharge summaries, laboratory tests, and vitals) that becomes available only late in the encounter, limiting usefulness for real-time, in-hospital intervention. This creates a timeliness-accuracy trade-off: models that are most accurate retrospectively may arrive too late to act upon. Objective: This study tests whether a clinically meaningful predictive signal for 30-day all-cause readmission is present within a limited set of structured clinical codes recorded during the patient’s hospital stay. This approach evaluates whether predictive signals are retained when using a restricted set of structured clinical codes. Methods: We conducted a retrospective comparative modeling study using a large, deidentified electronic health record dataset of 50,000 inpatient encounters from the 2019 New York State Emergency Department Database. Two feature sets were constructed: (1) a limited set consisting of the first 5 () diagnosis codes, the first 5 Current Procedural Terminology (CPT) codes, and Charlson Comorbidity Index (CCI; 11 input features); and (2) a rich set using all available and CPT codes plus CCI (up to 135 input features). We trained 4 models: random forest, CatBoost, multilayer perceptron, and DistilBERT (a distilled Bidirectional Encoder Representations from Transformers [BERT] model; structured codes mapped to text and tokenized with DistilBERT-base-uncased). Evaluation used an untouched hold-out set of 10,000 encounters, preserving the natural 21.1% readmission prevalence. Primary metrics were area under the receiver operating characteristic curve (AUROC), -score, and accuracy. To address class imbalance, the training split only was balanced via undersampling of the majority class and bootstrap oversampling of the minority class; validation/test distributions were left unchanged. Results: Models trained on the limited feature set achieved AUROC values ranging from 0.5369 to 0.5596 and -scores from 0.2555 to 0.3434. Across 3 of 4 architectures, models trained on the limited feature set matched or exceeded the discrimination of their rich counterparts. The best model (random forest, limited) achieved an area under the curve AUROC 0.5596 (95% CI 0.5440‐0.5739) compared to the best performing rich model (DistilBERT) at 0.5703 (95% CI 0.5565‐0.5842), an absolute difference of 0.0107. The highest -score (0.3434) was achieved by DistilBERT on the limited feature set. Differences across architectures were small in absolute terms, with threshold-dependent metrics (eg, -score) being comparable. Conclusions: The findings suggest that models using a limited set of structured clinical codes can achieve performance comparable to those using more comprehensive coding information.
<img src="https://jmir-production.s3.us-east-2.amazonaws.com/thumbs/2c6a3535f15393463cd4b4c04e53b2e6" />