Die stille Gefahr

Deleted Journal, Год журнала: 2025, Номер unknown

Опубликована: Янв. 31, 2025

---

Towards interpretable drug interaction prediction via dual-stage attention and Bayesian calibration with active learning

PeerJ Computer Science, Год журнала: 2025, Номер 11, С. e2847 - e2847

Опубликована: Апрель 22, 2025

Background Drug-drug interactions (DDIs) account for 17–23% of adverse drug reactions leading to hospitalization, with over 74,000 DDI-related events reported in the FDA Adverse Event Reporting System (FAERS) during 2023. While recent computational methods focus on improving prediction accuracy, they suffer from high false-positive rates (>45%) and often function as black-box models without biological interpretability. Methods We propose Dual-stage attention Bayesian calibration active learning Drug-Drug Interaction (DABI-DDI), a novel framework integrating: (1) A dual-stage mechanism LSTM networks capturing temporal dependencies interactions, (2) approach beta-binomial modeling refining interaction signals reducing false positives, (3) an strategy efficient sample selection, (4) network pharmacology component linking underlying mechanisms. The model was validated using data FAERS, DrugBank, STRING databases, comprehensive evaluation both performance Results DABI-DDI achieved superior (AUC = 0.947, PR_AUC 0.944). improved event detection accuracy (94% vs . 54% AUC), while revealed key molecular mechanisms through enzyme-transporter interactions. Ablation studies demonstrated each component’s significance, maintaining training requirements. Conclusion present DABI-DDI, integrated feature extraction that successfully addresses challenges DDIs three major innovations: Temporal pattern recognition, Most importantly, demonstrates strong clinical applicability by efficiently identifying high-risk combinations providing mechanistic insights pathway analysis. This bridges gap between understanding, offering promising tool safer combination therapy.

Язык: Английский

---

Drug–Drug Interactions in Nosocomial Infections: An Updated Review for Clinicians

Pharmaceutics, Год журнала: 2024, Номер 16(9), С. 1137 - 1137

Опубликована: Авг. 28, 2024

Prevention, assessment, and identification of drug–drug interactions (DDIs) represent a challenge for healthcare professionals, especially in nosocomial settings. This narrative review aims to provide thorough assessment the most clinically significant DDIs antibiotics used healthcare-associated infections. Complex poly-pharmaceutical regimens, targeting multiple pathogens or one pathogen presence another comorbidity, have an increased predisposition result life-threatening DDIs. Recognising, assessing, limiting infections offers promising opportunities improving health outcomes. The objective this is clinicians with practical advice prevent mitigate DDIs, aim increasing safety effectiveness therapy. DDI management importance individualising therapy according patient, disease status, associated comorbidities.

Язык: Английский

---

Assessment of potential drug-drug interactions in hospitalized patients with infectious diseases: an experience from a secondary care hospital

F1000Research, Год журнала: 2024, Номер 13, С. 164 - 164

Опубликована: Авг. 7, 2024

Background Polypharmacy is common among hospitalized patients with infectious infections owing to comorbidities or concomitant illnesses. This raises the likelihood of drug-drug interactions and creates uncertainty for healthcare providers. study aimed assess potential (pDDIs) diseases in a secondary care hospital. Methods A prospective observational was conducted internal medicine ward six months after ethics committee’s approval. Data were collected from patient case records, prescriptions screened pDDIs portable electronic physician information database (PEPID) resource analyzed using SPSS, version 27.0. Results In total, 148 records analyzed, 549 identified, 66.8% having at least one more DDIs. The mean number drug 3.70 ± 4.58 per prescription. most frequently encountered combinations such as bisoprolol atorvastatin aspirin tazobactam/piperacillin. Bivariate analysis showed that age, comorbidities, length hospital stay, drugs prescribed risk factors associated DDIs (p<0.05). multiple binary logistic regression analysis, significantly medications (p<0.0001). Conclusions observed prevalence ‘moderate’ severity. Prescription screening assists early identification prevention DDIs, enhancing safety quality patient-centered care.

Язык: Английский

---

Potential Adverse Drug Events Identified with Decision Support Algorithms from Janusmed Risk Profile—A Retrospective Population-Based Study in a Swedish Region

Pharmacy, Год журнала: 2024, Номер 12(6), С. 168 - 168

Опубликована: Ноя. 15, 2024

Adverse drug events (ADEs) occur frequently and are a common cause of suffering, hospitalizations, or death, can be caused by harmful combinations medications. One method used to prevent ADEs is using

Язык: Английский

---