Browse Articles
Discover research articles across all indexed journals
Accurate discharge summary generation using fine tuned large language models with self evaluation
Abstract Discharge summaries are critical for patient care continuity, clinical decision-making, and legal documentation, yet their creation is labor-intensive. Clinicians must manually integrate diverse data from multiple sources under time constraints, often leading to delays, inconsistencies, and potential omissions. This study introduces a novel framework to automate discharge summary generation using advanced natural language processing (NLP) techniques, aiming to reduce clinician workload while ensuring accurate, complete, and standardized documentation. We combine the Decomposed Low-Rank Adaptation (DoRA) fine-tuning method with a novel self-evaluation mechanism to enhance large language models (LLMs) for medical text generation. DoRA efficiently adapts pre-trained LLMs to the specialized medical domain, demonstrating superior performance over traditional methods such as LoRA and QLoRA, with a enhancement in BERTScore and a reduction in Perplexity across all evaluated models. The self-evaluation mechanism, inspired by cognitive psychology, iteratively re-feeds generated summaries together with segmented clinical data into the model, allowing it to systematically detect and correct omissions in each data segment, thereby ensuring the outputs accurately and comprehensively represent the original input. This approach was rigorously compared against few-shot prompting and Chain of Thought (CoT) methods. Extensive experiments show that self-evaluation improves BERTScore by 6.9% and 4.1% and increases ROUGE-L by 69.6% and 0.4% relative to few-shot and CoT baselines, respectively, while qualitative metrics also demonstrate consistent gains in accuracy and completeness. Our results demonstrate substantial enhancements in the quality and consistency of generated discharge summaries while reducing the time required for their creation. This research underscores the potential of AI-driven tools in healthcare documentation, reducing the time required for generating discharge summaries while improving their quality and consistency. The findings indicate promising prospects for automating medical documentation that adheres to high standards of accuracy and relevance.
The impact of urine output trajectory on clinical outcomes in female patients with genitourinary infections in the intensive care unit
Urinary complement C3 fragment levels and their clinical relevance in MPO-ANCA-associated vasculitis
Comprehensive analysis of florasulam: crystal structure, reactivity, sensitivity, and bioactivity using structural, spectroscopic, and computational approaches
Chemical composition, metabolomics, and functional potential of Pholiota nameko pre-cooking liquid
Reliable directional relaying for T-Connected Series-Compensated transmission lines
Abstract This article introduces an extensive directional relaying algorithm designed for series-compensated transmission lines. The proposed algorithm relies on observing the locus of the calculated impedance of the change of the positive-sequence circuit. Both the voltage and current signals in positive sequence circuit are observed to acquire the impedance perceived by the relay. The fault direction is identified by the quadrant in which the locus of the obtained impedance settles. The proposed algorithm is distinguished by its high performance under varying prefault power factor and active power flow direction. It is tested for a range of fault types, varying fault resistance values, diverse fault locations, and different compensation ratios. All tests have been conducted while considering the nonlinear behaviour of the shunt metal oxide varistor (MOV) with the series-compensation capacitor. Furthermore, the proposed scheme is validated with a T-connected multi-terminal Egyptian transmission system. It is extensively tested and compared with other relevant schemes in the literature.
Consensus for the most suitable trial design to assess therapy for rare vascular malformations: a Delphi study
Multi-modal ultrasound for evaluating structure and function of neurogenic bladder in pediatrics
The association between probable sarcopenia, non-motor symptoms, and fall-related outcomes in parkinson’s disease
Identifying and predicting gait stability metrics in people with stroke in uneven-surface walking using machine learning
Abstract People with stroke (PwS) face increased fall risk on uneven surfaces; however, gait stability under such conditions remains unexplored. This study used machine learning (ML) to identify acceleration features distinguishing PwS from healthy controls (HC) during uneven-surface walking and to predict them from even-surface gait parameters. Trunk acceleration data from 71 PwS and 39 HC were analyzed using classification and regression models. The ML classifiers achieved an accuracy of over 95%. The key discriminative features included the vertical root mean square (RMS_VT), anterior-posterior sample entropy (SampEn_AP), and harmonic ratio (HR_AP). In PwS, even-surface gait speed < 0.8 m/s predicted reduced speed and higher RMS_VT on uneven surfaces. SampEn_AP and HR_AP were influenced by ankle kinematics and their even-surface values, respectively, showing nonlinear associations. These findings support the use of wearable sensor data and interpretable ML to assess gait stability and adaptability, facilitating development of digital biomarkers for personalized stroke rehabilitation aimed at improving outdoor mobility.
Climatic and governance determinants of malaria transmission in Rivers State, Nigeria
Efficacy of ambulance air purifiers with different photocatalytic oxidation components in the removal of Bacillus subtilis spores
Abstract Ambulances are enclosed environments that carry a high risk of airborne and surface microbial transmission, yet effective disinfection technologies remain limited. This study evaluated four photocatalytic oxidation (PCO) configurations—O₃+UVA + TiO₂, UVA + TiO₂, O₃+UVC + ZnO, and UVC + ZnO—against B. subtilis spores. The testing employed a prototype air purification system for ambulance applications, where the photocatalyst TiO 2 or ZnO was integrated into a filter medium. This system operated in combination with its corresponding UV light source (UVA or UVC) and an optional ozone generator; all housed within a laboratory-simulated ambulance cabin. (8.998 m³), where spores at 1.5 × 10⁸ CFU/mL (8 mL) were spray misted using a nebulizer and sampled using an Andersen Impactor, following the NIOSH method. Disinfection efficacy was quantified as the percentage reduction of B. subtilis spores in the air and on surfaces. Among the tested systems, efficacy ranked as UVA + TiO₂ > O₃+UVA + TiO₂ > O₃+UVC + ZnO > UVC + ZnO. UVA + TIO 2 achieved the most rapid and stable disinfection among the tested systems under controlled conditions, reducing airborne spores by > 80% within 15 min, achieving complete removal within 90 min, and reducing surface contamination by 96.77% at 120 min. In contrast, ZnO- and UVC-based systems exhibited lower or inconsistent performance. These findings identify UVA + TiO₂ photocatalysis as a safe, ozone-free, and highly effective strategy for ambulance air purification. Its rapid and durable antimicrobial action demonstrates clear advantages over approaches based on ozone or UVC, offering practical benefits for infection control in emergency medical services and providing a foundation for further optimization of photocatalytic technologies in healthcare settings.