Healthcare services are facing challenges in increasing their efficiency, quality of care, and coping with surges in demand. To this end, some hospitals have implemented lean healthcare. The aim of this systematic review is to evaluate the effects of lean healthcare (LH) interventions on inpatient care and determine whether patient flow and efficiency outcomes improve. The review was performed according to PRISMA. We used six databases to search for studies published from 2002 to 2019. Out of 5732 studies, 39 measuring one or more defined outcomes were included. Hospital length of stay (LOS) was measured in 23 studies, 16 of which reported a reduction, turnover time (TOT) decreased in six out of eight studies, while the turnaround time (TAT) and on-time starts (OTS) improved in all five and seven studies, respectively. Moreover, eight out of nine studies reported an earlier discharge time, and the boarding time decreased in all four cases. Meanwhile, the readmission rate did not increase in all nine studies. Lastly, staff and patient satisfaction improved in all eight studies. Our findings show that by focusing on reducing non-value-added activities, LH contributed to improving patient flow and efficiency within inpatient care.

Purpose: This research aims to implement Lean Manufacturing (LM) techniques and tools (T&T) in a retail distribution center located in Mariano Roque Alonso, Paraguay, targeting waste reduction and efficiency improvements in fresh and frozen product (FFP) handling to reduce long lead times.Design/methodology/approach: This study was conducted in two phases. First, a literature review revealed the key gaps to be addressed, the most commonly faced problems, and the LM tools applicable to solving these issues. Second, a case study was conducted in which LM T&T were applied to reduce the long lead time and propose solutions to address difficulties in the FFP process. Five problem-detection tools (flowchart, brainstorming, Ishikawa diagram, Pareto, and value stream mapping) and three improvement tools (kaizen, 5S, and Plan-Do-Check-Act) were applied to streamline FFP process from order generation to store reception.Findings: The analysis revealed significant inefficiencies: four were workforce-related, ten related to work methods, six environmental, and three product-related. Addressing these issues could substantially reduce operational bottlenecks and improve process throughput.Practical implications: The application of LM techniques significantly minimized waste, enhancing time management and human resource utilization, which led to a 96% improvement in FFP processing efficiency. These changes are expected to shorten lead times, bolster competitive advantage, and increase customer satisfaction.Originality/value: This study underscores the versatility of LM techniques, adapting them from manufacturing to retail distribution. The innovative combination of kaizen with 5S and PDCA offers a robust framework for ongoing improvements, promising for broader adoption in the retail sector.

Background/Objectives: Efficient management of intensive care unit (ICU) resources is a critical challenge for modern healthcare systems, which must balance high-quality patient care with operational and financial performance. ICU length of stay (LOS) is a key metric of clinical complexity and hospital efficiency. However, traditional methods for predicting LOS often fail to capture the complex, nonlinear interactions among physiological, demographic, and treatment-related variables. Machine learning (ML) and deep learning (DL) models have emerged as promising tools for enhancing predictive accuracy and supporting data-driven decision-making. Methods: This study presents a systematic review and meta-analysis of ML and DL approaches for predicting ICU LOS in adult patients. Following PRISMA guidelines, eight scientific databases were searched, yielding 33 eligible studies published between 2015 and 2025. Results: Mixed medical–surgical ICUs were the most common setting (51.5%), and 45.5% of datasets were sourced from public repositories. Most studies (19/33) focused on binary classification of prolonged stays, although thresholds ranged from >48 h to ≥14 days. The pooled results from ten studies yielded an AUROC of 0.9005 (95% CI: 0.8890–0.9121), indicating strong predictive capability across diverse clinical contexts. Subgroup analyses showed comparable performance between specialized surgical and general ICUs. Conclusions: These findings suggest that AI-driven LOS prediction models exhibit strong discriminatory power for ICU LOS prediction, supporting hospital capacity planning. However, to translate this into reliable clinical support, the methodological heterogeneity, scarcity of external validation, and near absence of calibration reporting identified in this review need to be addressed.

Introduction. Cardiovascular disease represents a substantial global health burden, and delays in diagnosis can further compromise clinical outcomes. Objective. This review aimed to evaluate the diagnostic effectiveness of artificial intelligence models across major cardiovascular conditions, with a specific focus on coronary artery disease, acute coronary syndromes and myocardial infarction, and heart failure. Method. Following PRISMA guidelines, we conducted a systematic review and applied meta-analytical techniques to pool measures of diagnostic performance. Random effects models with restricted maximum likelihood and Hartung–Knapp adjustments were used to calculate summary estimates, and analyses were stratified by cardiovascular condition. Only effects with variance (reported 95% CI or derivable denominators) were pooled. Areas under the receiver operating characteristic curve without variance were synthesized narratively. Results. Out of 39,383 records, thirty-five studies were eligible. The pooled area under the receiver operating characteristic curve was 0.823 (95% confidence interval 0.754–0.892). Heterogeneity was considerable (I² = 98.4%), and the 95% prediction interval was 0.59 to 1. However, 9 of 35 included studies reported external/independent validation. The pooled estimate therefore represents an overall descriptive average, reflecting the expected range of diagnostic performance rather than a single universal value. Stratified results showed pooled values of 0.88 for acute coronary syndromes and myocardial infarction, 0.87 for heart failure, and 0.86 for coronary artery disease. Sensitivity often reached 0.90 or higher in electrocardiography deep learning models, while specificity remained variable. Conclusion. Artificial intelligence models demonstrated promising discrimination across cardiovascular conditions, with variable specificity and predictive value, indicating that clinical utility requires external validation and local calibration.

El presente estudio evaluó la implementación de Lean Healthcare en el Departamento de Conservación y Mantenimiento de un centro de salud de atención primaria en México, con el objetivo de mejorar la capacidad de respuesta a las solicitudes de mantenimiento y mejorar la eficiencia operativa. Para lograrlo, se desarrolló un proyecto piloto durante siete meses, utilizando herramientas como 5S, Mapeo de procesos y Gestión visual. A través de 200 encuestas estructuradas, se evaluó la percepción post-implementación y se midieron distintas dimensiones clave, como la eficiencia en el proceso de solicitud, comunicación y coordinación, y la satisfacción del personal. La intervención resultó en mejoras significativas, incluyendo la reducción en el tiempo de respuesta de 48 a 33,6 h, la distancia total recorrida del personal de 801 a 211,5 metros y un incremento en la capacidad de respuesta del 65 a un 87%. Además, se obtuvo una tasa de respuesta de la encuesta del 81%, indicando una percepción general positiva de las prácticas Lean (α ≥ 0,799). Los hallazgos de esta investigación revelaron que Lean Healthcare es una herramienta eficaz para optimizar los procesos de mantenimiento en el sector de la salud; no obstante, para lograr un rendimiento sostenido y el éxito de futuras iniciativas, es indispensable fortalecer la cultura organizacional, fomentar la participación activa del personal y establecer mecanismos de seguimiento y evaluación continua.

En los últimos años, la inteligencia artificial y el aprendizaje automático se expandieron rápidamente en el sector de la salud, no obstante, la literatura reportó los algoritmos de forma inconsistente. El objetivo de este estudio fue identificar y caracterizar las familias de algoritmos con mayor frecuencia reportadas en aplicaciones clínicas, estratificadas por modalidad de datos y tarea clínica. Para este fin, se realizó una revisión de alcance y se recuperó literatura publicada entre 2020 y 2025 en bases de datos médicas. Se incluyeron artículos de revistas que reportaron modelos de inteligencia artificial entrenados o evaluados y que presentaron métricas cuantitativas. Se extrajeron datos sobre arquitecturas de modelos, dominios clínicos, tareas, conjuntos de datos, validación y rendimiento. Se identificaron 20 estudios primarios de implementación que abarcaron tres modalidades principales: imágenes médicas (n = 9), datos tabulares de historia clínica electrónica (n = 7) y señales fisiológicas (n = 4). Las redes neuronales convolucionales dominaron en imágenes, mientras que los modelos híbridos prevalecieron en señales. Los modelos de aprendizaje automático clásico fueron comunes en predicción tabular. En conjunto, los modelos de inteligencia artificial en salud mostraron patrones claros de especialización por modalidad de datos, con arquitecturas profundas que dominaron datos no estructurados y métodos clásicos que persistieron en datos tabulares.