Digital health · 2026
Predicting cancer treatment outcomes using machine learning enabled clinical decision support systems.
СтатьяОбзорНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Hossain Adib, Rashid Md Mohaimin, Alam Towsif, Riipa Muslima Begom, Hassan Mahafuj, Imran Md Ahsan Ullah, Mohammad Nur, Ahmed Fahad, Parvin Mst Rina
Аннотация
BACKGROUND: Cancer treatment poses significant challenges due to variability in patient responses, disease progression, and therapy outcomes. Traditional decision-making frameworks often fall short in integrating complex data streams, underscoring the need for intelligent systems. Machine learning (ML)-enabled clinical decision support systems (CDSS) offer a promising solution by enabling personalized, predictive, and data-driven oncology care.
AIM: This scoping review aimed to explore how ML-powered CDSS are applied to predict treatment outcomes across different types of cancer. It sought to identify the types of machine learning models used, data modalities involved, predictive endpoints targeted, and the extent of clinical implementation and validation.
METHODS: A systematic search was conducted across six databases covering studies from 2010 to 2025. Eligible studies included those deploying ML algorithms within CDSS frameworks for outcome prediction in oncology. Data extraction followed a structured charting process, and studies were assessed using the Mixed Methods Appraisal Tool (MMAT). Findings were synthesized narratively and through thematic categorization.
RESULTS: A total of 32 studies were included. Predictive objectives ranged from survival estimation and therapy response to toxicity risk and recurrence detection. ML techniques varied from decision trees and vector machines to deep learning models such as convolutional neural networks. While technical performance was promising, few studies demonstrated external validation or integration into clinical workflows. Interpretability, ethical considerations, and patient involvement were frequently underreported.
CONCLUSION: ML-enabled CDSS have shown significant potential in predicting cancer treatment outcomes, yet their adoption in practice remains limited. Advancing these systems requires focus on validation, interpretability, data integration, and ethical design to bridge the gap between innovation and clinical utility.
AIM: This scoping review aimed to explore how ML-powered CDSS are applied to predict treatment outcomes across different types of cancer. It sought to identify the types of machine learning models used, data modalities involved, predictive endpoints targeted, and the extent of clinical implementation and validation.
METHODS: A systematic search was conducted across six databases covering studies from 2010 to 2025. Eligible studies included those deploying ML algorithms within CDSS frameworks for outcome prediction in oncology. Data extraction followed a structured charting process, and studies were assessed using the Mixed Methods Appraisal Tool (MMAT). Findings were synthesized narratively and through thematic categorization.
RESULTS: A total of 32 studies were included. Predictive objectives ranged from survival estimation and therapy response to toxicity risk and recurrence detection. ML techniques varied from decision trees and vector machines to deep learning models such as convolutional neural networks. While technical performance was promising, few studies demonstrated external validation or integration into clinical workflows. Interpretability, ethical considerations, and patient involvement were frequently underreported.
CONCLUSION: ML-enabled CDSS have shown significant potential in predicting cancer treatment outcomes, yet their adoption in practice remains limited. Advancing these systems requires focus on validation, interpretability, data integration, and ethical design to bridge the gap between innovation and clinical utility.