Survival prediction landscape: an in-depth systematic literature review on activities, methods, tools, diseases, and databases

Area of Science:

• Computational biology
• Bioinformatics
• Medical informatics

Background:

• Survival prediction integrates molecular and clinical data for forecasting disease events.
• AI methods have advanced survival prediction, but disease-specific models are crucial.
• A centralized platform is needed to consolidate knowledge on existing survival predictors.

Purpose of the Study:

• To comprehensively analyze recent survival predictors across diverse diseases.
• To provide insights into methods, data modalities, and features used in disease-specific predictors.
• To identify trends and gaps in survival prediction research.

Main Methods:

• Systematic review of 23 existing review studies.
• Analysis of 90 recent survival predictors across 44 diseases.
• Examination of data modalities, clinical features, feature engineering, and AI/ML approaches.

Main Results:

• Detailed insights into diverse methods for disease-specific survival prediction.
• Analysis of data sources, feature subsets, and statistical/machine learning techniques.
• Overview of open-source predictors and survival prediction frameworks.

Conclusions:

• The study synthesizes current knowledge on survival prediction models.
• Identifies key components and methodologies in developing effective predictors.
• Provides a foundation for future advancements in AI-driven precision medicine.