The core message of this article is that by leveraging the inherent hierarchical structure of whole slide images (WSIs) and transcriptomic data, and efficiently modeling intra-modal and inter-modal interactions at different granularity levels, the proposed SurvMamba framework can achieve superior performance in survival prediction compared to existing state-of-the-art methods, while also being computationally more efficient.
A novel binary version of the Child Drawing Development Optimization (BCDDO) algorithm is proposed for efficient feature selection to improve classification accuracy on various medical datasets.
Federated learning models can accurately predict major postoperative complications using electronic health record data from multiple institutions, while preserving data privacy.
Bayesian neural network architectures can provide uncertainty estimates in survival analysis, improving prediction performance and calibration compared to traditional non-Bayesian approaches.
TransformerLSR is a novel deep learning framework that can jointly model longitudinal measurements, recurrent events, and survival outcomes, while accounting for their complex dependencies and incorporating known clinical knowledge.
A novel deep learning approach, Transformer-based Diffusion Probabilistic Model for Sparse Time Series Forecasting (TDSTF), achieves state-of-the-art performance in predicting heart rate, systolic blood pressure, and diastolic blood pressure in the intensive care unit setting.
Our Bayesian neural controlled differential equation (BNCDE) model provides meaningful posterior predictive distributions of potential outcomes, enabling reliable decision-making in personalized medicine.
A multilevel stochastic optimization approach based on computational applied mathematics techniques can accurately and efficiently impute missing values in massive medical datasets, significantly outperforming current state-of-the-art methods.
Appropriate machine learning algorithms with carefully selected features and balanced data can accurately predict mortality, ICU requirement, and ventilation support for COVID-19 patients.