Popular deep learning algorithms for disease prediction: a review.

Area of Science:

• Artificial Intelligence
• Medical Informatics
• Computational Biology

Background:

• Deep learning (DL) is a dominant AI subset, recognized for automatic feature learning and high performance.
• DL applications are expanding across various fields, notably in medicine where its accuracy in disease prediction can exceed that of clinicians.

Purpose of the Study:

• To provide a comprehensive overview of deep learning algorithms used in disease prediction.
• To analyze current challenges and solutions in the disease prediction landscape.
• To explore future trends in medical AI, specifically Digital Twins and precision medicine.

Main Methods:

• Introduction to key deep learning algorithms: Artificial Neural Network (NN), FM-Deep Learning, Convolutional NN, and Recurrent NN.
• Discussion of their theoretical underpinnings, historical development, and practical applications in disease prediction.
• Analysis of existing limitations and proposed solutions within the disease prediction domain.

Main Results:

• Deep learning algorithms demonstrate significant potential in disease prediction accuracy.
• Identified limitations in current disease prediction methods.
• Highlighted emerging trends like Digital Twins and precision medicine.

Conclusions:

• Deep learning offers powerful tools for advancing disease prediction and medical research.
• The integration of Digital Twins and precision medicine represents the future trajectory for medical AI.
• This review aims to guide researchers in understanding and developing advanced disease prediction models.