Applications of Deep Learning in Endocrine Neoplasms

Area of Science:

• Medical Informatics
• Computational Pathology
• Oncology

Background:

• Machine learning (ML) and deep learning (DL) are increasingly vital in medical applications.
• DL enables computational analysis of histological samples, supporting diagnosis and characterization in various diseases.
• DL shows promise in endocrine cancer research for tasks like tumor grading and gene expression prediction.

Purpose of the Study:

• To review the current applications of deep learning (DL) in endocrine cancer histopathology.
• To emphasize experimental design, key findings, and limitations in existing DL research for endocrine cancers.

Main Methods:

• Systematic review of recent literature on deep learning in endocrine cancer histopathology.
• Analysis of studies focusing on experimental design, methodologies, and reported outcomes.
• Identification of common themes, successes, and challenges in the field.

Main Results:

• DL methods have been successfully applied to various endocrine cancer histopathology tasks.
• Significant findings include improved diagnostic accuracy and predictive capabilities.
• Identified limitations relate to data variability, model interpretability, and generalizability.

Conclusions:

• Deep learning holds significant potential to advance endocrine cancer diagnosis and research.
• Further research is needed to address current limitations and facilitate clinical translation.
• Standardization of experimental design and validation is crucial for robust DL applications in endocrine cancer histopathology.