Deep Learning in Colorectal Cancer Classification: A Scoping Review

Area of Science:

• Oncology
• Medical Imaging
• Artificial Intelligence

Background:

• Colorectal cancer (CRC) diagnosis and classification present challenges for medical professionals.
• Accurate CRC classification is crucial for effective treatment planning and patient outcomes.
• Artificial intelligence (AI), particularly deep learning, offers potential advancements in diagnostic accuracy and speed.

Purpose of the Study:

• To conduct a scoping review on the application of deep learning for classifying different types of colorectal cancer.
• To explore the current research landscape and identify trends in deep learning for CRC classification.

Main Methods:

• A systematic search was conducted across five databases.
• 45 relevant studies were selected based on predefined inclusion criteria.
• Data extraction focused on deep learning models, data types, and classification performance for CRC.

Main Results:

• Deep learning models have been applied to classify colorectal cancer using diverse data sources.
• Histopathology and endoscopy images are the most frequently utilized data types in these studies.
• Convolutional Neural Networks (CNNs) were the predominant deep learning architecture employed for CRC classification.

Conclusions:

• Deep learning, especially CNNs, demonstrates significant potential in enhancing the classification of colorectal cancer.
• Further research utilizing histopathology and endoscopy images with deep learning models is warranted.
• AI-driven tools could support pathologists and imaging specialists in improving CRC diagnosis and care.