Augmented histopathology: Enhancing colon cancer detection through deep learning and ensemble techniques

Area of Science:

• Oncology
• Medical Imaging
• Artificial Intelligence

Background:

• Colon cancer presents a significant global health challenge with high mortality rates.
• Early and accurate detection is critical for effective treatment and improved patient survival.
• Histopathological image analysis is key for diagnosing colon cancer subtypes.

Purpose of the Study:

• To develop and evaluate a deep learning model for enhanced colon cancer detection and classification.
• To improve the accuracy and efficiency of diagnosing colon cancer from histopathological images.
• To leverage advanced AI techniques for better patient outcomes in colon cancer care.

Main Methods:

• Utilized the CRC-VAL-HE-7K dataset for histopathological colon cancer images.
• Implemented a hybrid deep learning model combining Convolutional Neural Networks (CNN) and transformers for feature extraction.
• Employed a Siamese network with an attention mechanism for improved classification accuracy and incorporated hybrid Particle Swarm Optimization (PSO) and Dwarf Mongoose Optimization (DMO) for model tuning.

Main Results:

• The proposed deep learning model achieved a highest accuracy rate of 98.84% in multi-class classification of colon cancer tissues.
• The model demonstrated superior performance compared to existing methods across various analytical metrics.
• The hybrid PSO-DMO optimization algorithm effectively fine-tuned model parameters, enhancing classification capabilities.

Conclusions:

• The developed deep learning approach significantly enhances colon cancer detection and classification accuracy.
• The hybrid CNN-transformer model with attention and Siamese network offers a robust solution for histopathological image analysis.
• This research highlights the potential of advanced AI in improving diagnostic precision for colon cancer.