Lung and Colon Cancer Detection Using a Deep AI Model.

Area of Science:

• Oncology
• Medical Imaging
• Artificial Intelligence

Background:

• Lung and colon cancers are leading causes of mortality worldwide.
• Accurate and early cancer detection is critical for effective treatment and patient survival.
• Traditional tissue sample analysis is complex, time-consuming, and prone to errors.

Purpose of the Study:

• To develop a novel, efficient, and mobile-embedded deep learning model for accurate lung and colon cancer detection.
• To address the limitations of existing deep learning models in terms of accuracy and resource requirements.
• To classify specific lung and colon cancer types from digital pathology images.

Main Methods:

• A lightweight, parameter-efficient 1D convolutional neural network (CNN) with squeeze-and-excitation layers was proposed.
• The model was trained and validated on the histopathological (LC25000) lung and colon datasets.
• The model was designed for efficient computation and mobile embedding.

Main Results:

• The proposed model achieved 100% accuracy in detecting lung cancer, colon cancer, and combined lung and colon cancers.
• This 100% accuracy was achieved with approximately 0.35 million trainable parameters and 6.4 million floating-point operations (FLOPs).
• The model demonstrated state-of-the-art performance compared to existing methods on the benchmark datasets.

Conclusions:

• The developed deep learning model offers a highly accurate and efficient solution for diagnosing lung and colon cancers.
• Its lightweight and parameter-efficient design makes it suitable for mobile-embedded applications, potentially increasing accessibility.
• This research represents a significant advancement in AI-driven cancer detection, promising improved diagnostic capabilities and patient outcomes.