CNN-Based Classification of Histopathology Images for the Diagnosis of Lung Cancer

CNN-Based Classification of Histopathology Images for the Diagnosis of Lung Cancer

T. Thirupati Rao 1, J. Deepthi2, D. Prabhas3, G. M. Krishnam Naidu4, B. Teja5

1 T. Thirupati rao

2J. Deepthi

3 D. Prabhas

4 G. M. Krishnam Naidu

5 B. Teja

1Assistant Professor,

[2-5] B. Tech Student, LIET

[1,2,3,4,5] Computer Science & Engineering, Lendi Institute of Engineering and Technology, Vizianagaram

---------------------------------------------------------------------***---------------------------------------------------------------------

Abstract - Cancer is the uncontrollable division of abnormal cells that can spread to other organs. Lung cancer, the most common type in both men and women, includes two main types: small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC). NSCLC constitutes about 80 percent of lung cancer cases and typically spreads to other body parts, with subtypes including adenocarcinoma and squamous cell carcinoma. Diagnostic methods for lung cancer include X-rays, CT scans, bronchoscopy, and biopsy. Histopathology, the examination of biopsy samples by a pathologist, is crucial for diagnosing diseases by analysing tissue and cells. Histopathology image analysis offers more comprehensive diagnostic information compared to mammography and CT scans, especially for cancer grading. To expedite lung cancer diagnosis and alleviate the pathologist's workload, deep learning techniques are employed. Our proposed system utilizes Convolutional Neural Networks (CNN) combined with Rectified Linear Unit (ReLU) and Batch Normalization (BN) to enhance the diagnostic process.

Key Words:  CNN, Rectified Linear Unit (ReLU), Batch Normalization (BN), Histopathology image.