An Algorithmic Approach for the Detection and Classification of Lung Cancer from the Histopathological Images
Karthikeyan Shanmugam, Harikumar Rajaguru · 2023
Lung cancer, a prevalent malignancy affecting both genders, demands early detection to mitigate its significant human toll. Unfortunately, symptoms often manifest in advanced stages, necessitating reliance on medical imaging for early diagnosis. Researchers have devised algorithms geared towards identifying lung cancer in histopathological images. This study introduces a computer-aided detection method using Flower Pollination Optimization for feature extraction from histopathological images. The study employs standard histopathological images sourced from the Kaggle database. In this investigation, five classifiers, including the Gaussian Mixture Model (GMM), Non-Linear Regression (NLR), Naïve Bayes Classifier (NBC), Logistic Regression (LR), and Fisher Discriminant Classifier (FDC), are utilized to assess and categorize the Benign and Adenocarcinoma classes. Results show that the Naïve Bayes Classifier, combined with Flower Pollination, achieves the highest accuracy at 87.86%, according to established benchmark metrics.