Colon Cancer Disease Diagnosis Based on CNN and Machine Learning
6th International Interdisciplinary Symposium on Chaos and Complex Systems, SCCS 2025, İstanbul, Türkiye, 8 - 10 Mayıs 2025, ss.425-433, (Tam Metin Bildiri)
- Yayın Türü: Bildiri / Tam Metin Bildiri
- Doi Numarası: 10.1007/978-3-032-09101-7_36
- Basıldığı Şehir: İstanbul
- Basıldığı Ülke: Türkiye
- Sayfa Sayıları: ss.425-433
- Anahtar Kelimeler: Colon cancer, Convolutional neural networks, Feature selection, Machine learning
- İstanbul Kültür Üniversitesi Adresli: Evet
Özet
Colon cancer is a leading cause of cancer-related mortality worldwide, underscoring the urgent need for effective and accurate diagnostic techniques. This study proposes a novel hybrid approach for colon cancer diagnosis using convolutional neural networks (CNNs) combined with the Bitterling Fish Optimization (BFO) algorithm for enhanced feature selection and classification. Utilizing a publicly available histopathological image dataset from Kaggle, the proposed method begins with extracting high-level features through pre-trained CNN models such as VGGNet, AlexNet, and SqueezeNet. These features are then refined using the BFO algorithm to select the most informative attributes. Multiple machine learning classifiers—including Support Vector Machine (SVM), Decision Tree, Ensemble methods, Naive Bayes, Logistic Regression, and K-Nearest Neighbors (KNN)—are employed for the final classification of images into cancerous and non-cancerous categories. The VGGNet-BFO model paired with the SVM classifier achieved the highest accuracy of 98.21%, demonstrating the model’s robustness and efficiency. The findings highlight the potential of deep learning models integrated with bio-inspired optimization algorithms for reliable and automated colon cancer diagnosis.