Comparative Analysis of Stomach Cancer Detection Using Endoscopy and CT Scan Images with Machine Learning

Author

Lalitha K, Thrisha M C, Vaishnavi R S, Varshini M Devu, Vinu Ramyasree S A


Abstract

Early and accurate identification of stomach cancer plays a vital role in improving patient survival rates. Medical imaging techniques such as CT scans and gastrointestinal endoscopy are widely used for diagnosis; however, their manual interpretation requires significant clinical expertise and is often affected by observer variability. To address these challenges, this paper proposes a dual-modal deep learning framework for automated stomach cancer classification using separate MobileNetV2-based models for CT and endoscopic images. Each imaging modality undergoes tailored preprocessing, including window-level optimization and contrast enhancement for CT scans, as well as color normalization and illumination correction for endoscopy images. Model robustness is further improved through data augmentation and progressive fine-tuning strategies. A unified Gradio-based interface is developed to facilitate real-time image upload and prediction for both modalities. Experimental evaluations show strong classification performance for both CT and endoscopy pipelines, highlighting the benefits of modality-specific learning. Overall, the proposed system provides an efficient and clinically applicable solution for automated stomach cancer detection


Keywords

Stomach Cancer Detection, Deep Learning, MobileNetV2, CT scan, Endoscopy Images, Dual-Modal Classification, Medical Image Analysis, Gradio Interface



Full Text:

Download Paper PDF


References


S. Zhou, Y. Xie, X. Feng, Y. Li, L. Shen, and Y. Chen, “Artificial intelligence in gastrointestinal cancer research: Image learning advances and applications,” Cancer Letters, vol. 614, Feb. 2025.

2. Wang, C. Zhou, W. Wang, and H. Zhang, “A multimodal deep learning model for detecting endoscopic images of near-infrared fluorescence capsules,” Biosensors and Bioelectronics, vol. 278, Feb. 2025.

3. C. Robles-Medranda, I. Verplan, D. Schulz, and M. Spadaccini, “Artificial Intelligence in Biliopancreatic Disorders: Applications in Cross-Imaging and Endoscopy,” Gastroenterology, Apr. 11, 2025.

 4. E. U. Haq, Z. Yuan, H. Jianjun, R. Rizwan, and U. Haq, “Accurate detection and segmentation of gastric cancer based on a hybrid cascaded deep learning model with a vision transformer from endoscopic images,” Information Sciences, vol. 670, Apr. 12, 2024.

a. M. Rai, J. Yoo, and A. Razaque, “Comparative analysis of machine learning and deep learning models for improved cancer detection: A comprehensive review of recent advancements in diagnostic techniques,” Expert Systems with Applications, vol. 255, Jul. 23, 2024.

 5. N. B. Marya, S. Pawa, and N. R. Thiruvengadam, “American Society for Gastrointestinal Endoscopy guideline on the role of therapeutic EUS in the management of biliary tract disorders: methodology and review of evidence,” Gastrointestinal Journal, vol. 100, Mar. 2024.

 6. D. Fan, H. Liang, and C. Qu, “A novel approach for early gastric cancer detection using a hybrid of AlexNet, Extreme Learning Machine, and Adjusted Gorilla Troops Optimization,” Biomedical Signal and Control, vol. 95, Feb. 24, 2024.

7. D. K. T. Thoi, K. Seo, and J. H. Lim, “Diagnosis of invasive pancreatic cancer in endoscopic ultrasound images leveraging translation models,” Biomedical Signal Processing and Control, vol. 102, Dec. 24, 2024.

8. He, M. Zhang, W. Yu, and Y. Peng, “SaB-Net: Self- attention backward mechanism for gastric tumor segmentation in CT images,” Computers in Medicine and Imaging, vol. 105, Dec. 19, 2023.

 9. M. Rai, J. Yoo, and S. Dashkevych, “Advancements in traditional machine learning techniques for detection and diagnosis of fatal cancer types: Comprehensive review of biomedical imaging datasets,” Measurement, vol. 225, Dec. 22, 2023.

10. https://www.kaggle.com/datasets/orvile/coloncancerct- 2025-abdominal-ct-s-cans

11. https://www.kaggle.com/datasets/yasserhessein/the- kvasir-dataset

Share your valuable work from Social Media Buttons