Computer-Aided Diagnosis System for Detection of Stomach Cancer with Image Processing Techniques

Area of Science:

• Gastroenterology
• Oncology
• Medical Imaging
• Computer Science

Background:

• Stomach cancer incidence is rising globally and in Turkey.
• Early detection of gastric cancer is difficult due to minimal initial symptoms.
• Endoscopic diagnosis, while sensitive, can miss cancerous areas due to human observation limitations, potentially leading to recurrence.

Purpose of the Study:

• To develop a computerized decision support system (CDS) to assist in early stomach cancer detection.
• To improve the accuracy of identifying cancerous regions in endoscopic images.
• To aid in targeted biopsy collection for definitive diagnosis and treatment planning.

Main Methods:

• Implementation of a computerized decision support system (CDS) utilizing image processing techniques.
• Collaboration between specialist physicians and computer scientists to develop the system.
• Integration of the CDS as an assistant tool for gastroenterologists during endoscopic procedures.

Main Results:

• The CDS system assists doctors in identifying potential cancerous areas within endoscopic images.
• The system aids in guiding biopsy sampling from suspicious regions.
• The developed model is considered useful for improving gastric cancer diagnosis and management.

Conclusions:

• Computerized decision support systems can enhance the accuracy of stomach cancer detection during endoscopy.
• The integration of image processing and expert knowledge improves diagnostic capabilities.
• This approach has the potential to reduce missed diagnoses and cancer recurrence.