ARCHIVES

Stomach cancer is one of the major causes of cancer- related mortality worldwide due to the absence of effective early screening techniques. Conventional diagnostic methods such as biopsy, endoscopy, and imaging are invasive, expensive, and de- pendent on expert interpretation. This paper presents a detailed multimodal artificial intelligence-based system for early detection of stomach cancer by integrating biomarker data, medical image processing, and voice-based symptom analysis. Machine learning models are applied for biomarker and voice classification, while a deep Convolutional Neural Network is employed for gastric image classification. A decision-level fusion strategy is used to combine the individual predictions to enhance diagnostic accuracy and reliability. The experimental evaluation demonstrates that the proposed multi-modal system significantly outperforms single- modality approaches. The proposed framework provides a reliable, non-invasive, and cost-effective solution for early diagnosis and remote healthcare applications.