Objective: To present a knowledge-based classification system, GASTRO, for gastric lesion diagnosis.
Study design: The study group consisted of 150 patients with various gastric lesions (25 relatively normal, 30 gastritic changed, 25 ulcerous, 25 adenomatous, 25 cancerous and 20 metastatic cases). After a subtotal gastrectomy, the lesions were routinely diagnosed and quantitatively evaluated. The control group consisted of 25 patients with relatively normal mucosae. Both quantitative methods were used for determining the dynamics of malignant processes of gastric mucosa.
Results: During the consultation more than 20 factors influencing disease dynamics were considered. Statistics on cellular and nuclear parameter changes in stomach mucosa were also used in the diagnosis of gastric lesions.
Conclusion: Diagnoses were generated with confidence levels of 90%, 95%, and 99%. The final decision remained the responsibility of the physician.