An expert system was developed to support clinical decision-making by ultrasound diagnosticians in assessing imaging findings in cases of acute appendicitis (AA) in children. The knowledge base is implemented using production rules derived from interviews with two experienced sonographers. The system performs differential diagnosis between AA and other abdominal pathologies and classifies various forms of appendicitis. The prototype was tested on real data from 100 patients. The system achieved a sensitivity of 91.36% and specificity of 68.42% for diagnosing AA vs. non-AA; sensitivity of 83.72% and specificity of 64.52% for differentiating phlegmonous from other AA forms; and sensitivity of 92.31% and specificity of 68.75% for distinguishing between phlegmonous and gangrenous forms. However, even experienced ultrasound diagnostic specialists cannot always detect signs of destructive appendicitis, which was observed in two cases in this work. Developing of a computer system is justified by the need for timely recognition of complications in AA, a leading cause of surgery in pediatric patients. Delayed care may result in progression from catarrhal to destructive forms (phlegmonous, gangrenous, or gangrenous-perforated), potentially leading to peritonitis—a life-threatening condition. Sonography is the primary diagnostic method for suspected AA in children, but conclusions based on ultrasound to depend on physician expertise. This highlights the relevance of intelligent system for recognizing different types of AA forms using sonographic data.
DOI: 10.1007/978-3-032-13612-1_31
Download the proceedings from Springer Nature: https://link.springer.com/content/pdf/10.1007/978-3-032-13612-1.pdf
Bogatikova, M. I., Belyaeva, A. V., Kobrinskii, B. A. (2026). Prototype of an Expert System for the Diagnosis of Destructive Forms of Acute Appendicitis in Children // Proceedings of the Ninth International Scientific Conference “Intelligent Information Technologies for Industry” (IITI’25), Volume 2, pp. 350–357.