Design of a Computer System to Assist in Differential Preoperative Diagnosis for Pelvic Surgery

Abstract

To help in making nine common differential diagnoses for pelvic surgery, a computer system has been designed in three parts: a data-collection procedure, based on a checklist of symptoms obtained on admission, each with a weight between 1 and 9; a statistical diagnostic algorithm, for which medical records of over 500 patients were used to calculate the necessary coefficients; and a computer-based information output, which, using the first two parts, gives the physician a condensed history and the most probable diagnoses, before surgery. The system has been shown to be 66 per cent accurate —equal to results in major teaching hospitals. It is designed for further improvements by making very complete data collection and retrieval easy. It was developed for easy implementation at minimal cost.