StrokeDx: A Logic Programming System to Diagnose Stroke

Jeffrey L. Sponsler


expert system, stroke, prolog, monitoring


Decision support for stroke diagnosis is important given the complexity of diagnosis. Write a software system to diagnose stroke using logic programming technology. Using the conventional data elements for neurological workup, a set of benchmarks for stroke syndromes and a peripheral neuropathy syndrome have been encoded. Using a logic programming and back-chaining rule base, a program called StrokeDx to diagnose stroke has been created. Development of Stroke Benchmarks and rule sets was done at the same time using rapid prototyping. Each diagnostic rule base was applied to all benchmark datasets to compute a confidence factor from 0 to 1. Diagnoses included frontal stroke, occipital stroke, Wallenberg syndrome, and radial neuropathy. The sensitivity of each diagnostic rule set (for the appropriate diagnosis) was 100%. Our approach has demonstrated technical feasibility, the goals were met, and encoding of other neurological syndromes is in progress.

Important Links:

Go Back