Probabilistic Evaluation of Candidates and Symptom Clustering for Multidisorder Diagnosis
This work addresses the challenge of efficient diagnosis validation in medical AI, though it appears incremental as it builds on existing symptom clustering methods.
The paper tackles the problem of evaluating candidate disorder sets in multidisorder diagnosis by deriving a formula to compute their conditional probability, enabling validation or pruning of large candidate sets simultaneously without assumptions about disorders outside the candidate.
This paper derives a formula for computing the conditional probability of a set of candidates, where a candidate is a set of disorders that explain a given set of positive findings. Such candidate sets are produced by a recent method for multidisorder diagnosis called symptom clustering. A symptom clustering represents a set of candidates compactly as a cartesian product of differential diagnoses. By evaluating the probability of a candidate set, then, a large set of candidates can be validated or pruned simultaneously. The probability of a candidate set is then specialized to obtain the probability of a single candidate. Unlike earlier results, the equation derived here allows the specification of positive, negative, and unknown symptoms and does not make assumptions about disorders not in the candidate.