Comparing a patient case with other cases having a similar progress in time can prove valuable for medical decision making. However, as the population under study increases, the complexity of such comparisons can exceed the capabilities of human medical experts, especially when multiple parameters are involved. In this paper we propose a novel computational approach to aid the comparison of multimodal medical data obtained from different time-series patient examinations. The similarity between the patient cases is assessed by a fusion scheme involving the estimation of multiple dynamic time warping distances between temporal sequences of observed medical parameter values. The results obtained from its application on a reference dataset of hepatic infections demonstrated high precision even for low recall rates. Thus, the effectiveness of the proposed approach which is generic renders it applicable on various medical domains.
展开▼
