Algorithms for the analysis of polysomnographic recordings with customizable criteria

摘要: The diagnosis of Sleep Apnoea-Hypopnoea Syndrome requires the visual inspection a recording containing large number physiological parameters patient - polysomnogram. purpose this is identification and characterization different types pathological events that occur over these parameters. These are defined by set morphological criteria. Based on criteria, commercial tools have been developed to support clinicians in task visually inspecting polysomnograms. This article argues standard criteria just guiding recommendations experienced physicians often adapt each specific diagnostic context. Thus, analysis polysomnograms ideally should use flexible could be easily customizable clinicians. In paper, we propose algorithms capable identifying relevant SAHS using custom acquired directly from clinician. take advantage Fuzzy Set Theory capture handle vagueness uncertainty characteristics medical knowledge. Knowledge acquisition traditional linguistic approach Sets supported desktop tool. However, authors feel some need more nature than linguistic. An alternative mechanism for proposed. Finally, when presenting identified, tool uses several metaphors designed simplify We validated our proposal 69h polysomnographic recordings arising 12 patients were subjected sleep study. 95.7% identified correct detections. rate false negatives was 1.6%.