Multimodal Capture of Patient Behaviour for Improved Detection of Early Dementia: Clinical Feasibility and Preliminary Results

• DOI: 10.3389/fcomp.2021.642633
• Publication Date: 2021-04-19T16:07:51Z
• Authors: Patrik Jonell; Birger Moëll; Krister Håkansson; Gustav Eje Henter; Taras Kucherenko; Olga Mikheeva; Göran Hagman; Jasper Holleman; Miia Kivipelto; Hedvig Kjellström; Joakim Gustafson; Jonas Beskow
• Keywords: 03 medical and health sciences; mild cognitive impairment; 0302 clinical medicine; speech; Electronic computers. Computer science; Alzheimer; multimodal prediction; QA75.5-76.95; gaze; pupil dilation; 3. Good health

Non-invasive automatic screening for Alzheimer’s disease has the potential to improve diagnostic accuracy while lowering healthcare costs. Previous research shown that patterns in speech, language, gaze, and drawing can help detect early signs of cognitive decline. In this paper, we describe a highly multimodal system unobtrusively capturing data during real clinical interviews conducted as part assessments disease. The uses nine different sensor devices (smartphones, tablet, an eye tracker, microphone array, wristband) record interaction specialist’s first interview with patient, is currently use at Karolinska University Hospital Stockholm, Sweden. Furthermore, complementary information form brain imaging, psychological tests, speech therapist assessment, meta-data also available each patient. We detail our data-collection analysis procedure present preliminary findings relate measures extracted from recordings established biomarkers, based on 25 patients gathered thus far. Our demonstrate feasibility proposed methodology indicate collected be used dementia.