Abstract
Non-invasive automatic screening for Alzheimer’s disease has the potential to improve diagnostic accuracy while lowering healthcare costs. Previous research has 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 for unobtrusively capturing data during real clinical interviews conducted as part of cognitive assessments for Alzheimer’s disease. The system uses nine different sensor devices (smartphones, a tablet, an eye tracker, a microphone array, and a wristband) to record interaction data during a specialist’s first clinical interview with a patient, and is currently in use at Karolinska University Hospital in Stockholm, Sweden. Furthermore, complementary information in the form of brain imaging, psychological tests, speech therapist assessment, and clinical meta-data is also available for each patient. We detail our data-collection and analysis procedure and present preliminary findings that relate measures extracted from the multimodal recordings to clinical assessments and established biomarkers, based on data from 25 patients gathered thus far. Our findings demonstrate feasibility for our proposed methodology and indicate that the collected data can be used to improve clinical assessments of early dementia.
Highlights
Alzheimer’s disease and other neurocognitive disorders with a neuropathological origin develop gradually over many years before existing criteria of a clinical diagnosis are fulfilled (Blennow et al, 2006; Jack et al, 2018)
From the 25 patients 4 patients were diagnosed with Alzheimer’s disease, 7 with mild cognitive impairment and 14 received a diagnosis of subjective cognitive impairment, meaning the clinical examination found no clinical signs of impairment
Looking at the statistics of pen motion and pen pressure, we found that two features were interesting: mean drawing gap length correlated with diagnosis (0.62), Moca-MIS (−0.61) and Hippocampus total volume (−0.58), and mean pen pressure correlated with p-tau (−0.88) and Hippocampus total volume (0.86)
Summary
Alzheimer’s disease and other neurocognitive disorders with a neuropathological origin develop gradually over many years before existing criteria of a clinical diagnosis are fulfilled (Blennow et al, 2006; Jack et al, 2018). Approaches based on machine learning have proved successful for processing complex information and assisting in medical decisions in several diseases (Hamet and Tremblay, 2017) In recent years, such methods have been developed for neurocognitive disorders (Bruun et al, 2019; Koikkalainen et al, 2019; Lee et al, 2019a). Machine learning has been used to combine many types of clinical data to further aid in the diagnosis of neurocognitive disorders (Bruun et al, 2019; Koikkalainen et al, 2019; Lee et al, 2019a) Another potential application of machine learning for neurocognitive disorders could be the automatic capture and analysis of behavioural signals of potential clinical relevance, both for reducing the risk that such signals are missed by the clinician and for adding new and complementary information beyond what normally is collected in the medical examination.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
