Digital biomarker‐based individualized prognosis for people at risk of dementia

Abstract

Background Research investigating treatments and interventions for cognitive decline fail due to difficulties in accurately recognizing behavioral signatures in the presymptomatic stages of the disease. For this validation study, we took our previously constructed digital biomarker‐based prognostic models and focused on generalizability and robustness of the models. Method We validated prognostic models characterizing subjects using digital biomarkers in a longitudinal, multi‐site, 40‐month prospective study collecting data in memory clinics, general practitioner offices, and home environments. Results Our models were able to accurately discriminate between healthy subjects and individuals at risk to progress to dementia within 3 years. The model was also able to differentiate between people with or without amyloid neuropathology and classify fast and slow cognitive decliners with a very good diagnostic performance. Conclusion Digital biomarker prognostic models can be a useful tool to assist large‐scale population screening for the early detection of cognitive impairment and patient monitoring over time.

Innovative motor and cognitive dual-task approaches combining upper and lower limbs may improve dementia early detection

“…That justifies the increasing efforts to identify reliable predictors of the disease 3 . Besides, the scientific community advocates the identification of valid, inexpensive, and non-invasive markers, that may improve the early detection of these conditions 4 , 5 . In this framework, Mild Cognitive Impairment (MCI), a 10-fold risk condition of progressing to dementia 6 , represents a crucial target for clinical research.…”

Artificial intelligence as an aid to diagnosing dementia: an overview

“…The development of MCI often manifests as behavioural change with caregivers being first to recognise subtle alterations in baseline activities 36 . At present, exclusive application of neuropsychological assessments can fail to reveal information regarding these changes related to early MCI, and despite their widespread use and reported efficacy they cannot predict an individual's behaviour progression in a real‐life scenario 37,38 . This only highlights a need to develop systematic tools to perform behavioural analysis during early stages of MCI and integrate this with future clinical practice.…”

“…With an ever‐growing accessibility to smartphone applications and tablets that have potential to test behavioural change and cognition, the addition of digital biomarkers alongside cognitive testing and known biological diagnostic factors has presented itself 37 . Early detection of MCI based on behavioural analysis includes technological approaches, including ML, virtual reality (VR) and gait kinematic analysis, which are currently being explored as part of a new conventional assessment.…”

“…An example of this described by Buegler et al . (2020) 37 is a smartphone application called ‘Altoida instrumental activities of daily living (iADL) test’; this provides interactive tasks at varying difficulties through use of augmented reality and aims to test cognitive control, prospective memory, executive function, spatial memory and processing speed. Users are guided by technology to perform fine and gross motor tasks on a touch screen, ie hiding objects that appear on touchscreen, exploring their direct environment via a live camera.…”

“…Digital biomarkers as discussed above have shown promising results regarding prognostic predictive value and can be incorporated in a widespread manner both in the community and clinical settings. This is particularly important with a recognised potential threefold difference in conversion rate from MCI to dementia between recruits from community and memory clinics 37,38 . The growth of this approach could offer patients and caregivers an optimised quality of life and treatment plan 40 …”