Multimodal learning of clinically accessible tests to aid diagnosis of neurodegenerative disorders: a scoping review

Abstract

With ageing populations around the world, there is a rapid rise in the number of people with Alzheimer’s disease (AD) and Parkinson’s disease (PD), the two most common types of neurodegenerative disorders. There is an urgent need to find new ways of aiding early diagnosis of these conditions. Multimodal learning of clinically accessible data is a relatively new approach that holds great potential to support early precise diagnosis. This scoping review follows the PRSIMA guidelines and we analysed 46 papers, comprising 11,750 participants, 3569 with AD, 978 with PD, and 2482 healthy controls; the recency of this topic was highlighted by nearly all papers being published in the last 5 years. It highlights the effectiveness of combining different types of data, such as brain scans, cognitive scores, speech and language, gait, hand and eye movements, and genetic assessments for the early detection of AD and PD. The review also outlines the AI methods and the model used in each study, which includes feature extraction, feature selection, feature fusion, and using multi-source discriminative features for classification. The review identifies knowledge gaps around the need to validate findings and address limitations such as small sample sizes. Applying multimodal learning of clinically accessible tests holds strong potential to aid the development of low-cost, reliable, and non-invasive methods for early detection of AD and PD.

Brief webcam test of hand movements predicts episodic memory, executive function, and working memory in a community sample of cognitively asymptomatic older adults

“…The use of computer vision methods to extract finger‐tapping features to predict cognitive performance is novel 3 , 5 , 16 and efficient 12 but similar methods are increasingly being used to quantify various movement disorders as they are objective and granular, unlike clinical rating scales, 44 and provide evaluation of movements in 3D space through simple smartphones or laptops. 14 , 15 , 45 Computer vision approaches facilitate wide reach and accessibility, including to those in rural and remote regions, as webcams are so ubiquitous in phones and computers around the world.…”

“…Alzheimer's disease (AD) accounts for more than 70% of cases, and the pathology begins 10–20 years before any overt cognitive symptoms emerge 2 . Developing simple, low‐cost tests to detect this preclinical stage is a research priority for drug development and enriching cohorts for dementia prevention 3 . However, we currently lack cost‐effective population‐level tests to risk stratify asymptomatic community samples; cognitive test performance is highly correlated with intelligence and education, and some people are more resilient to cognitive decline, carrying a higher burden of pathology for longer before cognitive test performance starts to decline 4 .…”

“… 2 Developing simple, low‐cost tests to detect this preclinical stage is a research priority for drug development and enriching cohorts for dementia prevention. 3 However, we currently lack cost‐effective population‐level tests to risk stratify asymptomatic community samples; cognitive test performance is highly correlated with intelligence and education, and some people are more resilient to cognitive decline, carrying a higher burden of pathology for longer before cognitive test performance starts to decline. 4 Positron emission tomography scans, cerebrospinal fluid (CSF) analysis, and blood‐based biomarkers are too costly, invasive, or specialist for widespread use.…”

Development of a smartphone screening test for preclinical Alzheimer’s disease and validation across the dementia continuum

“…Specifically, we plan to combine analysis of repetitive hand movements (finger-to-thumb tapping) with analysis of speech-like movements (through the DDK test) to provide a more sensitive and inclusive means for detecting AD pathology. This ‘multimodal’ approach will use digital video data for finger tapping analysis combined with audio data for DDK data and is expected to improve the accuracy of the test compared to using a single type of test; a scoping review of 46 studies and 11,750 participants found that, applying a multimodal approach improves the sensitivity and specificity of tests to detect AD and other neurodegenerative disorders [ 36 ]. In further support of this proposal, it is noteworthy that the hand and lips-tongue-palate areas in the cerebral cortex share a close topographical relationship, suggesting shared movement pathways [ 37 ].…”

miRNA‐137‐5p improves spatial memory and cognition in Alzheimer's mice by targeting ubiquitin‐specific peptidase 30

“…Over the past several decades, despite extensive research efforts, the early detection, effective treatment, and understanding of the molecular mechanisms underlying AD remain largely elusive. 4 , 5 Therefore, it is important to develop dependable diagnostic methods for the identification of AD biomarkers, particularly in the initial phases.…”