ORCID Profile
0000-0002-2520-454X
Current Organisation
University of Cambridge
Does something not look right? The information on this page has been harvested from data sources that may not be up to date. We continue to work with information providers to improve coverage and quality. To report an issue, use the Feedback Form.
Publisher: BMJ
Date: 30-07-2020
Publisher: Cold Spring Harbor Laboratory
Date: 16-07-2023
DOI: 10.1101/2023.07.14.23292648
Abstract: PREVENT is a multi-centre prospective cohort study in the UK and Ireland that aims to examine mid-life risk factors for dementia, identify and describe the earliest indices of disease development. The PREVENT dementia programme is one of the original epidemiological initiatives targeting midlife as a critical window for intervention in neurodegenerative conditions. This paper provides an overview of the study protocol and presents the first summary results from the initial baseline data to describe the cohort. Participants in the PREVENT cohort provide demographic data, biological s les (blood, saliva, urine and optional cerebrospinal fluid), lifestyle and psychological questionnaires, undergo a comprehensive cognitive test battery, and are imaged using multi-modal 3T magnetic resonance imaging (MRI) scanning, with both structural and functional sequences. The PREVENT cohort governance structure is described, which includes a steering committee, a scientific advisory board and core patient and public involvement groups. A number of sub-studies which supplement the main PREVENT cohort are also described. The PREVENT cohort baseline data includes 700 participants recruited between 2014 and 2020 across five sites in the UK and Ireland (Cambridge, Dublin, Edinburgh, London and Oxford). At baseline, participants had a mean age of 51.2 years (range 40-59, SD ±5.47), with the majority female (n=433, 61.9%). There was a near equal distribution of participants with and without a parental history of dementia (51.4% vs 48.6%) and a relatively high prevalence of APOE⍰4 carriers (n=264, 38.0%). Participants were highly educated (16.7 ± 3.44 years of education), were mainly of European Ancestry (n=672, 95.9%) and were cognitively healthy as measured by the Addenbrookes Cognitive Examination-III (ACE-III) (Total score 95.6 ±4.06). Mean white matter hyperintensity (WMH) volume at recruitment was 2.26 ± 2.77 ml (median = 1.39ml), with hippoc al volume 8.15 ± 0.79ml. There was good representation of known dementia risk factors in the cohort. The PREVENT cohort offers a novel dataset to explore midlife risk factors and early signs of neurodegenerative disease. Data are available open access at no cost via the Alzheimer’s Disease Data Initiative (ADDI) platform and Dementia Platforms UK (DPUK) platform pending approval of the data access request from the PREVENT steering group committee.
Publisher: Wiley
Date: 10-08-2023
DOI: 10.1002/ALZ.13412
Abstract: Artificial intelligence (AI) and neuroimaging offer new opportunities for diagnosis and prognosis of dementia. We systematically reviewed studies reporting AI for neuroimaging in diagnosis and/or prognosis of cognitive neurodegenerative diseases. A total of 255 studies were identified. Most studies relied on the Alzheimer's Disease Neuroimaging Initiative dataset. Algorithmic classifiers were the most commonly used AI method (48%) and discriminative models performed best for differentiating Alzheimer's disease from controls. The accuracy of algorithms varied with the patient cohort, imaging modalities, and stratifiers used. Few studies performed validation in an independent cohort. The literature has several methodological limitations including lack of sufficient algorithm development descriptions and standard definitions. We make recommendations to improve model validation including addressing key clinical questions, providing sufficient description of AI methods and validating findings in independent datasets. Collaborative approaches between experts in AI and medicine will help achieve the promising potential of AI tools in practice. There has been a rapid expansion in the use of machine learning for diagnosis and prognosis in neurodegenerative disease Most studies (71%) relied on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset with no other in idual dataset used more than five times There has been a recent rise in the use of more complex discriminative models (e.g., neural networks) that performed better than other classifiers for classification of AD vs healthy controls We make recommendations to address methodological considerations, addressing key clinical questions, and validation We also make recommendations for the field more broadly to standardize outcome measures, address gaps in the literature, and monitor sources of bias
Location: United Kingdom of Great Britain and Northern Ireland
No related grants have been discovered for Audrey Low.