Robust and Interpretable AI-guided Marker for Early Dementia Prediction in Real-world Clinical Settings

Overview

Journal EClinicalMedicine

Specialty General Medicine

Date 2025 Jan 7

PMID 39764178

Authors

Liz Yuanxi Lee

Delshad Vaghari

Michael C Burkhart

Peter Tino

Marcella Montagnese

Zhuoyu Li

Katharina Zuhlsdorff

Joseph Giorgio

Guy Williams

Eddie Chong

Christopher Chen

Benjamin R Underwood

Timothy Rittman

Zoe Kourtzi

Affiliations

Soon will be listed here.

Abstract

Background: Predicting dementia early has major implications for clinical management and patient outcomes. Yet, we still lack sensitive tools for stratifying patients early, resulting in patients being undiagnosed or wrongly diagnosed. Despite rapid expansion in machine learning models for dementia prediction, limited model interpretability and generalizability impede translation to the clinic.

Methods: We build a robust and interpretable predictive prognostic model (PPM) and validate its clinical utility using real-world, routinely-collected, non-invasive, and low-cost (cognitive tests, structural MRI) patient data. To enhance scalability and generalizability to the clinic, we: 1) train the PPM with clinically-relevant predictors (cognitive tests, grey matter atrophy) that are common across research and clinical cohorts, 2) test PPM predictions with independent multicenter real-world data from memory clinics across countries (UK, Singapore).

Findings: PPM robustly predicts (accuracy: 81.66%, AUC: 0.84, sensitivity: 82.38%, specificity: 80.94%) whether patients at early disease stages (MCI) will remain stable or progress to Alzheimer's Disease (AD). PPM generalizes from research to real-world patient data across memory clinics and its predictions are validated against longitudinal clinical outcomes. PPM allows us to derive an individualized AI-guided multimodal marker (i.e. predictive prognostic index) that predicts progression to AD more precisely than standard clinical markers (grey matter atrophy, cognitive scores; PPM-derived marker: hazard ratio = 3.42, p = 0.01) or clinical diagnosis (PPM-derived marker: hazard ratio = 2.84, p < 0.01), reducing misdiagnosis.

Interpretation: Our results provide evidence for a robust and explainable clinical AI-guided marker for early dementia prediction that is validated against longitudinal, multicenter patient data across countries, and has strong potential for adoption in clinical practice.

Funding: Wellcome Trust, Royal Society, Alzheimer's Research UK, Alzheimer's Drug Discovery Foundation Diagnostics Accelerator, Alan Turing Institute.

References

Dubois B, Padovani A, Scheltens P, Rossi A, Dellagnello G . Timely Diagnosis for Alzheimer's Disease: A Literature Review on Benefits and Challenges. J Alzheimers Dis. 2015; 49(3):617-31. PMC: 4927869. DOI: 10.3233/JAD-150692. View

Myszczynska M, Ojamies P, Lacoste A, Neil D, Saffari A, Mead R . Applications of machine learning to diagnosis and treatment of neurodegenerative diseases. Nat Rev Neurol. 2020; 16(8):440-456. DOI: 10.1038/s41582-020-0377-8. View

Leming M, Das S, Im H . Adversarial confound regression and uncertainty measurements to classify heterogeneous clinical MRI in Mass General Brigham. PLoS One. 2023; 18(3):e0277572. PMC: 9980829. DOI: 10.1371/journal.pone.0277572. View

Schneider P, Biehl M, Hammer B . Adaptive relevance matrices in learning vector quantization. Neural Comput. 2009; 21(12):3532-61. DOI: 10.1162/neco.2009.11-08-908. View

Rathore S, Habes M, Iftikhar M, Shacklett A, Davatzikos C . A review on neuroimaging-based classification studies and associated feature extraction methods for Alzheimer's disease and its prodromal stages. Neuroimage. 2017; 155():530-548. PMC: 5511557. DOI: 10.1016/j.neuroimage.2017.03.057. View

. Estimation of the global prevalence of dementia in 2019 and forecasted prevalence in 2050: an analysis for the Global Burden of Disease Study 2019. Lancet Public Health. 2022; 7(2):e105-e125. PMC: 8810394. DOI: 10.1016/S2468-2667(21)00249-8. View

van Dyck C, Swanson C, Aisen P, Bateman R, Chen C, Gee M . Lecanemab in Early Alzheimer's Disease. N Engl J Med. 2022; 388(1):9-21. DOI: 10.1056/NEJMoa2212948. View

Garre-Olmo J . [Epidemiology of Alzheimer's disease and other dementias]. Rev Neurol. 2018; 66(11):377-386. View

Young A, Oxtoby N, Daga P, Cash D, Fox N, Ourselin S . A data-driven model of biomarker changes in sporadic Alzheimer's disease. Brain. 2014; 137(Pt 9):2564-77. PMC: 4132648. DOI: 10.1093/brain/awu176. View

10.

Jack Jr C, Wiste H, Vemuri P, Weigand S, Senjem M, Zeng G . Brain beta-amyloid measures and magnetic resonance imaging atrophy both predict time-to-progression from mild cognitive impairment to Alzheimer's disease. Brain. 2010; 133(11):3336-48. PMC: 2965425. DOI: 10.1093/brain/awq277. View

11.

Pellegrini E, Ballerini L, Valdes Hernandez M, Chappell F, Gonzalez-Castro V, Anblagan D . Machine learning of neuroimaging for assisted diagnosis of cognitive impairment and dementia: A systematic review. Alzheimers Dement (Amst). 2018; 10:519-535. PMC: 6197752. DOI: 10.1016/j.dadm.2018.07.004. View

12.

Topol E . High-performance medicine: the convergence of human and artificial intelligence. Nat Med. 2019; 25(1):44-56. DOI: 10.1038/s41591-018-0300-7. View

13.

Giorgio J, Jagust W, Baker S, Landau S, Tino P, Kourtzi Z . A robust and interpretable machine learning approach using multimodal biological data to predict future pathological tau accumulation. Nat Commun. 2022; 13(1):1887. PMC: 8989879. DOI: 10.1038/s41467-022-28795-7. View

14.

Cadar D, Lassale C, Davies H, Llewellyn D, Batty G, Steptoe A . Individual and Area-Based Socioeconomic Factors Associated With Dementia Incidence in England: Evidence From a 12-Year Follow-up in the English Longitudinal Study of Ageing. JAMA Psychiatry. 2018; 75(7):723-732. PMC: 6145673. DOI: 10.1001/jamapsychiatry.2018.1012. View

15.

Moguilner S, Whelan R, Adams H, Valcour V, Tagliazucchi E, Ibanez A . Visual deep learning of unprocessed neuroimaging characterises dementia subtypes and generalises across non-stereotypic samples. EBioMedicine. 2023; 90:104540. PMC: 10066533. DOI: 10.1016/j.ebiom.2023.104540. View

16.

Varoquaux G, Raamana P, Engemann D, Hoyos-Idrobo A, Schwartz Y, Thirion B . Assessing and tuning brain decoders: Cross-validation, caveats, and guidelines. Neuroimage. 2016; 145(Pt B):166-179. DOI: 10.1016/j.neuroimage.2016.10.038. View

17.

Jack Jr C, Bennett D, Blennow K, Carrillo M, Dunn B, Budd Haeberlein S . NIA-AA Research Framework: Toward a biological definition of Alzheimer's disease. Alzheimers Dement. 2018; 14(4):535-562. PMC: 5958625. DOI: 10.1016/j.jalz.2018.02.018. View

18.

Edmonds E, Delano-Wood L, Clark L, Jak A, Nation D, McDonald C . Susceptibility of the conventional criteria for mild cognitive impairment to false-positive diagnostic errors. Alzheimers Dement. 2014; 11(4):415-24. PMC: 4241187. DOI: 10.1016/j.jalz.2014.03.005. View

19.

Borchert R, Azevedo T, Badhwar A, Bernal J, Betts M, Bruffaerts R . Artificial intelligence for diagnostic and prognostic neuroimaging in dementia: A systematic review. Alzheimers Dement. 2023; 19(12):5885-5904. DOI: 10.1002/alz.13412. View

20.

Abdelnour C, Agosta F, Bozzali M, Fougere B, Iwata A, Nilforooshan R . Perspectives and challenges in patient stratification in Alzheimer's disease. Alzheimers Res Ther. 2022; 14(1):112. PMC: 9375274. DOI: 10.1186/s13195-022-01055-y. View