Visual deep learning of unprocessed neuroimaging characterises dementia subtypes and generalises across non-stereotypic samples

Sebastian Moguilner, Robert Whelan, Hieab Adams, Victor Valcour, Enzo Tagliazucchi, Agustín Ibáñez*

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

21 Citations (Scopus)
49 Downloads (Pure)

Abstract

Background: Dementia's diagnostic protocols are mostly based on standardised neuroimaging data collected in the Global North from homogeneous samples. In other non-stereotypical samples (participants with diverse admixture, genetics, demographics, MRI signals, or cultural origins), classifications of disease are difficult due to demographic and region-specific sample heterogeneities, lower quality scanners, and non-harmonised pipelines. Methods: We implemented a fully automatic computer-vision classifier using deep learning neural networks. A DenseNet was applied on raw (unpreprocessed) data from 3000 participants (behavioural variant frontotemporal dementia-bvFTD, Alzheimer's disease-AD, and healthy controls; both male and female as self-reported by participants). We tested our results in demographically matched and unmatched samples to discard possible biases and performed multiple out-of-sample validations. Findings: Robust classification results across all groups were achieved from standardised 3T neuroimaging data from the Global North, which also generalised to standardised 3T neuroimaging data from Latin America. Moreover, DenseNet also generalised to non-standardised, routine 1.5T clinical images from Latin America. These generalisations were robust in samples with heterogenous MRI recordings and were not confounded by demographics (i.e., were robust in both matched and unmatched samples, and when incorporating demographic variables in a multifeatured model). Model interpretability analysis using occlusion sensitivity evidenced core pathophysiological regions for each disease (mainly the hippocampus in AD, and the insula in bvFTD) demonstrating biological specificity and plausibility. Interpretation: The generalisable approach outlined here could be used in the future to aid clinician decision-making in diverse samples. Funding: The specific funding of this article is provided in the acknowledgements section.

Original languageEnglish
Article number104540
JournalEBioMedicine
Volume90
DOIs
Publication statusPublished - Apr 2023

Bibliographical note

Acknowledgments:
S.M. is an Atlantic Fellow for Equity in Brain Health at the Global Brain Health Institute (GBHI) and is supported with funding from GBHI, Alzheimer's Association, and Alzheimer's Society (GBHI ALZ UK-21-721776). A.I. is partially supported by grants of Takeda CW2680521; CONICET; FONCYT-PICT (2017-1818, 2017-1820); ANID/FONDECYT Regular (1210195, 1210176, 1220995); ANID/FONDAP (15150012); ANID/FONDEF (ID20I10152 and ID22I10029), ANID/PIA/ANILLOS ACT210096; and the Multi-Partner Consortium to Expand Dementia Research in Latin America (ReDLat), funded by the National Institutes of Aging of the National Institutes of Health under award number R01AG057234, an Alzheimer's Association grant (SG-20-725707-ReDLat), the Rainwater Foundation, and the GBHI.

Publisher Copyright: © 2023 The Author(s)

Fingerprint

Dive into the research topics of 'Visual deep learning of unprocessed neuroimaging characterises dementia subtypes and generalises across non-stereotypic samples'. Together they form a unique fingerprint.

Cite this