Alzheimer's disease diagnosis support for brain perfusion SPECT scans in a real-world clinical cohort
Alzheimer's disease diagnosis support for brain perfusion SPECT scans in a real-world clinical cohort
Background: dementia diagnosis is challenging and often delayed. Brain imaging techniques such as single-photon emission computed tomography (SPECT) imaging can help identify subtle changes in brain perfusion. Artificial intelligence methods may support results interpretation for early diagnosis.
Objective: to develop and validate multivariate models for the early diagnosis of Alzheimer's disease (AD), using brain perfusion SPECT imaging and interpretable artificial intelligence methods in a real-world clinical setting.
Methods: two logistic regression models were developed using a training dataset of 420 SPECT scans and tested on an independent clinical dataset of 443 scans. Model 1 was designed to identify abnormal perfusion patterns, while Model 2 identified perfusion changes associated with AD. Input features were extracted from anatomical volumes of interest, with feature selection performed using the Minimum Redundancy Maximum Relevance (MRMR) algorithm.
Results: the models demonstrated good classification performance using real-world clinical data. Model 1 achieved an area under receiver operator characteristic (AUROC) Curve of 0.89 (Sensitivity 76%, Specificity 87%) in identifying abnormal brain perfusion. Model 2 achieved an AUROC of 0.86 (Sensitivity 87%, Specificity 72%) in identifying AD.
Conclusions: multivariate logistic regression models trained on real-world clinical data show promise as clinical decision support tools for the diagnosis of AD from brain perfusion SPECT imaging. The models use features from clinically relevant brain regions, which enhances interpretability. Future research should focus on expanding model applicability to other dementia types and on prospective evaluation of their utility in improving diagnostic accuracy, consistency, and care pathways in diverse clinical environments.
Alzheimer's disease, SPECT imaging, artificial intelligence, brain perfusion, clinical decision support, real-world data
192-200
Michopoulou, Sofia
f21ba2a3-f5d3-4998-801f-1ae72ff5d92c
Prosser, Angus
de1efee5-67f5-478e-8cfa-12a8e78a68e5
O’Brien, Neil
d946b58d-b9d9-45f2-b0e5-b703421a0ac2
Dickson, John
627f7f54-97e9-4cc1-812c-728c3973265d
Guy, Matthew
1a40b2ed-3aec-4fce-9954-396840471c28
Teeling, Jessica L.
fcde1c8e-e5f8-4747-9f3a-6bdb5cd87d0a
Kipps, Christopher M.
e43be016-2dc2-45e6-9a02-ab2a0e0208d5
March 2026
Michopoulou, Sofia
f21ba2a3-f5d3-4998-801f-1ae72ff5d92c
Prosser, Angus
de1efee5-67f5-478e-8cfa-12a8e78a68e5
O’Brien, Neil
d946b58d-b9d9-45f2-b0e5-b703421a0ac2
Dickson, John
627f7f54-97e9-4cc1-812c-728c3973265d
Guy, Matthew
1a40b2ed-3aec-4fce-9954-396840471c28
Teeling, Jessica L.
fcde1c8e-e5f8-4747-9f3a-6bdb5cd87d0a
Kipps, Christopher M.
e43be016-2dc2-45e6-9a02-ab2a0e0208d5
Michopoulou, Sofia, Prosser, Angus, O’Brien, Neil, Dickson, John, Guy, Matthew, Teeling, Jessica L. and Kipps, Christopher M.
(2026)
Alzheimer's disease diagnosis support for brain perfusion SPECT scans in a real-world clinical cohort.
Journal of Alzheimer’s Disease, 110 (1), .
(doi:10.1177/13872877251413790).
Abstract
Background: dementia diagnosis is challenging and often delayed. Brain imaging techniques such as single-photon emission computed tomography (SPECT) imaging can help identify subtle changes in brain perfusion. Artificial intelligence methods may support results interpretation for early diagnosis.
Objective: to develop and validate multivariate models for the early diagnosis of Alzheimer's disease (AD), using brain perfusion SPECT imaging and interpretable artificial intelligence methods in a real-world clinical setting.
Methods: two logistic regression models were developed using a training dataset of 420 SPECT scans and tested on an independent clinical dataset of 443 scans. Model 1 was designed to identify abnormal perfusion patterns, while Model 2 identified perfusion changes associated with AD. Input features were extracted from anatomical volumes of interest, with feature selection performed using the Minimum Redundancy Maximum Relevance (MRMR) algorithm.
Results: the models demonstrated good classification performance using real-world clinical data. Model 1 achieved an area under receiver operator characteristic (AUROC) Curve of 0.89 (Sensitivity 76%, Specificity 87%) in identifying abnormal brain perfusion. Model 2 achieved an AUROC of 0.86 (Sensitivity 87%, Specificity 72%) in identifying AD.
Conclusions: multivariate logistic regression models trained on real-world clinical data show promise as clinical decision support tools for the diagnosis of AD from brain perfusion SPECT imaging. The models use features from clinically relevant brain regions, which enhances interpretability. Future research should focus on expanding model applicability to other dementia types and on prospective evaluation of their utility in improving diagnostic accuracy, consistency, and care pathways in diverse clinical environments.
Text
michopoulou-et-al-2026-alzheimer-s-disease-diagnosis-support-for-brain-perfusion-spect-scans-in-a-real-world-clinical
- Version of Record
More information
Accepted/In Press date: 12 December 2025
e-pub ahead of print date: 30 January 2026
Published date: March 2026
Keywords:
Alzheimer's disease, SPECT imaging, artificial intelligence, brain perfusion, clinical decision support, real-world data
Identifiers
Local EPrints ID: 511591
URI: http://eprints.soton.ac.uk/id/eprint/511591
PURE UUID: 678f26e6-f2fd-4185-9577-ad99e421a977
Catalogue record
Date deposited: 22 May 2026 16:32
Last modified: 23 May 2026 02:14
Export record
Altmetrics
Contributors
Author:
Sofia Michopoulou
Author:
Neil O’Brien
Author:
John Dickson
Author:
Matthew Guy
Author:
Christopher M. Kipps
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics