Deep learning improves utility of tau PET in the study of Alzheimer's disease

Abstract

Introduction Positron emission tomography (PET) imaging targeting neurofibrillary tau tangles is increasingly used in the study of Alzheimer's disease (AD), but its utility may be limited by conventional quantitative or qualitative evaluation techniques in earlier disease states. Convolutional neural networks (CNNs) are effective in learning spatial patterns for image classification. Methods 18F‐MK6240 (n = 320) and AV‐1451 (n = 446) PET images were pooled from multiple studies. We performed iterations with differing permutations of radioligands, heuristics, and architectures. Performance was compared to a standard region of interest (ROI)‐based approach on prediction of memory impairment. We visualized attention of the network to illustrate decision making. Results Overall, models had high accuracy (> 80%) with good average sensitivity and specificity (75% and 82%, respectively), and had comparable or higher accuracy to the ROI standard. Visualizations of model attention highlight known characteristics of tau radioligand binding. Discussion CNNs could improve tau PET's role in early disease and extend the utility of tau PET across generations of radioligands.

Deep learning application for the classification of Alzheimer’s disease using 18F-flortaucipir (AV-1451) tau positron emission tomography

“…In other words, it is difficult to classify tau deposition measurements as representative of cognitive decline including MCI and AD compared to CU through the ROIs of stage III/IV. We performed a systematic review of the existing literature to summarize the most common CU versus AD classification techniques that include comparison of CU versus MCI (Table 3 ) 17 , 18 , 22 – 27 . Notably, the classification between CU and MCI in this study showed better performance than other previously published methods.…”

Advancing Tau-PET quantification in Alzheimer's disease with machine learning: introducing THETA, a novel tau summary measure

“…The application of deep learning and ML using Tau-PET has become common in recent years, either to improve PET image acquisition 13 , to classify spatial patterns 14,15 , to study the association between Aβ and Tau-PET scans 16 , or to predict pathological tau accumulation from clinical measures 17,18 . ML-based indices have also been introduced such as Spatial Pattern of Abnormality for Recognition of Early Tauopathy (SPARE-Tau) 19 and Alzheimer's disease resemblance atrophy index (AD-RAI) 20 .…”

Improved interpretation of 18F-florzolotau PET in progressive supranuclear palsy using a normalization-free deep-learning classifier

“…While a growing number of DL-based clinical applications have been proposed for PET imaging, 15 their use in tau PET imaging has been limited to 18 F-flortaucipir and 18 F-MK-6240 for AD. 16 , 17 , 18 To the best of our knowledge, no studies have investigated the performance of DL algorithms using raw PET images without spatial and intensity normalization as input. In the context of PSP, where elevated tracer uptake mainly occurs in small, deep brain structures such as the subcortical nucleus, Endo et al.…”