Federated Morphometry Feature Selection for Hippocampal Morphometry Associated Beta-Amyloid and Tau Pathology

Abstract

Amyloid-β (Aβ) plaques and tau protein tangles in the brain are now widely recognized as the defining hallmarks of Alzheimer's disease (AD), followed by structural atrophy detectable on brain magnetic resonance imaging (MRI) scans. One of the particular neurodegenerative regions is the hippocampus to which the influence of Aβ/tau on has been one of the research focuses in the AD pathophysiological progress. This work proposes a novel framework, Federated Morphometry Feature Selection (FMFS) model, to examine subtle aspects of hippocampal morphometry that are associated with Aβ/tau burden in the brain, measured using positron emission tomography (PET). FMFS is comprised of hippocampal surface-based feature calculation, patch-based feature selection, federated group LASSO regression, federated screening rule-based stability selection, and region of interest (ROI) identification. FMFS was tested on two ADNI cohorts to understand hippocampal alterations that relate to Aβ/tau depositions. Each cohort included pairs of MRI and PET for AD, mild cognitive impairment (MCI) and cognitively unimpaired (CU) subjects. Experimental results demonstrated that FMFS achieves an 89x speedup compared to other published state-of-the-art methods under five independent hypothetical institutions. In addition, the subiculum and cornu ammonis 1 (CA1 subfield) were identified as hippocampal subregions where atrophy is strongly associated with abnormal Aβ/tau. As potential biomarkers for Aβ/tau pathology, the features from the identified ROIs had greater power for predicting cognitive assessment and for survival analysis than five other imaging biomarkers. All the results indicate that FMFS is an efficient and effective tool to reveal associations between Aβ/tau burden and hippocampal morphometry.

Demystifying Swarm Learning: A New Paradigm of Blockchain-based Decentralized Federated Learning

“…As far as we are concerned, there are few empirical studies on SL or blockchain-based decentralized FL paradigm, therefore, we conduct this measurement to fill the knowledge gap between SL deployment and developers, and provide practical suggestions to developers and researchers. As for applications, researchers utilize SL to help diagnosis of COVID-19, tuberculosis, leukaemia and lung pathologies [40], feature selection for beta-amyloid and TAU pathology [42], the Internet of Vehicles [37], skin lesion classification fairness [5], genomics data sharing [28], and risk prediction of cardiovascular events [41].…”

“…Apart from Swarm edge nodes, there are Swarm coordinator nodes responsible for maintaining metadata like the model state, training progress, and licenses, without model parameters. Despite the increasing attention on the SL paradigm both in industries and academics [5,28,37,41,42], there lacks comprehensive knowledge of best practices and precautions of deploying SL in real-world scenarios. Although the Swarm Learning Library (SLL) is open-sourced in binary format for non-commercial use 1 and has been followed by many researchers and developers, numerous issues about deployment have been raised, such as adaption on different datasets or data distributions, encapsulated features, license assignment, connectivity, heterogeneity in operation systems, hardware infrastructures, DL platforms, and models.…”

Integrating Transcriptomics, Genomics, and Imaging in Alzheimer’s Disease: A Federated Model

“…In recent work (Wang et al, 2021;Wu et al, 2021), the authors created tools to generate a univariate morphometry index (UMI) for surface morphometry features on regions of interest (ROIs) that are related to beta-amyloid deposition. This induced UMI may reflect intrinsic morphological changes induced by processes of amyloid accumulation in AD.…”