Generalizable, Reproducible, and Neuroscientifically Interpretable Imaging Biomarkers for Alzheimer's Disease

Abstract

Precision medicine for Alzheimer's disease (AD) necessitates the development of personalized, reproducible, and neuroscientifically interpretable biomarkers, yet despite remarkable advances, few such biomarkers are available. Also, a comprehensive evaluation of the neurobiological basis and generalizability of the end‐to‐end machine learning system should be given the highest priority. For this reason, a deep learning model (3D attention network, 3DAN) that can simultaneously capture candidate imaging biomarkers with an attention mechanism module and advance the diagnosis of AD based on structural magnetic resonance imaging is proposed. The generalizability and reproducibility are evaluated using cross‐validation on in‐house, multicenter ( n = 716), and public ( n = 1116) databases with an accuracy up to 92%. Significant associations between the classification output and clinical characteristics of AD and mild cognitive impairment (MCI, a middle stage of dementia) groups provide solid neurobiological support for the 3DAN model. The effectiveness of the 3DAN model is further validated by its good performance in predicting the MCI subjects who progress to AD with an accuracy of 72%. Collectively, the findings highlight the potential for structural brain imaging to provide a generalizable, and neuroscientifically interpretable imaging biomarker that can support clinicians in the early diagnosis of AD.

Regional radiomics similarity networks (R2SNs) in the human brain: Reproducibility, small-world properties and a biological basis

“…These results further suggested that gene expression can be reflected by R2SNs. In brief, R2SNs have a genetic basis, and the present findings provide the possibility of estimating the risk of a gene by using R2SNs and indicate a certain degree of evidence for genetic disease research, such as that conducted regarding AD (Jin et al, 2020b;Zhao et al, 2020).…”

Cortical and Subcortical Grey Matter Abnormalities in White Matter Hyperintensities and Subsequent Cognitive Impairment

“…First, the sample size of the in-house dataset was relatively small, particularly for the WMH-MCI group, and the sizes of the WMH-MCI and WMH-nCI groups were unbalanced due to the study design, which may lead to lower statistical power. Based on studies focused on the neuroimaging features of AD [ 71 , 72 ], the characteristic GM alteration pattern in WMH patients with MCI needs further cross-validation by large-scale multi-center studies in the future. Second, we did not conduct a comprehensive evaluation of motor and mood disturbances which are common in patients with WMH.…”

Application of Structural and Functional Connectome Mismatch for Classification and Individualized Therapy in Alzheimer Disease

“…The development of personalized, reproducible, non-invasive, and neuroscientifically interpretable biomarkers are urgently needed for AD precision medicine (16,42), yet despite remarkable advances, few such biomarkers are available. Neuroimaging using DTI and fMRI in conjunction provides objective information on the structure and function that for assessing network connectivity of the brain.…”

“…Machine learning is an application of artificial intelligence that allows computers to learn automatically and improve from experience. It is one of today's most rapidly growing technical fields (13), which performs throughout science including health care (14) such as identification and classification for diseases like AD (15)(16)(17), traffic programming (18), and marketing apps designing (19), which allows us to process largescale, multidimensional, complex datasets in this information explosion of an era. Machine learning-based analysis of connectomic data created from neuroimaging studies in patients AD has been extensively studied in the literature (5,9,12,20,21).…”