Recently, deep-learning-based approaches have been proposed for the classification of neuroimaging data related to Alzheimer’s disease (AD), and significant progress has been made. However, end-to-end learning that is capable of maximizing the impact of deep learning has yet to receive much attention due to the endemic challenge of neuroimaging caused by the scarcity of data. Thus, this study presents an approach meant to encourage the end-to-end learning of a volumetric convolutional neural network (CNN) model for four binary classification tasks (AD vs. normal control (NC), progressive mild cognitive impairment (pMCI) vs. NC, stable mild cognitive impairment (sMCI) vs. NC and pMCI vs. sMCI) based on magnetic resonance imaging (MRI) and visualizes its outcomes in terms of the decision of the CNNs without any human intervention. In the proposed approach, we use convolutional autoencoder (CAE)-based unsupervised learning for the AD vs. NC classification task, and supervised transfer learning is applied to solve the pMCI vs. sMCI classification task. To detect the most important biomarkers related to AD and pMCI, a gradient-based visualization method that approximates the spatial influence of the CNN model’s decision was applied. To validate the contributions of this study, we conducted experiments on the ADNI database, and the results demonstrated that the proposed approach achieved the accuracies of 86.60% and 73.95% for the AD and pMCI classification tasks respectively, outperforming other network models. In the visualization results, the temporal and parietal lobes were identified as key regions for classification.

IMPORTANCE Amyloid-β (Aβ), tau, and cerebral small vessel disease (CSVD), which occasionally coexist, are the most common causes of cognitive impairments in older people. However, whether tau is observed in patients with subcortical vascular cognitive impairment (SVCI), as well as its associations with Aβ and CSVD, are not yet established. More importantly, the role of tau underlying cognitive impairments in SVCI is unknown. OBJECTIVE To investigate the extent and the role of tau in patients with SVCI using 18 F-AV1451, which is a new ligand to detect neurofibrillary tangles in vivo.

The Antarctic marine environment is characterized by extreme seasonality in primary production, and herbivores must cope with a prolonged winter period of food shortage. In this study, tissue mass and biochemical composition were determined for various tissues of the bivalve Laternula elliptica (King & Broderip) over a 2 yr period, and its storage and use of energy reserves were investigated with respect to seasonal changes in food level and water temperature. Total ash-free dry mass (AFDM) accumulated rapidly following phytoplankton blooms (with peak values immediately before and after spawning) and was depleted considerably during the spawning and winter periods. Most of the variation was in the muscle, gonads and digestive gland. Spawning peaked in January and February and caused considerable protein and lipid losses in the muscle, gonads and digestive gland. In winter (March to August), the muscle and digestive gland lost considerable mass, while gonad mass increased; this suggests that the muscle tissue and digestive gland serve as major energy depots for both maintenance metabolism and gonad development in winter. There were also marked year-to-year differences in the seasonal patterns of mass variation and reproduction. Overall, the relative and absolute tissue-mass values were positively correlated with chlorophyll concentration, and were not related to water temperature; thus, for the first time, this study clearly shows that food is an important factor governing growth and gonad maturation in this bivalve. It is also noteworthy that protein, constituting ~75% of AFDM, served as the major energy reserve throughout the study, closely following the AFDM variation. In particular, during the winter months, protein comprised >60% of AFDM loss, while lipids and glycogen served as minor (< 20% each) reserves. Protein loss was most substantial in the muscle tissue, which comprised half of the body tissue. Thus, protein use, with muscle tissues as a depot for protein reserves, may be a result of selective pressure on Antarctic marine herbivores undergoing a prolonged period of food shortage in winter.