Search for: All records

Recent deep clustering algorithms take advantage of self-supervised learning and self-training techniques to map the original data into a latent space, where the data embedding and clustering assignment can be jointly optimized. However, as many recent datasets are enormous and noisy, getting a clear boundary between different clusters is challenging with existing methods that mainly focus on contracting similar samples together and overlooking samples near boundary of clusters in the latent space. In this regard, we propose an end-to-end deep clustering algorithm, i.e., Locally Normalized Soft Contrastive Clustering (LNSCC). It takes advantage of similarities among each sample’s local neighborhood and globally disconnected samples to leverage positiveness and negativeness of sample pairs in a contrastive way to separate different clusters. Experimental results on various datasets illustrate that our proposed approach achieves outstanding clustering performance over most of the state-of-the-art clustering methods for both image and non-image data even without convolution. 

The emergence of Positron Emission Tomography (PET) imaging allows us to quantify the burden of amyloid plaques in-vivo, which is one of the hallmarks of Alzheimer’s disease (AD). However, the invasive exposure to radiation and high imaging cost significantly restrict the application of PET in characterizing the evolution of pathology burden which often requires longitudinal PET image sequences. In this regard, we propose a proof-of-concept solution to generate the complete trajectory of pathological events throughout the brain based on very limited number of PET scans. We present a novel variational autoencoder model to learn a latent population-level representation of neurodegeneration process based on the longitudinal β-amyloid measurements at each brain region and longitudinal diagnostic stages. As the propagation of pathological burdens follow the topology of brain connectome, we further cast our neural network into a supervised sequential graph VAE, where we use the brain network to guide the representation learning. Experiments show that the disentangled representation can capture disease-related dynamics of amyloid and forecast the level of amyloid depositions at future time points. 

Abstract The DAMA/LIBRA collaboration has reported the observation of an annual modulation in the event rate that has been attributed to dark matter interactions over the last two decades. However, even though tremendous efforts to detect similar dark matter interactions were pursued, no definitive evidence has been observed to corroborate the DAMA/LIBRA signal. Many studies assuming various dark matter models have attempted to reconcile DAMA/LIBRA’s modulation signals and null results from other experiments, however no clear conclusion can be drawn. Apart from the dark matter hypothesis, several studies have examined the possibility that the modulation is induced by variations in detector’s environment or their specific analysis methods. In particular, a recent study presents a possible cause of the annual modulation from an analysis method adopted by the DAMA/LIBRA experiment in which the observed annual modulation could be reproduced by a slowly varying time-dependent background. Here, we study the COSINE-100 data using an analysis method similar to the one adopted by the DAMA/LIBRA experiment and observe a significant annual modulation, however the modulation phase is almost opposite to that of the DAMA/LIBRA data. Assuming the same background composition for COSINE-100 and DAMA/LIBRA, simulated experiments for the DAMA/LIBRA without dark matter signals also provide significant annual modulation with an amplitude similar to DAMA/LIBRA with opposite phase. Even though this observation does not directly explain the DAMA/LIBRA results directly, this interesting phenomenon motivates more profound studies of the time-dependent DAMA/LIBRA background data. 

Brain development in adolescence is synthetically influenced by various factors such as age, education, and socioeconomic conditions. To identify an independent effect from a variable of interest (e.g., socioeconomic conditions), statistical models such as General Linear Model (GLM) are typically adopted to account for covariates (e.g., age and gender). However, statistical models may be vulnerable with insufficient sample size and outliers, and multiple tests for a whole brain analysis lead to inevitable false-positives without sufficient sensitivity. Hence, it is necessary to develop a unified framework for multiple tests that robustly fits the observation and increases sensitivity. We therefore propose a unified flexible neural network that optimizes on the contribution from the main variable of interest as introduced in original GLM, which leads to improved statistical outcomes. The results on group analysis with fractional anisotropy (FA) from Diffusion Tensor Images from Adolescent Brain Cognitive Development (ABCD) study demonstrate that the proposed method provides much more selective and meaningful detection of ROIs related to socioeconomic status over conventional methods. 

Given a population longitudinal neuroimaging measurements defined on a brain network, exploiting temporal dependencies within the sequence of data and corresponding latent variables defined on the graph (i.e., network encoding relationships between regions of interest (ROI)) can highly benefit characterizing the brain. Here, it is important to distinguish time-variant (e.g., longitudinal measures) and time-invariant (e.g., gender) components to analyze them individually. For this, we propose an innovative and ground-breaking Disentangled Sequential Graph Autoencoder which leverages the Sequential Variational Autoencoder (SVAE), graph convolution and semi-supervising framework together to learn a latent space composed of time-variant and time-invariant latent variables to characterize disentangled representation of the measurements over the entire ROIs. Incorporating target information in the decoder with a supervised loss let us achieve more effective representation learning towards improved classification. We validate our proposed method on the longitudinal cortical thickness data from Alzheimer’s Disease Neuroimaging Initiative (ADNI) study. Our method outperforms baselines with traditional techniques demonstrating benefits for effective longitudinal data representation for predicting labels and longitudinal data generation. 

Tremendous recent literature show that associations between different brain regions, i.e., brain connectivity, provide early symptoms of neurological disorders. Despite significant efforts made for graph neural network (GNN) techniques, their focus on graph nodes makes the state-of-the-art GNN methods not suitable for classifying brain connectivity as graphs where the objective is to characterize disease-relevant network dysfunction patterns on graph links. To address this issue, we propose Multi-resolution Edge Network (MENET) to detect disease-specific connectomic benchmarks with high discrimination power across diagnostic categories. The core of MENET is a novel graph edge-wise transform that we propose, which allows us to capture multi-resolution “connectomic” features. Using a rich set of the connectomic features, we devise a graph learning framework to jointly select discriminative edges and assign diagnostic labels for graphs. Experiments on two real datasets show that MENET accurately predicts diagnostic labels and identify brain connectivities highly associated with neurological disorders such as Alzheimer’s Disease and Attention-Deficit/Hyperactivity Disorder. 

Weakly labeled data are inevitable in various research areas in artificial intelligence (AI) where one has a modicum of knowledge about the complete dataset. One of the reasons for weakly labeled data in AI is insufficient accurately labeled data. Strict privacy control or accidental loss may also cause missing-data problems. However, supervised machine learning (ML) requires accurately labeled data in order to successfully solve a problem. Data labeling is difficult and time-consuming as it requires manual work, perfect results, and sometimes human experts to be involved (e.g., medical labeled data). In contrast, unlabeled data are inexpensive and easily available. Due to there not being enough labeled training data, researchers sometimes only obtain one or few data points per category or label. Training a supervised ML model from the small set of labeled data is a challenging task. The objective of this research is to recover missing labels from the dataset using state-of-the-art ML techniques using a semisupervised ML approach. In this work, a novel convolutional neural network-based framework is trained with a few instances of a class to perform metric learning. The dataset is then converted into a graph signal, which is recovered using a recover algorithm (RA) in graph Fourier transform. The proposed approach was evaluated on a Fashion dataset for accuracy and precision and performed significantly better than graph neural networks and other state-of-the-art methods