Causal effect estimation on a graph of connected units is often complicated by entangled treatments, where the treatment assignment is not independent for each individual. This presents multiple challenges: accurately modeling treatment assignment mechanisms, adjusting for both observed and unobserved confounders to mitigate confounding bias, and constructing instrumental variables to adjust unobserved confounders within a graph structure. Prior research on estimating the causal effects of entangled treatments either assumed no unobserved confounders or relied on the manual selection of IVs, leading to gaps in the methodology. To bridge these gaps and build upon previous work, we introduce the Graph-Disentanglement Instrumental Variable (GDIV) model, a novel approach employing both Graph Neural Networks (GNNs) and Adversarial Networks to assess the causal effects on nodes in a graph, considering observed/unobserved confounders and the intricacies of treatment entanglement. Our GDIV estimator is validated through extensive experiments across synthetic and semisynthetic datasets, demonstrating its better performance over state-of-the-art methods. The ablation studies and robustness experiments verify the benefits of leveraging adversarial networks to generate IVs that satisfy the required assumptions.

Cancer has been characterized as a heterogeneous disease, and the classification of cancer subtypes has become a necessity in cancer research, as it can facilitate the subsequent clinical management of patients and provide clinical decision support for clinicians. With the advance of machine learning in the last decade, many researchers employ machine learning to tackle the cancer classification problem. Importantly, traditional machine learning algorithms require a large amount of annotated data for model training. However, collection of large amounts of annotated data is time-consuming and expensive and may not be realistic in real-world activities. Facing data scarcity, meta-learning is proposed to tackle this problem. Meta-learning utilizes prior knowledge learned from related tasks and generalizes to new tasks of limited supervised experience, and it has been applied in many fields to tackle scarce annotated data problem, such as few-shot image classification, drug discovery, etc. As data scarcity is common in cancer research and diagnosis studies, and there are only few previous studies that classify cancers based on limited annotated data. We explore the meta-learning algorithm (MAML) to tackle the scenario where only limited annotated data are available. In this work, our objective is to comprehensively compare MAML among few-shot learning methods (matching network and prototypical network) and traditional machine learning methods (random forest and K-nearest neighbor). Experimental results on The Cancer Genome Atlas (TCGA) cancer patient data demonstrates the effectiveness and superiority of MAML over other methods, including its ability to outperform the other methods using 4.5-fold fewer features.

As an important family of learning problems in healthcare domain, patient similarity learning has received much attention in recent years. Patient similarity learning aims to measure the similarity between a pair of patients according to their historical clinical information, which helps to improve the clinical predictions of the patient of interest. Although patient similarity learning has achieved tremendous success in many real-world applications, the lack of transparency behind the behavior of the learned patient similarity model impedes users from trusting the predicted results, which hampers its further applications in the real world. To tackle this problem, in this paper, we investigate how to enable interpretation in patient similarity learning and propose a global interpretation method for patient similarity learning. Based on the proposed global interpretation method, we can identify a minimal sufficient subset of data features that are sufficient in themselves to justify the global predictions made by the well-trained patient similarity model. The identified minimal sufficient feature subset can help us to better understand the overall behaviors of the learned model across different subpopulations of patients. We also conduct experiments on real-world datasets to evaluate the performance of the proposed global interpretation method.

Annotated data samples in real-world applications are often limited. Meta-learning, which utilizes prior knowledge learned from related tasks and generalizes to new tasks of limited supervised experience, is an effective approach for few-shot learning. However, standard meta-learning with globally shared knowledge cannot handle the task heterogeneity problem well, i.e., tasks lie in different distributions. Recent advances have explored several ways to trigger task-dependent initial parameters or metrics, in order to customize task-specific information. These approaches learn task contextual information from data, but ignore external domain knowledge that can help in the learning process. In this paper, we propose a task-adaptive network (TAdaNet) that makes use of a domain-knowledge graph to enrich data representations and provide task-specific customization. Specifically, we learn a task embedding that characterizes task relationships and tailors task-specific parameters, resulting in a task-adaptive metric space for classification. Experimental results on a few-shot image classification problem show the effectiveness of the proposed method. We also apply it on a real-world disease classification problem, and show promising results for clinical decision support.

With the arrival of the big data era, more and more data are becoming readily available in various real-world applications and those data are usually highly heterogeneous. Taking computational medicine as an example, we have both Electronic Health Records (EHR) and medical images for each patient. For complicated diseases such as Parkinson's and Alzheimer's, both EHR and neuroimaging information are very important for disease understanding because they contain complementary aspects of the disease. However, EHR and neuroimage are completely different. So far the existing research has been mainly focusing on one of them. In this paper, we proposed a framework, Memory-Based Graph Convolution Network (MemGCN), to perform integrative analysis with such multi-modal data. Specifically, GCN is used to extract useful information from the patients' neuroimages. The information contained in the patient EHRs before the acquisition of each brain image is captured by a memory network because of its sequential nature. The information contained in each brain image is combined with the information read out from the memory network to infer the disease state at the image acquisition timestamp. To further enhance the analytical power of MemGCN, we also designed a multi-hop strategy that allows multiple reading and updating on the memory can be performed at each iteration. We conduct experiments using the patient data from the Parkinson's Progression Markers Initiative (PPMI) with the task of classification of Parkinson's Disease (PD) cases versus controls. We demonstrate that superior classification performance can be achieved with our proposed framework, compared with existing approaches involving a single type of data.