RetainVis: Visual Analytics with Interpretable and Interactive Recurrent Neural Networks on Electronic Medical Records

A Visual Analytics Conceptual Framework for Explorable and Steerable Partial Dependence Analysis

Machine learning techniques are a driving force for research in various fields, from credit card fraud detection to stock analysis. Recently, a growing interest in increasing human involvement has emerged, with the primary goal of improving the interpretability of machine learning models. Among different techniques, Partial Dependence Plots (PDP) represent one of the main model-agnostic approaches for interpreting how the features influence the prediction of a machine learning model. However, its limitations (i.e., visual interpretation, aggregation of heterogeneous effects, inaccuracy, and computability) could complicate or misdirect the analysis. Moreover, the resulting combinatorial space can be challenging to explore both computationally and cognitively when analyzing the effects of more features at the same time. This article proposes a conceptual framework that enables effective analysis workflows, mitigating state-of-the-art limitations. The proposed framework allows for exploring and refining computed partial dependences, observing incrementally accurate results, and steering the computation of new partial dependences on user-selected subspaces of the combinatorial and intractable space. With this approach, the user can save both computational and cognitive costs, in contrast with the standard monolithic approach that computes all the possible combinations of features on all their domains in batch. The framework is the result of a careful design process involving experts' knowledge during its validation and informed the development of a prototype, W4SP1, that demonstrates its applicability traversing its different paths. A case study shows the advantages of the proposed approach.

TSOANet: Time-Sensitive Orthogonal Attention Network for medical event prediction

Medical Event Prediction (MEP) based on Electronic Medical Records (EMR) is an essential and valuable task for healthcare. For a patient, information in the EMR can be organized into a structured sequence, consisting of multiple visits each with details about visit time and various types of medical events. As the time intervals between neighboring visits are irregular and the medical events at different visits can vary significantly, MEP based on EMR is still challenging. Many studies have been proposed to model the irregular time intervals, relations among different types of medical events within each visit and relations among medical events across visits, and reported exciting results. However, most of these studies focus on two out of the three aspects mentioned above, with only a few addressing all the three aspects simultaneously. In this study, we propose a novel network, the Time-Sensitive Orthogonal Attention Network (TSOANet), which can fully utilize the irregular time intervals, relations among different types of medical events within and across visits. In particular, we design two key components: (1) Time-Sensitive Block, used to model the time intervals at both local and global levels to determine the impact of each visit in EMR; (2) Orthogonal Attention Block, used to model relations among different types of medical events within each visit and across visits in two axes, that is, event axis and time axis. Extensive experiments on two public real-world EMR datasets demonstrate that TSOANet outperforms the state-of-the-art models for various prediction tasks, thereby verifying the effectiveness of our approach. The source code of TSOANet is released at https://github.com/chh13502/TSOANet.

TrajVis: a visual clinical decision support system to translate artificial intelligence trajectory models in the precision management of chronic kidney disease

OBJECTIVE

Our objective is to develop and validate TrajVis, an interactive tool that assists clinicians in using artificial intelligence (AI) models to leverage patients' longitudinal electronic medical records (EMRs) for personalized precision management of chronic disease progression.

MATERIALS AND METHODS

We first perform requirement analysis with clinicians and data scientists to determine the visual analytics tasks of the TrajVis system as well as its design and functionalities. A graph AI model for chronic kidney disease (CKD) trajectory inference named DisEase PrOgression Trajectory (DEPOT) is used for system development and demonstration. TrajVis is implemented as a full-stack web application with synthetic EMR data derived from the Atrium Health Wake Forest Baptist Translational Data Warehouse and the Indiana Network for Patient Care research database. A case study with a nephrologist and a user experience survey of clinicians and data scientists are conducted to evaluate the TrajVis system.

RESULTS

The TrajVis clinical information system is composed of 4 panels: the Patient View for demographic and clinical information, the Trajectory View to visualize the DEPOT-derived CKD trajectories in latent space, the Clinical Indicator View to elucidate longitudinal patterns of clinical features and interpret DEPOT predictions, and the Analysis View to demonstrate personal CKD progression trajectories. System evaluations suggest that TrajVis supports clinicians in summarizing clinical data, identifying individualized risk predictors, and visualizing patient disease progression trajectories, overcoming the barriers of AI implementation in healthcare.

DISCUSSION

The TrajVis system provides a novel visualization solution which is complimentary to other risk estimators such as the Kidney Failure Risk Equations.

CONCLUSION

TrajVis bridges the gap between the fast-growing AI/ML modeling and the clinical use of such models for personalized and precision management of chronic diseases.

Recent advancements and applications of deep learning in heart failure: Α systematic review

BACKGROUND

Heart failure (HF), a global health challenge, requires innovative diagnostic and management approaches. The rapid evolution of deep learning (DL) in healthcare necessitates a comprehensive review to evaluate these developments and their potential to enhance HF evaluation, aligning clinical practices with technological advancements.

OBJECTIVE

This review aims to systematically explore the contributions of DL technologies in the assessment of HF, focusing on their potential to improve diagnostic accuracy, personalize treatment strategies, and address the impact of comorbidities.

METHODS

A thorough literature search was conducted across four major electronic databases: PubMed, Scopus, Web of Science and IEEE Xplore, yielding 137 articles that were subsequently categorized into five primary application areas: cardiovascular disease (CVD) classification, HF detection, image analysis, risk assessment, and other clinical analyses. The selection criteria focused on studies utilizing DL algorithms for HF assessment, not limited to HF detection but extending to any attempt in analyzing and interpreting HF-related data.

RESULTS

The analysis revealed a notable emphasis on CVD classification and HF detection, with DL algorithms showing significant promise in distinguishing between affected individuals and healthy subjects. Furthermore, the review highlights DL's capacity to identify underlying cardiomyopathies and other comorbidities, underscoring its utility in refining diagnostic processes and tailoring treatment plans to individual patient needs.

CONCLUSIONS

This review establishes DL as a key innovation in HF management, highlighting its role in advancing diagnostic accuracy and personalized care. The insights provided advocate for the integration of DL in clinical settings and suggest directions for future research to enhance patient outcomes in HF care.

An Infrastructure Framework for Remote Patient Monitoring Interventions and Research

Remote patient monitoring (RPM) enables clinicians to maintain and adjust their patients' plan of care by using remotely gathered data, such as vital signs, to proactively make medical decisions about a patient's care. RPM interventions have been touted as a means to improve patient care and well-being while reducing costs and resource needs within the health care ecosystem. However, multiple interworking components must be successfully implemented for an RPM intervention to yield the desired outcomes, and the design and key driver of each component can vary depending on the medical context. This viewpoint and perspective paper presents a 4-component RPM infrastructure framework based on a synthesis of existing literature and practice related to RPM. Specifically, these components are identified and considered: (1) data collection, (2) data transmission and storage, (3) data analysis, and (4) information presentation. Interaction points to consider between components include transmission, interoperability, accessibility, workflow integration, and transparency. Within each of the 4 components, questions affecting research and practice emerge that can affect the outcomes of RPM interventions. This framework provides a holistic perspective of the technologies involved in RPM interventions and how these core elements interact to provide an appropriate infrastructure for deploying RPM in health systems. Further, it provides a common vocabulary to compare and contrast RPM solutions across health contexts and may stimulate new research and intervention opportunities.

Targeted-BEHRT: Deep Learning for Observational Causal Inference on Longitudinal Electronic Health Records

Observational causal inference is useful for decision-making in medicine when randomized clinical trials (RCTs) are infeasible or nongeneralizable. However, traditional approaches do not always deliver unconfounded causal conclusions in practice. The rise of "doubly robust" nonparametric tools coupled with the growth of deep learning for capturing rich representations of multimodal data offers a unique opportunity to develop and test such models for causal inference on comprehensive electronic health records (EHRs). In this article, we investigate causal modeling of an RCT-established causal association: the effect of classes of antihypertensive on incident cancer risk. We develop a transformer-based model, targeted bidirectional EHR transformer (T-BEHRT) coupled with doubly robust estimation to estimate average risk ratio (RR). We compare our model to benchmark statistical and deep learning models for causal inference in multiple experiments on semi-synthetic derivations of our dataset with various types and intensities of confounding. In order to further test the reliability of our approach, we test our model on situations of limited data. We find that our model provides more accurate estimates of relative risk [least sum absolute error (SAE) from ground truth] compared with benchmark estimations. Finally, our model provides an estimate of class-wise antihypertensive effect on cancer risk that is consistent with results derived from RCTs.

Interpretable Inference and Classification of Tissue Types in Histological Colorectal Cancer Slides Based on Ensembles Adaptive Boosting Prototype Tree

Digital pathology images are treated as the "gold standard" for the diagnosis of colorectal lesions, especially colon cancer. Real-time, objective and accurate inspection results will assist clinicians to choose symptomatic treatment in a timely manner, which is of great significance in clinical medicine. However, Manual methods suffers from long inspection cycle and serious reliance on subjective interpretation. It is also a challenging task for existing computer-aided diagnosis methods to obtain models that are both accurate and interpretable. Models that exhibit high accuracy are always more complex and opaque, while interpretable models may lack the necessary accuracy. Therefore, the framework of ensemble adaptive boosting prototype tree is proposed to predict the colorectal pathology images and provide interpretable inference by visualizing the decision-making process in each base learner. The results showed that the proposed method could effectively address the "accuracy-interpretability trade-off" issue by ensemble of m adaptive boosting neural prototype trees. The superior performance of the framework provides a novel paradigm for interpretable inference and high-precision prediction of pathology image patches in computational pathology.

Predicting line of therapy transition via similar patient augmentation

Analyzing large EHR databases to predict cancer progression and treatments has become a hot trend in recent years. An increasing number of modern deep learning models have been proposed to find the milestones of essential patient medical journey characteristics to predict their disease status and give healthcare professionals valuable insights. However, most of the existing methods are lack of consideration for the inter-relationship among different patients. We believe that more valuable information can be extracted, especially when patients with similar disease statuses visit the same doctors. Towards this end, a similar patient augmentation-based approach named SimPA is proposed to enhance the learning of patient representations and further predict lines of therapy transition. Our experiment results on a real-world multiple myeloma dataset show that our proposed approach outperforms state-of-the-art baseline approaches in terms of standard evaluation metrics for classification tasks.

MVIRA: A model based on Missing Value Imputation and Reliability Assessment for mortality risk prediction

BACKGROUND

Mortality risk prediction is to predict whether a patient has the risk of death based on relevant diagnosis and treatment data. How to accurately predict patient mortality risk based on electronic health records (EHR) is currently a hot research topic in the healthcare field. In actual medical datasets, there are often many missing values, which can seriously interfere with the effect of model prediction. However, when missing values are interpolated, most existing methods do not take into account the fidelity or confidence of the interpolated values. Misestimation of missing variables can lead to modeling difficulties and performance degradation, while the reliability of the model may be compromised in clinical environments.

MATERIALS AND METHODS

We propose a model based on Missing Value Imputation and Reliability Assessment for mortality risk prediction (MVIRA). The model uses a combination of variational autoencoder and recurrent neural networks to complete the interpolation of missing values and enhance the characterization ability of EHR data, thus improving the performance of mortality risk prediction. In addition, we also introduce the Monte Carlo Dropout method to calculate the uncertainty of the model prediction results and thus achieve the reliability assessment of the model.

RESULTS

We perform performance validation of the model on the public datasets MIMIC-III and MIMIC-IV. The proposed model showed improved performance compared with competitive models in terms of overall specialties.

CONCLUSION

The proposed model can effectively improve the accuracy of mortality risk prediction, and can help medical institutions assess the condition of patients.

Prediction models for heart failure in the community: A systematic review and meta-analysis

AIMS

Multivariable prediction models can be used to estimate risk of incident heart failure (HF) in the general population. A systematic review and meta-analysis was performed to determine the performance of models.

METHODS AND RESULTS

From inception to 3 November 2022 MEDLINE and EMBASE databases were searched for studies of multivariable models derived, validated and/or augmented for HF prediction in community-based cohorts. Discrimination measures for models with c-statistic data from ≥3 cohorts were pooled by Bayesian meta-analysis, with heterogeneity assessed through a 95% prediction interval (PI). Risk of bias was assessed using PROBAST. We included 36 studies with 59 prediction models. In meta-analysis, the Atherosclerosis Risk in Communities (ARIC) risk score (summary c-statistic 0.802, 95% confidence interval [CI] 0.707-0.883), GRaph-based Attention Model (GRAM; 0.791, 95% CI 0.677-0.885), Pooled Cohort equations to Prevent Heart Failure (PCP-HF) white men model (0.820, 95% CI 0.792-0.843), PCP-HF white women model (0.852, 95% CI 0.804-0.895), and REverse Time AttentIoN model (RETAIN; 0.839, 95% CI 0.748-0.916) had a statistically significant 95% PI and excellent discrimination performance. The ARIC risk score and PCP-HF models had significant summary discrimination among cohorts with a uniform prediction window. 77% of model results were at high risk of bias, certainty of evidence was low, and no model had a clinical impact study.

CONCLUSIONS

Prediction models for estimating risk of incident HF in the community demonstrate excellent discrimination performance. Their usefulness remains uncertain due to high risk of bias, low certainty of evidence, and absence of clinical effectiveness research.

ClinicalPath: A Visualization Tool to Improve the Evaluation of Electronic Health Records in Clinical Decision-Making

Physicians work at a very tight schedule and need decision-making support tools to help on improving and doing their work in a timely and dependable manner. Examining piles of sheets with test results and using systems with little visualization support to provide diagnostics is daunting, but that is still the usual way for the physicians' daily procedure, especially in developing countries. Electronic Health Records systems have been designed to keep the patients' history and reduce the time spent analyzing the patient's data. However, better tools to support decision-making are still needed. In this article, we propose ClinicalPath, a visualization tool for users to track a patient's clinical path through a series of tests and data, which can aid in treatments and diagnoses. Our proposal is focused on patient's data analysis, presenting the test results and clinical history longitudinally. Both the visualization design and the system functionality were developed in close collaboration with experts in the medical domain to ensure a right fit of the technical solutions and the real needs of the professionals. We validated the proposed visualization based on case studies and user assessments through tasks based on the physician's daily activities. Our results show that our proposed system improves the physicians' experience in decision-making tasks, made with more confidence and better usage of the physicians' time, allowing them to take other needed care for the patients.

An intelligent adverse delivery outcomes prediction model based on the fusion of multiple obstetric clinical data

Adverse delivery outcomes is a major re-productive health problem that affects the physical and mental health of pregnant women. Obviously, obstetric clinical data has periodically time series characteristics. This paper proposed a three stage adverse delivery outcomes prediction model via the fusion of multiple time series clinical data. The first stage is data aggregation, in which the data set is collected from the obstetric clinical data and divided based on time series features. In the second stage, a multi-channel gated cycle unit is used to solve the calculation error caused by irregular sampling of time series data. The hidden layer feature vector is connected with the fully connected layer, reshaped into a new one-dimensional feature, and fused with the non-time series data into a new data set. The third stage is the prediction stage of adverse delivery outcomes. By connecting the multi-channel gated cycle unit with the extreme gradient lift, the data transmitted in the corresponding channel is used in the feature extraction stage, in which the weighted entropy-based feature extraction is adopted. With the help of the extracted features, a hybrid artificial neural network architecture (MGRU-XGB) was developed to predict adverse delivery outcomes. The experimental results showed that the hybrid model had the best prediction performance for adverse delivery outcomes compared with other single models in terms of sensitivity, specificity, AUC and other evaluation indexes.

RNA trafficking and subcellular localization-a review of mechanisms, experimental and predictive methodologies

RNA localization is essential for regulating spatial translation, where RNAs are trafficked to their target locations via various biological mechanisms. In this review, we discuss RNA localization in the context of molecular mechanisms, experimental techniques and machine learning-based prediction tools. Three main types of molecular mechanisms that control the localization of RNA to distinct cellular compartments are reviewed, including directed transport, protection from mRNA degradation, as well as diffusion and local entrapment. Advances in experimental methods, both image and sequence based, provide substantial data resources, which allow for the design of powerful machine learning models to predict RNA localizations. We review the publicly available predictive tools to serve as a guide for users and inspire developers to build more effective prediction models. Finally, we provide an overview of multimodal learning, which may provide a new avenue for the prediction of RNA localization.

An innovative ensemble model based on deep learning for predicting COVID-19 infection

Nowadays, global public health crises are occurring more frequently, and accurate prediction of these diseases can reduce the burden on the healthcare system. Taking COVID-19 as an example, accurate prediction of infection can assist experts in effectively allocating medical resources and diagnosing diseases. Currently, scholars worldwide use single model approaches or epidemiology models more often to predict the outbreak trend of COVID-19, resulting in poor prediction accuracy. Although a few studies have employed ensemble models, there is still room for improvement in their performance. In addition, there are only a few models that use the laboratory results of patients to predict COVID-19 infection. To address these issues, research efforts should focus on improving disease prediction performance and expanding the use of medical disease prediction models. In this paper, we propose an innovative deep learning model Whale Optimization Convolutional Neural Networks (CNN), Long-Short Term Memory (LSTM) and Artificial Neural Network (ANN) called WOCLSA which incorporates three models ANN, CNN and LSTM. The WOCLSA model utilizes the Whale Optimization Algorithm to optimize the neuron number, dropout and batch size parameters in the integrated model of ANN, CNN and LSTM, thereby finding the global optimal solution parameters. WOCLSA employs 18 patient indicators as predictors, and compares its results with three other ensemble deep learning models. All models were validated with train-test split approaches. We evaluate and compare our proposed model and other models using accuracy, F1 score, recall, AUC and precision metrics. Through many studies and tests, our results show that our prediction models can identify patients with COVID-19 infection at the AUC of 91%, 91%, and 93% respectively. Other prediction results achieve a respectable accuracy of 92.82%, 92.79%, and 91.66% respectively, f1-score of 93.41%, 92.79%, and 92.33% respectively, precision of 93.41%, 92.79%, and 92.33% respectively, recall of 93.41%, 92.79%, and 92.33% respectively. All of these exceed 91%, surpassing those of comparable models. The execution time of WOCLSA is also an advantage. Therefore, the WOCLSA ensemble model can be used to assist in verifying laboratory research results and predict and to judge various diseases in public health events.

Solving the explainable AI conundrum by bridging clinicians' needs and developers' goals

Explainable artificial intelligence (XAI) has emerged as a promising solution for addressing the implementation challenges of AI/ML in healthcare. However, little is known about how developers and clinicians interpret XAI and what conflicting goals and requirements they may have. This paper presents the findings of a longitudinal multi-method study involving 112 developers and clinicians co-designing an XAI solution for a clinical decision support system. Our study identifies three key differences between developer and clinician mental models of XAI, including opposing goals (model interpretability vs. clinical plausibility), different sources of truth (data vs. patient), and the role of exploring new vs. exploiting old knowledge. Based on our findings, we propose design solutions that can help address the XAI conundrum in healthcare, including the use of causal inference models, personalized explanations, and ambidexterity between exploration and exploitation mindsets. Our study highlights the importance of considering the perspectives of both developers and clinicians in the design of XAI systems and provides practical recommendations for improving the effectiveness and usability of XAI in healthcare.

A Study of Healthcare Team Communication Networks using Visual Analytics

Cooperation among teams or individuals of healthcare professionals (HCPs) is one of the crucial factors towards patients' survival outcome. However, it is challenging to uncover and understand such factors in the complex Multiteam System (MTS) communication networks representing daily HCP cooperation. In this paper, we present a study on MTS communication networks constructed with real-world cancer patients' Electronic Health Record (EHR) access logs. We adopt a visual analytics workflow to extract associations between semantic characteristics of MTS communication networks and the patients' survival outcomes. The workflow consists of a neural network learning phase to classify the data based on the chosen input and output attributes, a dimensionality reduction and optimization phase to produce a simplified set of results for examination, and finally an interpreting phase conducted by the user through an interactive visualization interface. We provide the insights found using this workflow with two case studies and an expert interview.

A fair and interpretable network for clinical risk prediction: a regularized multi-view multi-task learning approach

In healthcare domain, complication risk profiling which can be seen as multiple clinical risk prediction tasks is challenging due to the complex interaction between heterogeneous clinical entities. With the availability of real-world data, many deep learning methods are proposed for complication risk profiling. However, the existing methods face three open challenges. First, they leverage clinical data from a single view and then lead to suboptimal models. Second, most existing methods lack an effective mechanism to interpret predictions. Third, models learned from clinical data may have inherent pre-existing biases and exhibit discrimination against certain social groups. We then propose a multi-view multi-task network (MuViTaNet) to tackle these issues. MuViTaNet complements patient representation by using a multi-view encoder to exploit more information. Moreover, it uses a multi-task learning to generate more generalized representations using both labeled and unlabeled datasets. Last, a fairness variant (F-MuViTaNet) is proposed to mitigate the unfairness issues and promote healthcare equity. The experiments show that MuViTaNet outperforms existing methods for cardiac complication profiling. Its architecture also provides an effective mechanism for interpreting the predictions, which helps clinicians discover the underlying mechanism triggering the complication onsets. F-MuViTaNet can also effectively mitigate the unfairness with only negligible impact on accuracy.

Prediction of Brain Metastases Development in Patients With Lung Cancer by Explainable Artificial Intelligence From Electronic Health Records

PURPOSE

Early detection of brain metastases (BMs) is critical for prompt treatment and optimal control of the disease. In this study, we seek to predict the risk of developing BM among patients diagnosed with lung cancer on the basis of electronic health record (EHR) data and to understand what factors are important for the model to predict BM development through explainable artificial intelligence approaches accurately.

MATERIALS AND METHODS

We trained a recurrent neural network model, REverse Time AttentIoN (RETAIN), to predict the risk of developing BM using structured EHR data. To interpret the model's decision process, we analyzed the attention weights in the RETAIN model and the SHAP values from a feature attribution method, Kernel SHAP, to identify the factors contributing to BM prediction.

RESULTS

We developed a high-quality cohort with 4,466 patients with BM from the Cerner Health Fact database, which contains over 70 million patients from more than 600 hospitals. RETAIN uses this data set to achieve the best area under the receiver operating characteristic curve at 0.825, a significant improvement over the baseline model. We also extended a feature attribution method, Kernel SHAP, to structured EHR data for model interpretation. Both RETAIN and Kernel SHAP can identify important features related to BM prediction.

CONCLUSION

To the best of our knowledge, this is the first study to predict BM using structured EHR data. We achieved decent prediction performance for BM prediction and identified factors highly relevant to BM development. The sensitivity analysis demonstrated that both RETAIN and Kernel SHAP could discriminate unrelated features and put more weight on the features important to BM. Our study explored the potential of applying explainable artificial intelligence for future clinical applications.

Time interval uncertainty-aware and text-enhanced based disease prediction

Deep learning methods have achieved success in disease prediction using electronic health records (EHR) data. Most of the existing methods have some limitations. First, most of the methods adopt a homogeneous decay way to deal with the effect of time interval on patient's previous visits information. However, the effect of the time interval between patient's visits is not always negative. For example, although the time interval between visits for patients with chronic diseases is relatively long, the importance of the previous visit to the next visit is high, and we may not be able to consider the effect of the time interval as negative at this point. That is, the effect of the time interval on previous visits is exerted in a nonmonotonic manner, and it is either positive, negative, or neutral. In addition, the effect of text information on prediction results is not taken into account in most of methods. The text in EHR contains a description of the patient's past medical history and current symptoms of the disease, which is important for prediction results. In order to solve these issues, we propose a Time Interval Uncertainty-Aware and Text-Enhanced Based Disease Prediction Model, which utilizes the uncertain effects of time intervals and patient's text information for disease prediction. Firstly, we apply a cross-attention mechanism to generate a global representation of the patient using the patient's disease and text information from the EHR. Then, we use the key-query attention mechanism to obtain the two importance weights of the two visit sequences with and without time intervals, respectively. Furthermore, we achieve disease prediction by making slight adjustments to the encode part of the Transformer, a deep learning model based on a self-attention mechanism. We compare with various state-of-the-art models on two publicly available datasets, MIMIC-III and MIMIC-IV, and select the top 10 diseases with the highest frequency in the dataset as the target diseases. On the MIMIC-III dataset, our model is up to three percent higher than the optimal baseline in terms of evaluation metrics.

Polyphony: an Interactive Transfer Learning Framework for Single-Cell Data Analysis

Reference-based cell-type annotation can significantly reduce time and effort in single-cell analysis by transferring labels from a previously-annotated dataset to a new dataset. However, label transfer by end-to-end computational methods is challenging due to the entanglement of technical (e.g., from different sequencing batches or techniques) and biological (e.g., from different cellular microenvironments) variations, only the first of which must be removed. To address this issue, we propose Polyphony, an interactive transfer learning (ITL) framework, to complement biologists' knowledge with advanced computational methods. Polyphony is motivated and guided by domain experts' needs for a controllable, interactive, and algorithm-assisted annotation process, identified through interviews with seven biologists. We introduce anchors, i.e., analogous cell populations across datasets, as a paradigm to explain the computational process and collect user feedback for model improvement. We further design a set of visualizations and interactions to empower users to add, delete, or modify anchors, resulting in refined cell type annotations. The effectiveness of this approach is demonstrated through quantitative experiments, two hypothetical use cases, and interviews with two biologists. The results show that our anchor-based ITL method takes advantage of both human and machine intelligence in annotating massive single-cell datasets.

Extending the Nested Model for User-Centric XAI: A Design Study on GNN-based Drug Repurposing

Whether AI explanations can help users achieve specific tasks efficiently (i.e., usable explanations) is significantly influenced by their visual presentation. While many techniques exist to generate explanations, it remains unclear how to select and visually present AI explanations based on the characteristics of domain users. This paper aims to understand this question through a multidisciplinary design study for a specific problem: explaining graph neural network (GNN) predictions to domain experts in drug repurposing, i.e., reuse of existing drugs for new diseases. Building on the nested design model of visualization, we incorporate XAI design considerations from a literature review and from our collaborators' feedback into the design process. Specifically, we discuss XAI-related design considerations for usable visual explanations at each design layer: target user, usage context, domain explanation, and XAI goal at the domain layer; format, granularity, and operation of explanations at the abstraction layer; encodings and interactions at the visualization layer; and XAI and rendering algorithm at the algorithm layer. We present how the extended nested model motivates and informs the design of DrugExplorer, an XAI tool for drug repurposing. Based on our domain characterization, DrugExplorer provides path-based explanations and presents them both as individual paths and meta-paths for two key XAI operations, why and what else. DrugExplorer offers a novel visualization design called MetaMatrix with a set of interactions to help domain users organize and compare explanation paths at different levels of granularity to generate domain-meaningful insights. We demonstrate the effectiveness of the selected visual presentation and DrugExplorer as a whole via a usage scenario, a user study, and expert interviews. From these evaluations, we derive insightful observations and reflections that can inform the design of XAI visualizations for other scientific applications.

HetVis: A Visual Analysis Approach for Identifying Data Heterogeneity in Horizontal Federated Learning

Horizontal federated learning (HFL) enables distributed clients to train a shared model and keep their data privacy. In training high-quality HFL models, the data heterogeneity among clients is one of the major concerns. However, due to the security issue and the complexity of deep learning models, it is challenging to investigate data heterogeneity across different clients. To address this issue, based on a requirement analysis we developed a visual analytics tool, HetVis, for participating clients to explore data heterogeneity. We identify data heterogeneity through comparing prediction behaviors of the global federated model and the stand-alone model trained with local data. Then, a context-aware clustering of the inconsistent records is done, to provide a summary of data heterogeneity. Combining with the proposed comparison techniques, we develop a novel set of visualizations to identify heterogeneity issues in HFL. We designed three case studies to introduce how HetVis can assist client analysts in understanding different types of heterogeneity issues. Expert reviews and a comparative study demonstrate the effectiveness of HetVis.

Survey on Visual Analysis of Event Sequence Data

Event sequence data record series of discrete events in the time order of occurrence. They are commonly observed in a variety of applications ranging from electronic health records to network logs, with the characteristics of large-scale, high-dimensional and heterogeneous. This high complexity of event sequence data makes it difficult for analysts to manually explore and find patterns, resulting in ever-increasing needs for computational and perceptual aids from visual analytics techniques to extract and communicate insights from event sequence datasets. In this paper, we review the state-of-the-art visual analytics approaches, characterize them with our proposed design space, and categorize them based on analytical tasks and applications. From our review of relevant literature, we have also identified several remaining research challenges and future research opportunities.

AutoMed: Automated Medical Risk Predictive Modeling on Electronic Health Records

Electronic health records (EHR) have been widely applied to various tasks in the medical domain such as risk predictive modeling, which aims to predict further health conditions by analyzing patients' historical EHR. Existing work mainly focuses on modeling the sequential and temporal characteristics of EHR data with advanced deep learning techniques. However, the network architectures of these models are all manually designed based on experts' prior knowledge, which largely impedes non-experts from exploring this task. To address this issue, in this paper, we propose a novel automated risk prediction model named AutoMed to automatically search the optimal model architecture for modeling the complex EHR data and improving the performance of the risk prediction task. In particular, we follow the idea of neural architecture search to design a search space that contains three separate searchable modules. Two of them are used for analyzing sequential and temporal features of EHR data, respectively. The third is to automatically fuse both features together. Besides these three modules, AutoMed contains an embedding module and a prediction module. All the three searchable modules are jointly optimized in the search stage to derive the optimal model architecture. In such a way, the model design can be automatically achieved with few human interventions. Experimental results on three real-world datasets show that AutoMed outperforms state-of-the-art baselines in terms of PR-AUC, F1, and Cohen's Kappa. Moreover, the ablation study shows that AutoMed can obtain reasonable model architectures and offer useful insights to the future risk prediction model design.

CATNet: Cross-event attention-based time-aware network for medical event prediction

Medical event prediction (MEP) is a fundamental task in the healthcare domain, which needs to predict medical events, including medications, diagnosis codes, laboratory tests, procedures, outcomes, and so on, according to historical medical records of patients. Many researchers have tried to build MEP models to overcome the challenges caused by the heterogeneous and irregular temporal characteristics of EHR data. However, most of them consider the heterogenous and temporal medical events separately and ignore the correlations among different types of medical events, especially relations between heterogeneous historical medical events and target medical events. In this paper, we propose a novel neural network based on attention mechanism called Cross-event Attention-based Time-aware Network (CATNet) for MEP. It is a time-aware, event-aware and task-adaptive method with the following advantages: 1) modeling heterogeneous information and temporal information in a unified way and considering irregular temporal characteristics locally and globally respectively, 2) taking full advantage of correlations among different types of events via cross-event attention. Experiments on two public datasets (MIMIC-III and eICU) show CATNet outperforms other state-of-the-art methods on various MEP tasks. The source code of CATNet is released at https://github.com/sherry6247/CATNet.git.

Real-world data: a brief review of the methods, applications, challenges and opportunities

Background

The increased adoption of the internet, social media, wearable devices, e-health services, and other technology-driven services in medicine and healthcare has led to the rapid generation of various types of digital data, providing a valuable data source beyond the confines of traditional clinical trials, epidemiological studies, and lab-based experiments.

Methods

We provide a brief overview on the type and sources of real-world data and the common models and approaches to utilize and analyze real-world data. We discuss the challenges and opportunities of using real-world data for evidence-based decision making This review does not aim to be comprehensive or cover all aspects of the intriguing topic on RWD (from both the research and practical perspectives) but serves as a primer and provides useful sources for readers who interested in this topic.

Results and Conclusions

Real-world hold great potential for generating real-world evidence for designing and conducting confirmatory trials and answering questions that may not be addressed otherwise. The voluminosity and complexity of real-world data also call for development of more appropriate, sophisticated, and innovative data processing and analysis techniques while maintaining scientific rigor in research findings, and attentions to data ethics to harness the power of real-world data.

Explainable AI for clinical and remote health applications: a survey on tabular and time series data

Nowadays Artificial Intelligence (AI) has become a fundamental component of healthcare applications, both clinical and remote, but the best performing AI systems are often too complex to be self-explaining. Explainable AI (XAI) techniques are defined to unveil the reasoning behind the system’s predictions and decisions, and they become even more critical when dealing with sensitive and personal health data. It is worth noting that XAI has not gathered the same attention across different research areas and data types, especially in healthcare. In particular, many clinical and remote health applications are based on tabular and time series data, respectively, and XAI is not commonly analysed on these data types, while computer vision and Natural Language Processing (NLP) are the reference applications. To provide an overview of XAI methods that are most suitable for tabular and time series data in the healthcare domain, this paper provides a review of the literature in the last 5 years, illustrating the type of generated explanations and the efforts provided to evaluate their relevance and quality. Specifically, we identify clinical validation, consistency assessment, objective and standardised quality evaluation, and human-centered quality assessment as key features to ensure effective explanations for the end users. Finally, we highlight the main research challenges in the field as well as the limitations of existing XAI methods.

Evaluation of a clinical decision support system for detection of patients at risk after kidney transplantation

Patient care after kidney transplantation requires integration of complex information to make informed decisions on risk constellations. Many machine learning models have been developed for detecting patient outcomes in the past years. However, performance metrics alone do not determine practical utility. We present a newly developed clinical decision support system (CDSS) for detection of patients at risk for rejection and death-censored graft failure. The CDSS is based on clinical routine data including 1,516 kidney transplant recipients and more than 100,000 data points. In a reader study we compare the performance of physicians at a nephrology department with and without the CDSS. Internal validation shows AUC-ROC scores of 0.83 for rejection, and 0.95 for graft failure. The reader study shows that predictions by physicians converge toward the CDSS. However, performance does not improve (AUC-ROC; 0.6413 vs. 0.6314 for rejection; 0.8072 vs. 0.7778 for graft failure). Finally, the study shows that the CDSS detects partially different patients at risk compared to physicians. This indicates that the combination of both, medical professionals and a CDSS might help detect more patients at risk for graft failure. However, the question of how to integrate such a system efficiently into clinical practice remains open.

Modeling electronic health record data using an end-to-end knowledge-graph-informed topic model

The rapid growth of electronic health record (EHR) datasets opens up promising opportunities to understand human diseases in a systematic way. However, effective extraction of clinical knowledge from EHR data has been hindered by the sparse and noisy information. We present Graph ATtention-Embedded Topic Model (GAT-ETM), an end-to-end taxonomy-knowledge-graph-based multimodal embedded topic model. GAT-ETM distills latent disease topics from EHR data by learning the embedding from a constructed medical knowledge graph. We applied GAT-ETM to a large-scale EHR dataset consisting of over 1 million patients. We evaluated its performance based on topic quality, drug imputation, and disease diagnosis prediction. GAT-ETM demonstrated superior performance over the alternative methods on all tasks. Moreover, GAT-ETM learned clinically meaningful graph-informed embedding of the EHR codes and discovered interpretable and accurate patient representations for patient stratification and drug recommendations. GAT-ETM code is available at https://github.com/li-lab-mcgill/GAT-ETM .

An Explainable Transformer-Based Deep Learning Model for the Prediction of Incident Heart Failure

Predicting the incidence of complex chronic conditions such as heart failure is challenging. Deep learning models applied to rich electronic health records may improve prediction but remain unexplainable hampering their wider use in medical practice. We aimed to develop a deep-learning framework for accurate and yet explainable prediction of 6-month incident heart failure (HF). Using 100,071 patients from longitudinal linked electronic health records across the U.K., we applied a novel Transformer-based risk model using all community and hospital diagnoses and medications contextualized within the age and calendar year for each patient's clinical encounter. Feature importance was investigated with an ablation analysis to compare model performance when alternatively removing features and by comparing the variability of temporal representations. A post-hoc perturbation technique was conducted to propagate the changes in the input to the outcome for feature contribution analyses. Our model achieved 0.93 area under the receiver operator curve and 0.69 area under the precision-recall curve on internal 5-fold cross validation and outperformed existing deep learning models. Ablation analysis indicated medication is important for predicting HF risk, calendar year is more important than chronological age, which was further reinforced by temporal variability analysis. Contribution analyses identified risk factors that are closely related to HF. Many of them were consistent with existing knowledge from clinical and epidemiological research but several new associations were revealed which had not been considered in expert-driven risk prediction models. In conclusion, the results highlight that our deep learning model, in addition high predictive performance, can inform data-driven risk factor identification.

A survey on the interpretability of deep learning in medical diagnosis

Deep learning has demonstrated remarkable performance in the medical domain, with accuracy that rivals or even exceeds that of human experts. However, it has a significant problem that these models are "black-box" structures, which means they are opaque, non-intuitive, and difficult for people to understand. This creates a barrier to the application of deep learning models in clinical practice due to lack of interpretability, trust, and transparency. To overcome this problem, several studies on interpretability have been proposed. Therefore, in this paper, we comprehensively review the interpretability of deep learning in medical diagnosis based on the current literature, including some common interpretability methods used in the medical domain, various applications with interpretability for disease diagnosis, prevalent evaluation metrics, and several disease datasets. In addition, the challenges of interpretability and future research directions are also discussed here. To the best of our knowledge, this is the first time that various applications of interpretability methods for disease diagnosis have been summarized.

Human-centered explainability for life sciences, healthcare, and medical informatics

Rapid advances in artificial intelligence (AI) and availability of biological, medical, and healthcare data have enabled the development of a wide variety of models. Significant success has been achieved in a wide range of fields, such as genomics, protein folding, disease diagnosis, imaging, and clinical tasks. Although widely used, the inherent opacity of deep AI models has brought criticism from the research field and little adoption in clinical practice. Concurrently, there has been a significant amount of research focused on making such methods more interpretable, reviewed here, but inherent critiques of such explainability in AI (XAI), its requirements, and concerns with fairness/robustness have hampered their real-world adoption. We here discuss how user-driven XAI can be made more useful for different healthcare stakeholders through the definition of three key personas-data scientists, clinical researchers, and clinicians-and present an overview of how different XAI approaches can address their needs. For illustration, we also walk through several research and clinical examples that take advantage of XAI open-source tools, including those that help enhance the explanation of the results through visualization. This perspective thus aims to provide a guidance tool for developing explainability solutions for healthcare by empowering both subject matter experts, providing them with a survey of available tools, and explainability developers, by providing examples of how such methods can influence in practice adoption of solutions.

TAHDNet: Time-Aware Hierarchical Dependency Network for Medication Recommendation

Medication recommendation is a hot topic in the research of applying neural networks to the healthcare area. Although extensive progressions have been made, current researches still face the following challenges: i). Existing methods are poor at efficiently capturing and leveraging local and global dependency information from patient visit records. ii). Current time-aware models based on irregularly interval medical records tend to ignore periodic variability in patient conditions, which limits the representational learning capability of these models. Therefore, we propose a Dynamic Time-aware Hierarchical Dependency Network (TAHDNet) for the medication recommendation task to address these challenges. Firstly, we use a Transformer-based model to learn the global information of the whole patient record through a self-supervised pre-training process. Secondly, a 1D-CNN model is used to learn the local dependencies on visitation level. Thirdly, we propose a dynamic time-aware module with a fused temporal decay function to assign different weights among different time intervals dynamically through a key-value attention mechanism. Experimental results on real-world datasets demonstrate the effectiveness of the model proposed in this paper.

HealthFetch: An Influence-Based, Context-Aware Prefetch Scheme in Citizen-Centered Health Storage Clouds

Over the past few years, increasing attention has been given to the health sector and the integration of new technologies into it. Cloud computing and storage clouds have become essentially state of the art solutions for other major areas and have started to rapidly make their presence powerful in the health sector as well. More and more companies are working toward a future that will allow healthcare professionals to engage more with such infrastructures, enabling them a vast number of possibilities. While this is a very important step, less attention has been given to the citizens. For this reason, in this paper, a citizen-centered storage cloud solution is proposed that will allow citizens to hold their health data in their own hands while also enabling the exchange of these data with healthcare professionals during emergency situations. Not only that, in order to reduce the health data transmission delay, a novel context-aware prefetch engine enriched with deep learning capabilities is proposed. The proposed prefetch scheme, along with the proposed storage cloud, is put under a two-fold evaluation in several deployment and usage scenarios in order to examine its performance with respect to the data transmission times, while also evaluating its outcomes compared to other state of the art solutions. The results show that the proposed solution shows significant improvement of the download speed when compared with the storage cloud, especially when large data are exchanged. In addition, the results of the proposed scheme evaluation depict that the proposed scheme improves the overall predictions, considering the coefficient of determination (R2 > 0.94) and the mean of errors (RMSE < 1), while also reducing the training data by 12%.

Development of a Visualization Tool for Healthcare Decision-Making using Electronic Medical Records: A Systems Approach to Viewing a Patient Record

Healthcare delivery systems are widely accepted as socio-technical systems. Unlike other socio-technical systems, healthcare systems leave very little decision-making to technical automation and control. Instead, the healthcare delivery system relies on human healthcare resources for decision-making. Human decision-making is imperative to the clinical delivery of care to patients and to the operational processes that support care delivery, quality improvement, and other organizational management activities. For these clinical and operational activities, human resources make healthcare decisions using healthcare data typically housed in electronic medical records (EMRs). Unfortunately, EMR systems were first designed with the functional capability to store data, and, second to a lesser degree, to retrieve data. The literature recognizes the need to improve the retrieval of information from EMR systems. More specifically, there remains the need to directly view a patient's holistic health and healthcare trajectory. At this time, decision-makers are left to mentally build this holistic picture in their mind by sequentially clicking through many sections of the EMR. Therefore, in this paper, we develop a visualization tool to organize and present an individual's health and healthcare trajectory by describing a patient record holistically from a system architecture perspective. This approach is based on a previously developed system model for healthcare delivery and individual health outcomes.

Tribe or Not? Critical Inspection of Group Differences Using TribalGram

With the rise of AI and data mining techniques, group profiling and group-level analysis have been increasingly used in many domains, including policy making and direct marketing. In some cases, the statistics extracted from data may provide insights to a group’s shared characteristics; in others, the group-level analysis can lead to problems, including stereotyping and systematic oppression. How can analytic tools facilitate a more conscientious process in group analysis? In this work, we identify a set of accountable group analytics design guidelines to explicate the needs for group differentiation and preventing overgeneralization of a group. Following the design guidelines, we develop TribalGram , a visual analytic suite that leverages interpretable machine learning algorithms and visualization to offer inference assessment, model explanation, data corroboration, and sense-making. Through the interviews with domain experts, we showcase how our design and tools can bring a richer understanding of “groups” mined from the data.

Multi-channel Fusion LSTM for Medical Event Prediction using HERs

Automatic medical event prediction (MEP), e.g. diagnosis prediction, medication prediction, using electronic health records (EHRs) is a popular research direction in health informatics. In many cases, MEP relies on the determinations from different types of medical events, which demonstrates the heterogeneous nature of EHRs. However, most existing methods for MEP fail to distinguishingly model the type of event that is highly associated with the prediction task, i.e. task-wise event, which usually plays a more significant role than other events. In this paper, we proposed a Long Short-Term Memory network (LSTM)-based method for MEP, named Multi-Channel Fusion LSTM (MCF-LSTM), which models the correlations between different types of medical events using multiple network channels. To this end, we designed a task-wise fusion module, in which a gated network is applied to select how much information can be transferred between events. Furthermore, the irregular temporal interval between adjacent medical visits is also modeled in an individual channel, which is combined with other events in a unified manner. We compared MCF-LSTM with state-of-the-art methods on four MEP tasks on two public datasets: MIMIC-III and eICU. Experimental results show that MCF-LSTM achieves promising results on AUC(receiver operating characteristic curve), AUPR (area under the precision-recall curve), and top-k recall, and outperforms other methods with high stability.

Verification of De-Identification Techniques for Personal Information Using Tree-Based Methods with Shapley Values

With the development of big data and cloud computing technologies, the importance of pseudonym information has grown. However, the tools for verifying whether the de-identification methodology is correctly applied to ensure data confidentiality and usability are insufficient. This paper proposes a verification of de-identification techniques for personal healthcare information by considering data confidentiality and usability. Data are generated and preprocessed by considering the actual statistical data, personal information datasets, and de-identification datasets based on medical data to represent the de-identification technique as a numeric dataset. Five tree-based regression models (i.e., decision tree, random forest, gradient boosting machine, extreme gradient boosting, and light gradient boosting machine) are constructed using the de-identification dataset to effectively discover nonlinear relationships between dependent and independent variables in numerical datasets. Then, the most effective model is selected from personal information data in which pseudonym processing is essential for data utilization. The Shapley additive explanation, an explainable artificial intelligence technique, is applied to the most effective model to establish pseudonym processing policies and machine learning to present a machine-learning process that selects an appropriate de-identification methodology.

A Systematic Review of Explainable Artificial Intelligence in Terms of Different Application Domains and Tasks

Artificial intelligence (AI) and machine learning (ML) have recently been radically improved and are now being employed in almost every application domain to develop automated or semi-automated systems. To facilitate greater human acceptability of these systems, explainable artificial intelligence (XAI) has experienced significant growth over the last couple of years with the development of highly accurate models but with a paucity of explainability and interpretability. The literature shows evidence from numerous studies on the philosophy and methodologies of XAI. Nonetheless, there is an evident scarcity of secondary studies in connection with the application domains and tasks, let alone review studies following prescribed guidelines, that can enable researchers’ understanding of the current trends in XAI, which could lead to future research for domain- and application-specific method development. Therefore, this paper presents a systematic literature review (SLR) on the recent developments of XAI methods and evaluation metrics concerning different application domains and tasks. This study considers 137 articles published in recent years and identified through the prominent bibliographic databases. This systematic synthesis of research articles resulted in several analytical findings: XAI methods are mostly developed for safety-critical domains worldwide, deep learning and ensemble models are being exploited more than other types of AI/ML models, visual explanations are more acceptable to end-users and robust evaluation metrics are being developed to assess the quality of explanations. Research studies have been performed on the addition of explanations to widely used AI/ML models for expert users. However, more attention is required to generate explanations for general users from sensitive domains such as finance and the judicial system.

A Survey on Artificial Intelligence (AI) and eXplainable AI in Air Traffic Management: Current Trends and Development with Future Research Trajectory

Air Traffic Management (ATM) will be more complex in the coming decades due to the growth and increased complexity of aviation and has to be improved in order to maintain aviation safety. It is agreed that without significant improvement in this domain, the safety objectives defined by international organisations cannot be achieved and a risk of more incidents/accidents is envisaged. Nowadays, computer science plays a major role in data management and decisions made in ATM. Nonetheless, despite this, Artificial Intelligence (AI), which is one of the most researched topics in computer science, has not quite reached end users in ATM domain. In this paper, we analyse the state of the art with regards to usefulness of AI within aviation/ATM domain. It includes research work of the last decade of AI in ATM, the extraction of relevant trends and features, and the extraction of representative dimensions. We analysed how the general and ATM eXplainable Artificial Intelligence (XAI) works, analysing where and why XAI is needed, how it is currently provided, and the limitations, then synthesise the findings into a conceptual framework, named the DPP (Descriptive, Predictive, Prescriptive) model, and provide an example of its application in a scenario in 2030. It concludes that AI systems within ATM need further research for their acceptance by end-users. The development of appropriate XAI methods including the validation by appropriate authorities and end-users are key issues that needs to be addressed.

Geo-Context Aware Study of Vision-Based Autonomous Driving Models and Spatial Video Data

Vision-based deep learning (DL) methods have made great progress in learning autonomous driving models from large-scale crowd-sourced video datasets. They are trained to predict instantaneous driving behaviors from video data captured by on-vehicle cameras. In this paper, we develop a geo-context aware visualization system for the study of Autonomous Driving Model (ADM) predictions together with large-scale ADM video data. The visual study is seamlessly integrated with the geographical environment by combining DL model performance with geospatial visualization techniques. Model performance measures can be studied together with a set of geospatial attributes over map views. Users can also discover and compare prediction behaviors of multiple DL models in both city-wide and street-level analysis, together with road images and video contents. Therefore, the system provides a new visual exploration platform for DL model designers in autonomous driving. Use cases and domain expert evaluation show the utility and effectiveness of the visualization system.

Sibyl: Understanding and Addressing the Usability Challenges of Machine Learning In High-Stakes Decision Making

Machine learning (ML) is being applied to a diverse and ever-growing set of domains. In many cases, domain experts - who often have no expertise in ML or data science - are asked to use ML predictions to make high-stakes decisions. Multiple ML usability challenges can appear as result, such as lack of user trust in the model, inability to reconcile human-ML disagreement, and ethical concerns about oversimplification of complex problems to a single algorithm output. In this paper, we investigate the ML usability challenges that present in the domain of child welfare screening through a series of collaborations with child welfare screeners. Following the iterative design process between the ML scientists, visualization researchers, and domain experts (child screeners), we first identified four key ML challenges and honed in on one promising explainable ML technique to address them (local factor contributions). Then we implemented and evaluated our visual analytics tool, Sibyl, to increase the interpretability and interactivity of local factor contributions. The effectiveness of our tool is demonstrated by two formal user studies with 12 non-expert participants and 13 expert participants respectively. Valuable feedback was collected, from which we composed a list of design implications as a useful guideline for researchers who aim to develop an interpretable and interactive visualization tool for ML prediction models deployed for child welfare screeners and other similar domain experts.

THALIS: Human-Machine Analysis of Longitudinal Symptoms in Cancer Therapy

Although cancer patients survive years after oncologic therapy, they are plagued with long-lasting or permanent residual symptoms, whose severity, rate of development, and resolution after treatment vary largely between survivors. The analysis and interpretation of symptoms is complicated by their partial co-occurrence, variability across populations and across time, and, in the case of cancers that use radiotherapy, by further symptom dependency on the tumor location and prescribed treatment. We describe THALIS, an environment for visual analysis and knowledge discovery from cancer therapy symptom data, developed in close collaboration with oncology experts. Our approach leverages unsupervised machine learning methodology over cohorts of patients, and, in conjunction with custom visual encodings and interactions, provides context for new patients based on patients with similar diagnostic features and symptom evolution. We evaluate this approach on data collected from a cohort of head and neck cancer patients. Feedback from our clinician collaborators indicates that THALIS supports knowledge discovery beyond the limits of machines or humans alone, and that it serves as a valuable tool in both the clinic and symptom research.

Unbox the black-box for the medical explainable AI via multi-modal and multi-centre data fusion: A mini-review, two showcases and beyond

Explainable Artificial Intelligence (XAI) is an emerging research topic of machine learning aimed at unboxing how AI systems' black-box choices are made. This research field inspects the measures and models involved in decision-making and seeks solutions to explain them explicitly. Many of the machine learning algorithms cannot manifest how and why a decision has been cast. This is particularly true of the most popular deep neural network approaches currently in use. Consequently, our confidence in AI systems can be hindered by the lack of explainability in these black-box models. The XAI becomes more and more crucial for deep learning powered applications, especially for medical and healthcare studies, although in general these deep neural networks can return an arresting dividend in performance. The insufficient explainability and transparency in most existing AI systems can be one of the major reasons that successful implementation and integration of AI tools into routine clinical practice are uncommon. In this study, we first surveyed the current progress of XAI and in particular its advances in healthcare applications. We then introduced our solutions for XAI leveraging multi-modal and multi-centre data fusion, and subsequently validated in two showcases following real clinical scenarios. Comprehensive quantitative and qualitative analyses can prove the efficacy of our proposed XAI solutions, from which we can envisage successful applications in a broader range of clinical questions.

VBridge: Connecting the Dots Between Features and Data to Explain Healthcare Models

Machine learning (ML) is increasingly applied to Electronic Health Records (EHRs) to solve clinical prediction tasks. Although many ML models perform promisingly, issues with model transparency and interpretability limit their adoption in clinical practice. Directly using existing explainable ML techniques in clinical settings can be challenging. Through literature surveys and collaborations with six clinicians with an average of 17 years of clinical experience, we identified three key challenges, including clinicians' unfamiliarity with ML features, lack of contextual information, and the need for cohort-level evidence. Following an iterative design process, we further designed and developed VBridge, a visual analytics tool that seamlessly incorporates ML explanations into clinicians' decision-making workflow. The system includes a novel hierarchical display of contribution-based feature explanations and enriched interactions that connect the dots between ML features, explanations, and data. We demonstrated the effectiveness of VBridge through two case studies and expert interviews with four clinicians, showing that visually associating model explanations with patients' situational records can help clinicians better interpret and use model predictions when making clinician decisions. We further derived a list of design implications for developing future explainable ML tools to support clinical decision-making.