Roses Have Thorns: Understanding the Downside of Oncological Care Delivery Through Visual Analytics and Sequential Rule Mining

Personalized head and neck cancer therapeutics have greatly improved survival rates for patients, but are often leading to understudied long-lasting symptoms which affect quality of life. Sequential rule mining (SRM) is a promising unsupervised machine learning method for predicting longitudinal patterns in temporal data which, however, can output many repetitive patterns that are difficult to interpret without the assistance of visual analytics. We present a data-driven, human-machine analysis visual system developed in collaboration with SRM model builders in cancer symptom research, which facilitates mechanistic knowledge discovery in large scale, multivariate cohort symptom data. Our system supports multivariate predictive modeling of post-treatment symptoms based on during-treatment symptoms. It supports this goal through an SRM, clustering, and aggregation back end, and a custom front end to help develop and tune the predictive models. The system also explains the resulting predictions in the context of therapeutic decisions typical in personalized care delivery. We evaluate the resulting models and system with an interdisciplinary group of modelers and head and neck oncology researchers. The results demonstrate that our system effectively supports clinical and symptom research.

Spatial and Temporal Data Visualisation for Mass Dissemination: Advances in the Era of COVID-19

The COVID-19 pandemic is the first major pandemic of the digital age and has been characterised by unprecedented public consumption of spatial and temporal disease data, which can enable greater transparency and accountability of governments to the public for their public health decisions. A variety of state and non-state actors have collated and presented maps, charts, and plots of data related to the pandemic in both static and dynamic formats. In particular, there has been a proliferation of online dashboards presenting data related to the pandemic. The sources and types of information displayed have evolved rapidly during the pandemic, with a general trend towards providing more specialised information pertinent to specific aspects of epidemiology or disease control, as opposed simply to disease and death notifications. Limited evaluation of the quality of COVID-19 data visualisation tools has been conducted and significant effort now needs to be spent on standardisation and quality improvement of national and international data visualisation systems including developing common indicators, data quality assurance mechanisms and visualisation approaches, and building compatible electronic systems for data collection and sharing. The increasing availability of disease data for public consumption presents challenges and opportunities for government, media organisations, academic research institutions, and the general public. A key challenge is ensuring consistency and effectiveness of public health messaging to ensure a coordinated response and public trust in intervention strategies. Capitalising on opportunities for greater government accountability for public health decision-making, and more effective mobilisation of public health interventions, is predicated on the provision of accurate and timely information.

Modelling temporal data in knowledge graphs: a systematic review protocol

Background: The benefits of having high-quality healthcare data are well established. However, high-dimensionality and irregularity of healthcare data pose challenges in their management. Knowledge graphs have gained increasing popularity in many domains, as a method for representing data to overcome such challenges. However, little is known about their suitability for use with healthcare data. One important factor in representing data is "time". Data with time related attributes are considered, temporal data. Temporal data are frequently observed in healthcare and the management of rapidly changing patient data is an ongoing challenge. Traditionally, data models have focused on presenting static data and do not account for temporal data. Temporal data models ensure time consistency in data models and assist analysing the history of data and predicting the future trends in data. Knowledge graphs can include temporal data models and are therefore of interest to the field of healthcare data management. As such, the herein aim is to outline a protocol for an inter-disciplinary systematic review of approaches, applications and challenges in modelling temporal data in knowledge graphs so that we can inform the application of knowledge graphs to healthcare data. Method: The research questions is, what are the existing approaches in modelling temporal data in RDF based knowledge graphs. Two sub-questions on applications, and challenges will also be evaluated. ACM digital library, IEEE Xplore and Scopus will be searched for this review. The search will be limited to peer-reviewed literature referring to knowledge graphs based on Resource Description Framework (RDF). A narrative synthesis of the papers will be conducted. Conclusion: The findings of this systematic review will be useful for data engineers to better represent data and perform analytics through temporal data modelling. They can be applied in the context of healthcare data and the current challenges faced in managing rapidly changing patient data.

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.

TSI-GNN: Extending Graph Neural Networks to Handle Missing Data in Temporal Settings

We present a novel approach for imputing missing data that incorporates temporal information into bipartite graphs through an extension of graph representation learning. Missing data is abundant in several domains, particularly when observations are made over time. Most imputation methods make strong assumptions about the distribution of the data. While novel methods may relax some assumptions, they may not consider temporality. Moreover, when such methods are extended to handle time, they may not generalize without retraining. We propose using a joint bipartite graph approach to incorporate temporal sequence information. Specifically, the observation nodes and edges with temporal information are used in message passing to learn node and edge embeddings and to inform the imputation task. Our proposed method, temporal setting imputation using graph neural networks (TSI-GNN), captures sequence information that can then be used within an aggregation function of a graph neural network. To the best of our knowledge, this is the first effort to use a joint bipartite graph approach that captures sequence information to handle missing data. We use several benchmark datasets to test the performance of our method against a variety of conditions, comparing to both classic and contemporary methods. We further provide insight to manage the size of the generated TSI-GNN model. Through our analysis we show that incorporating temporal information into a bipartite graph improves the representation at the 30% and 60% missing rate, specifically when using a nonlinear model for downstream prediction tasks in regularly sampled datasets and is competitive with existing temporal methods under different scenarios.

Boolean Feedforward Neural Network Modeling of Molecular Regulatory Networks for Cellular State Conversion

The molecular regulatory network (MRN) within a cell determines cellular states and transitions between them. Thus, modeling of MRNs is crucial, but this usually requires extensive analysis of time-series measurements, which is extremely difficult to obtain from biological experiments. However, single-cell measurement data such as single-cell RNA-sequencing databases have recently provided a new insight into resolving this problem by ordering thousands of cells in pseudo-time according to their differential gene expressions. Neural network modeling can be employed by using temporal data as learning data. In contrast, Boolean network modeling of MRNs has a growing interest, as it is a parameter-free logical modeling and thereby robust to noisy data while still capturing essential dynamics of biological networks. In this study, we propose a Boolean feedforward neural network (FFN) modeling by combining neural network and Boolean network modeling approach to reconstruct a practical and useful MRN model from large temporal data. Furthermore, analyzing the reconstructed MRN model can enable us to identify control targets for potential cellular state conversion. Here, we show the usefulness of Boolean FFN modeling by demonstrating its applicability through a toy model and biological networks.

Temporal scatterplots

Visualizing high-dimensional data on a 2D canvas is generally challenging. It becomes significantly more difficult when multiple time-steps are to be presented, as the visual clutter quickly increases. Moreover, the challenge to perceive the significant temporal evolution is even greater. In this paper, we present a method to plot temporal high-dimensional data in a static scatterplot; it uses the established PCA technique to project data from multiple time-steps. The key idea is to extend each individual displacement prior to applying PCA, so as to skew the projection process, and to set a projection plane that balances the directions of temporal change and spatial variance. We present numerous examples and various visual cues to highlight the data trajectories, and demonstrate the effectiveness of the method for visualizing temporal data.

The Effects of Quantitative Trait Architecture on Detection Power in Short-Term Artificial Selection Experiments

Evolve and resequence (E&R) experiments, in which artificial selection is imposed on organisms in a controlled environment, are becoming an increasingly accessible tool for studying the genetic basis of adaptation. Previous work has assessed how different experimental design parameters affect the power to detect the quantitative trait loci (QTL) that underlie adaptive responses in such experiments, but so far there has been little exploration of how this power varies with the genetic architecture of the evolving traits. In this study, we use forward simulation to build a more realistic model of an E&R experiment in which a quantitative polygenic trait experiences a short, but strong, episode of truncation selection. We study the expected power for QTL detection in such an experiment and how this power is influenced by different aspects of trait architecture, including the number of QTL affecting the trait, their starting frequencies, effect sizes, clustering along a chromosome, dominance, and epistasis patterns. We show that all of these parameters can affect allele frequency dynamics at the QTL and linked loci in complex and often unintuitive ways, and thus influence our power to detect them. One consequence of this is that existing detection methods based on models of independent selective sweeps at individual QTL often have lower detection power than a simple measurement of allele frequency differences before and after selection. Our findings highlight the importance of taking trait architecture into account when designing and interpreting studies of molecular adaptation with temporal data. We provide a customizable modeling framework that will enable researchers to easily simulate E&R experiments with different trait architectures and parameters tuned to their specific study system, allowing for assessment of expected detection power and optimization of experimental design.

Linking temporal medical records using non-protected health information data

Modern medical research relies on multi-institutional collaborations which enhance the knowledge discovery and data reuse. While these collaborations allow researchers to perform analytics otherwise impossible on individual datasets, they often pose significant challenges in the data integration process. Due to the lack of a unique identifier, data integration solutions often have to rely on patient's protected health information (PHI). In many situations, such information cannot leave the institutions or must be strictly protected. Furthermore, the presence of noisy values for these attributes may result in poor overall utility. While much research has been done to address these challenges, most of the current solutions are designed for a static setting without considering the temporal information of the data (e.g. EHR). In this work, we propose a novel approach that uses non-PHI for linking patient longitudinal data. Specifically, our technique captures the diagnosis dependencies using patterns which are shown to provide important indications for linking patient records. Our solution can be used as a standalone technique to perform temporal record linkage using non-protected health information data or it can be combined with Privacy Preserving Record Linkage solutions (PPRL) when protected health information is available. In this case, our approach can solve ambiguities in results. Experimental evaluations on real datasets demonstrate the effectiveness of our technique.

Using spatial, temporal and evidence-status data to improve ballistic imaging performance

Firearms identification imaging systems help solve crimes by comparing newly acquired images of cartridge casings or bullets to a database of images obtained from past crime scenes. We formulate an optimization problem that bases its matching decisions not only on the similarity between pairs of images, but also on the time and spatial location of each new acquisition and each database entry. The objective is to maximize the detection probability subject to a constraint on the false positive rate. We use data on all cartridge casings matches detected in Israel during 2006-2008 to estimate most of the model parameters. We estimate matching accuracy from two different studies and predict that the optimal use of extraneous information would increase the detection probability from 0.931 to 0.987 and from 0.707 to 0.844, respectively. These improvements are achieved by favoring pairs of images that are closer together in space and time.