Tasks and Visualizations Used for Data Profiling: A Survey and Interview Study.

The use of good-quality data to inform decision making is entirely dependent on robust processes to ensure it is fit for purpose. Such processes vary between organisations, and between those tasked with designing and following them. In this paper we report on a survey of 53 data analysts from many industry sectors, 24 of whom also participated in in-depth interviews, about computational and visual methods for characterizing data and investigating data quality. The paper makes contributions in two key areas. The first is to data science fundamentals, because our lists of data profiling tasks and visualization techniques are more comprehensive than those published elsewhere. The second concerns the application question "what does good profiling look like to those who routinely perform it?," which we answer by highlighting the diversity of profiling tasks, unusual practice and exemplars of visualization, and recommendations about formalizing processes and creating rulebooks.

Faster than light (microscopy): superiority of digital pathology over microscopy for assessment of immunohistochemistry.

AIMS: Digital pathology offers the potential for significant benefits in diagnostic pathology, but currently the efficiency of slide viewing is a barrier to adoption. We hypothesised that presenting digital slides for simultaneous viewing of multiple sections of tissue for comparison, as in those with immunohistochemical panels, would allow pathologists to review cases more quickly. METHODS: Novel software was developed to view synchronised parallel tissue sections on a digital pathology workstation. Sixteen histopathologists reviewed three liver biopsy cases including an immunohistochemical panel using the digital microscope, and three different liver biopsy cases including an immunohistochemical panel using the light microscope. The order of cases and interface was fully counterbalanced. Time to diagnosis was recorded and mean times are presented as data approximated to a normalised distribution. RESULTS: Mean time to diagnosis was 4 min 3 s using the digital microscope and 5 min 24 s using the light microscope, saving 1 min 21 s (95% CI 16 s to 2 min 26 s; p=0.02), using the digital microscope. Overall normalised mean time to diagnosis was 85% on the digital pathology workstation compared with 115% on the microscope, a relative reduction of 26%. CONCLUSIONS: With appropriate interface design, it is quicker to review immunohistochemical slides using a digital microscope than the conventional light microscope, without incurring any major diagnostic errors. As digital pathology becomes more integrated with routine clinical workflow and pathologists increase their experience of the technology, it is anticipated that other tasks will also become more time-efficient.

The DynAIRx Project Protocol: Artificial Intelligence for dynamic prescribing optimisation and care integration in multimorbidity.

Background: Structured Medication Reviews (SMRs) are intended to help deliver the NHS Long Term Plan for medicines optimisation in people living with multiple long-term conditions and polypharmacy. It is challenging to gather the information needed for these reviews due to poor integration of health records across providers and there is little guidance on how to identify those patients most urgently requiring review. Objective: To extract information from scattered clinical records on how health and medications change over time, apply interpretable artificial intelligence (AI) approaches to predict risks of poor outcomes and overlay this information on care records to inform SMRs. We will pilot this approach in primary care prescribing audit and feedback systems, and co-design future medicines optimisation decision support systems. Design: DynAIRx will target potentially problematic polypharmacy in three key multimorbidity groups, namely, people with (a) mental and physical health problems, (b) four or more long-term conditions taking ten or more drugs and (c) older age and frailty. Structured clinical data will be drawn from integrated care records (general practice, hospital, and social care) covering an ∼11m population supplemented with Natural Language Processing (NLP) of unstructured clinical text. AI systems will be trained to identify patterns of conditions, medications, tests, and clinical contacts preceding adverse events in order to identify individuals who might benefit most from an SMR. Discussion: By implementing and evaluating an AI-augmented visualisation of care records in an existing prescribing audit and feedback system we will create a learning system for medicines optimisation, co-designed throughout with end-users and patients.

Using set visualisation to find and explain patterns of missing values: a case study with NHS hospital episode statistics data.

OBJECTIVES: Missing data is the most common data quality issue in electronic health records (EHRs). Missing data checks implemented in common analytical software are typically limited to counting the number of missing values in individual fields, but researchers and organisations also need to understand multifield missing data patterns to better inform advanced missing data strategies for which counts or numerical summaries are poorly suited. This study shows how set-based visualisation enables multifield missing data patterns to be discovered and investigated. DESIGN: Development and evaluation of interactive set visualisation techniques to find patterns of missing data and generate actionable insights. The visualisations comprised easily interpretable bar charts for sets, heatmaps for set intersections and histograms for distributions of both sets and intersections. SETTING AND PARTICIPANTS: Anonymised admitted patient care health records for National Health Service (NHS) hospitals and independent sector providers in England. The visualisation and data mining software was run over 16 million records and 86 fields in the dataset. RESULTS: The dataset contained 960 million missing values. Set visualisation bar charts showed how those values were distributed across the fields, including several fields that, unexpectedly, were not complete. Set intersection heatmaps revealed unexpected gaps in diagnosis, operation and date fields because diagnosis and operation fields were not filled up sequentially and some operations did not have corresponding dates. Information gain ratio and entropy calculations allowed us to identify the origin of each unexpected pattern, in terms of the values of other fields. CONCLUSIONS: Our findings show how set visualisation reveals important insights about multifield missing data patterns in large EHR datasets. The study revealed both rare and widespread data quality issues that were previously unknown, and allowed a particular part of a specific hospital to be pinpointed as the origin of rare issues that NHS Digital did not know exist.

The Effect of Alignment on People's Ability to Judge Event Sequence Similarity.

Event sequences are central to the analysis of data in domains that range from biology and health, to logfile analysis and people's everyday behavior. Many visualization tools have been created for such data, but people are error-prone when asked to judge the similarity of event sequences with basic presentation methods. This article describes an experiment that investigates whether local and global alignment techniques improve people's performance when judging sequence similarity. Participants were divided into three groups (basic versus local versus global alignment), and each participant judged the similarity of 180 sets of pseudo-randomly generated sequences. Each set comprised a target, a correct choice and a wrong choice. After training, the global alignment group was more accurate than the local alignment group (98 versus 93 percent correct), with the basic group getting 95 percent correct. Participants' response times were primarily affected by the number of event types, the similarity of sequences (measured by the Levenshtein distance) and the edit types (nine combinations of deletion, insertion and substitution). In summary, global alignment is superior and people's performance could be further improved by choosing alignment parameters that explicitly penalize sequence mismatches.

Hidden labour: the skilful work of clinical audit data collection and its implications for secondary use of data via integrated health IT.

BACKGROUND: Secondary use of data via integrated health information technology is fundamental to many healthcare policies and processes worldwide. However, repurposing data can be problematic and little research has been undertaken into the everyday practicalities of inter-system data sharing that helps explain why this is so, especially within (as opposed to between) organisations. In response, this article reports one of the most detailed empirical examinations undertaken to date of the work involved in repurposing healthcare data for National Clinical Audits. METHODS: Fifty-four semi-structured, qualitative interviews were carried out with staff in five English National Health Service hospitals about their audit work, including 20 staff involved substantively with audit data collection. In addition, ethnographic observations took place on wards, in 'back offices' and meetings (102 h). Findings were analysed thematically and synthesised in narratives. RESULTS: Although data were available within hospital applications for secondary use in some audit fields, which could, in theory, have been auto-populated, in practice staff regularly negotiated multiple, unintegrated systems to generate audit records. This work was complex and skilful, and involved cross-checking and double data entry, often using paper forms, to assure data quality and inform quality improvements. CONCLUSIONS: If technology is to facilitate the secondary use of healthcare data, the skilled but largely hidden labour of those who collect and recontextualise those data must be recognised. Their detailed understandings of what it takes to produce high quality data in specific contexts should inform the further development of integrated systems within organisations.

QualDash: Adaptable Generation of Visualisation Dashboards for Healthcare Quality Improvement.

Adapting dashboard design to different contexts of use is an open question in visualisation research. Dashboard designers often seek to strike a balance between dashboard adaptability and ease-of-use, and in hospitals challenges arise from the vast diversity of key metrics, data models and users involved at different organizational levels. In this design study, we present QualDash, a dashboard generation engine that allows for the dynamic configuration and deployment of visualisation dashboards for healthcare quality improvement (QI). We present a rigorous task analysis based on interviews with healthcare professionals, a co-design workshop and a series of one-on-one meetings with front line analysts. From these activities we define a metric card metaphor as a unit of visual analysis in healthcare QI, using this concept as a building block for generating highly adaptable dashboards, and leading to the design of a Metric Specification Structure (MSS). Each MSS is a JSON structure which enables dashboard authors to concisely configure unit-specific variants of a metric card, while offloading common patterns that are shared across cards to be preset by the engine. We reflect on deploying and iterating the design of OualDash in cardiology wards and pediatric intensive care units of five NHS hospitals. Finally, we report evaluation results that demonstrate the adaptability, ease-of-use and usefulness of QualDash in a real-world scenario.

How, in what contexts, and why do quality dashboards lead to improvements in care quality in acute hospitals? Protocol for a realist feasibility evaluation.

INTRODUCTION: National audits are used to monitor care quality and safety and are anticipated to reduce unexplained variations in quality by stimulating quality improvement (QI). However, variation within and between providers in the extent of engagement with national audits means that the potential for national audit data to inform QI is not being realised. This study will undertake a feasibility evaluation of QualDash, a quality dashboard designed to support clinical teams and managers to explore data from two national audits, the Myocardial Ischaemia National Audit Project (MINAP) and the Paediatric Intensive Care Audit Network (PICANet). METHODS AND ANALYSIS: Realist evaluation, which involves building, testing and refining theories of how an intervention works, provides an overall framework for this feasibility study. Realist hypotheses that describe how, in what contexts, and why QualDash is expected to provide benefit will be tested across five hospitals. A controlled interrupted time series analysis, using key MINAP and PICANet measures, will provide preliminary evidence of the impact of QualDash, while ethnographic observations and interviews over 12 months will provide initial insight into contexts and mechanisms that lead to those impacts. Feasibility outcomes include the extent to which MINAP and PICANet data are used, data completeness in the audits, and the extent to which participants perceive QualDash to be useful and express the intention to continue using it after the study period. ETHICS AND DISSEMINATION: The study has been approved by the University of Leeds School of Healthcare Research Ethics Committee. Study results will provide an initial understanding of how, in what contexts, and why quality dashboards lead to improvements in care quality. These will be disseminated to academic audiences, study participants, hospital IT departments and national audits. If the results show a trial is feasible, we will disseminate the QualDash software through a stepped wedge cluster randomised trial.

Using Dashboard Networks to Visualize Multiple Patient Histories: A Design Study on Post-Operative Prostate Cancer.

In this design study, we present a visualization technique that segments patients' histories instead of treating them as raw event sequences, aggregates the segments using criteria such as the whole history or treatment combinations, and then visualizes the aggregated segments as static dashboards that are arranged in a dashboard network to show longitudinal changes. The static dashboards were developed in nine iterations, to show 15 important attributes from the patients' histories. The final design was evaluated with five non-experts, five visualization experts and four medical experts, who successfully used it to gain an overview of a 2,000 patient dataset, and to make observations about longitudinal changes and differences between two cohorts. The research represents a step-change in the detail of large-scale data that may be successfully visualized using dashboards, and provides guidance about how the approach may be generalized.

Requirements for a quality dashboard: Lessons from National Clinical Audits.

Healthcare organizations worldwide use quality dashboards to provide feedback to clinical teams and managers, in order to monitor care quality and stimulate quality improvement. However, there is limited evidence regarding the impact of quality dashboards and audit and feedback research focuses on feedback to individual clinicians, rather than to clinical and managerial teams. Consequently, we know little about what features a quality dashboard needs in order to provide benefit. We conducted 54 interviews across five healthcare organizations in the National Health Service in England, interviewing personnel at different levels of the organization, to understand how national (UK) clinical audit data are used for quality improvement and factors that support or constrain use of these data. The findings, organized around the themes of choosing performance indicators, assessing performance, identifying causes, communicating from ward to board, and data quality, have implications for the design of quality dashboards, which we have translated into a series of requirements.

Visualization System Requirements for Data Processing Pipeline Design and Optimization.

The rising quantity and complexity of data creates a need to design and optimize data processing pipelines-the set of data processing steps, parameters and algorithms that perform operations on the data. Visualization can support this process but, although there are many examples of systems for visual parameter analysis, there remains a need to systematically assess users' requirements and match those requirements to exemplar visualization methods. This article presents a new characterization of the requirements for pipeline design and optimization. This characterization is based on both a review of the literature and first-hand assessment of eight application case studies. We also match these requirements with exemplar functionality provided by existing visualization tools. Thus, we provide end-users and visualization developers with a way of identifying functionality that addresses data processing problems in an application. We also identify seven future challenges for visualization research that are not met by the capabilities of today's systems.

Barriers and facilitators to the introduction of digital pathology for diagnostic work.

Cellular pathologists are doctors who diagnose disease by using a microscope to examine glass slides containing thin sections of human tissue. These slides can be digitised and viewed on a computer, promising benefits in both efficiency and safety. Despite this, uptake of digital pathology for diagnostic work has been slow, with use largely restricted to second opinions, education, and external quality assessment schemes. To understand the barriers and facilitators to the introduction of digital pathology, we have undertaken an interview study with nine consultant pathologists. Interviewees were able to identify a range of potential benefits of digital pathology, with a particular emphasis on easier access to slides. Amongst the barriers to use, a key concern was lack of familiarity, not only in terms of becoming familiar with the technology but learning how to adjust their diagnostic skills to this new medium. The findings emphasise the need to ensure adequate training and support and the potential benefit of allowing parallel use of glass slides and digital while pathologists are on the learning curve.

Effect of display resolution on time to diagnosis with virtual pathology slides in a systematic search task.

Performing diagnoses using virtual slides can take pathologists significantly longer than with glass slides, presenting a significant barrier to the use of virtual slides in routine practice. Given the benefits in pathology workflow efficiency and safety that virtual slides promise, it is important to understand reasons for this difference and identify opportunities for improvement. The effect of display resolution on time to diagnosis with virtual slides has not previously been explored. The aim of this study was to assess the effect of display resolution on time to diagnosis with virtual slides. Nine pathologists participated in a counterbalanced crossover study, viewing axillary lymph node slides on a microscope, a 23-in 2.3-megapixel single-screen display and a three-screen 11-megapixel display consisting of three 27-in displays. Time to diagnosis and time to first target were faster on the microscope than on the single and three-screen displays. There was no significant difference between the microscope and the three-screen display in time to first target, while the time taken on the single-screen display was significantly higher than that on the microscope. The results suggest that a digital pathology workstation with an increased number of pixels may make it easier to identify where cancer is located in the initial slide overview, enabling quick location of diagnostically relevant regions of interest. However, when a comprehensive, detailed search of a slide has to be made, increased resolution may not offer any additional benefit.

Diagnosis of major cancer resection specimens with virtual slides: impact of a novel digital pathology workstation.

Digital pathology promises a number of benefits in efficiency in surgical pathology, yet the longer time required to review a virtual slide than a glass slide currently represents a significant barrier to the routine use of digital pathology. We aimed to create a novel workstation that enables pathologists to view a case as quickly as on the conventional microscope. The Leeds Virtual Microscope (LVM) was evaluated using a mixed factorial experimental design. Twelve consultant pathologists took part, each viewing one long cancer case (12-25 slides) on the LVM and one on a conventional microscope. Total time taken and diagnostic confidence were similar for the microscope and LVM, as was the mean slide viewing time. On the LVM, participants spent a significantly greater proportion of the total task time viewing slides and revisited slides more often. The unique design of the LVM, enabling real-time rendering of virtual slides while providing users with a quick and intuitive way to navigate within and between slides, makes use of digital pathology in routine practice a realistic possibility. With further practice with the system, diagnostic efficiency on the LVM is likely to increase yet more.

Virtual reality microscope versus conventional microscope regarding time to diagnosis: an experimental study.

AIMS: To create and evaluate a virtual reality (VR) microscope that is as efficient as the conventional microscope, seeking to support the introduction of digital slides into routine practice. METHODS AND RESULTS: A VR microscope was designed and implemented by combining ultra-high-resolution displays with VR technology, techniques for fast interaction, and high usability. It was evaluated using a mixed factorial experimental design with technology and task as within-participant variables and grade of histopathologist as a between-participant variable. Time to diagnosis was similar for the conventional and VR microscopes. However, there was a significant difference in the mean magnification used between the two technologies, with participants working at a higher level of magnification on the VR microscope. CONCLUSIONS: The results suggest that, with the right technology, efficient use of digital pathology for routine practice is a realistic possibility. Further work is required to explore what magnification is required on the VR microscope for histopathologists to identify diagnostic features, and the effect on this of the digital slide production process.

Working at the microscope: analysis of the activities involved in diagnostic pathology.

AIMS: To study the current work practice of histopathologists to inform the design of digital microscopy systems. METHODS AND RESULTS: Four gastrointestinal histopathologists were video-recorded as they undertook their routine work. Analysis of the video data shows a range of activities beyond viewing slides involved in reporting a case. There is much overlapping of activities, supported by the 'eyes free' nature of the pathologists' interaction with the microscope. The order and timing of activities varies according to consultant. CONCLUSIONS: In order to support the work of pathologists adequately, digital microscopy systems need to provide support for a range of activities beyond viewing slides. Digital microscopy systems should support multitasking, while also providing flexibility so that pathologists can adapt their use of the technology to their own working patterns.

Visualization of parameter space for image analysis.

Image analysis algorithms are often highly parameterized and much human input is needed to optimize parameter settings. This incurs a time cost of up to several days. We analyze and characterize the conventional parameter optimization process for image analysis and formulate user requirements. With this as input, we propose a change in paradigm by optimizing parameters based on parameter sampling and interactive visual exploration. To save time and reduce memory load, users are only involved in the first step--initialization of sampling--and the last step--visual analysis of output. This helps users to more thoroughly explore the parameter space and produce higher quality results. We describe a custom sampling plug-in we developed for CellProfiler--a popular biomedical image analysis framework. Our main focus is the development of an interactive visualization technique that enables users to analyze the relationships between sampled input parameters and corresponding output. We implemented this in a prototype called Paramorama. It provides users with a visual overview of parameters and their sampled values. User-defined areas of interest are presented in a structured way that includes image-based output and a novel layout algorithm. To find optimal parameter settings, users can tag high- and low-quality results to refine their search. We include two case studies to illustrate the utility of this approach.

The effect of landmark and body-based sensory information on route knowledge.

Two experiments investigated the effects of landmarks and body-based information on route knowledge. Participants made four out-and-back journeys along a route, guided only on the first outward trip and with feedback every time an error was made. Experiment 1 used 3-D virtual environments (VEs) with a desktop monitor display, and participants were provided with no supplementary landmarks, only global landmarks, only local landmarks, or both global and local landmarks. Local landmarks significantly reduced the number of errors that participants made, but global landmarks did not. Experiment 2 used a head-mounted display; here, participants who physically walked through the VE (translational and rotational body-based information) made 36% fewer errors than did participants who traveled by physically turning but changing position using a joystick. Overall, the experiments showed that participants were less sure of where to turn than which way, and journey direction interacted with sensory information to affect the number and types of errors participants made.

Virtual reality Powerwall versus conventional microscope for viewing pathology slides: an experimental comparison.

AIMS: Virtual slides could replace the conventional microscope. However, it can take 60% longer to make a diagnosis with a virtual slide, due to the small display size and inadequate user interface of current systems. The aim was to create and test a virtual reality (VR) microscope using a Powerwall (a high-resolution array of 28 computer screens) for viewing virtual slides more efficiently. METHODS AND RESULTS: A controlled user experiment was performed to compare the Powerwall with the microscope for four types of task: (i) a simple diagnosis, (ii) a decision about a lymph node, (iii) finding small objects, (iv) scoring a tissue microarray. User behaviour was recorded by video and questionnaire. Time taken to perform all four tasks and diagnostic confidence were similar using the Powerwall and conventional microscope. CONCLUSIONS: After just a few minutes' familiarization, a VR Powerwall allowed tasks to be performed as quickly and confidently as a microscope. Behavioural data indicated how histopathologists should be trained to make the best use of the large display provided by the VR microscope. Together with the potential for further improvements in the design of the VR microscope, future virtual slide systems could out-perform conventional microscopes in histopathological diagnosis.