AI Quality Standards in Health Care: Rapid Umbrella Review.

BACKGROUND: In recent years, there has been an upwelling of artificial intelligence (AI) studies in the health care literature. During this period, there has been an increasing number of proposed standards to evaluate the quality of health care AI studies. OBJECTIVE: This rapid umbrella review examines the use of AI quality standards in a sample of health care AI systematic review articles published over a 36-month period. METHODS: We used a modified version of the Joanna Briggs Institute umbrella review method. Our rapid approach was informed by the practical guide by Tricco and colleagues for conducting rapid reviews. Our search was focused on the MEDLINE database supplemented with Google Scholar. The inclusion criteria were English-language systematic reviews regardless of review type, with mention of AI and health in the abstract, published during a 36-month period. For the synthesis, we summarized the AI quality standards used and issues noted in these reviews drawing on a set of published health care AI standards, harmonized the terms used, and offered guidance to improve the quality of future health care AI studies. RESULTS: We selected 33 review articles published between 2020 and 2022 in our synthesis. The reviews covered a wide range of objectives, topics, settings, designs, and results. Over 60 AI approaches across different domains were identified with varying levels of detail spanning different AI life cycle stages, making comparisons difficult. Health care AI quality standards were applied in only 39% (13/33) of the reviews and in 14% (25/178) of the original studies from the reviews examined, mostly to appraise their methodological or reporting quality. Only a handful mentioned the transparency, explainability, trustworthiness, ethics, and privacy aspects. A total of 23 AI quality standard-related issues were identified in the reviews. There was a recognized need to standardize the planning, conduct, and reporting of health care AI studies and address their broader societal, ethical, and regulatory implications. CONCLUSIONS: Despite the growing number of AI standards to assess the quality of health care AI studies, they are seldom applied in practice. With increasing desire to adopt AI in different health topics, domains, and settings, practitioners and researchers must stay abreast of and adapt to the evolving landscape of health care AI quality standards and apply these standards to improve the quality of their AI studies.

Using Decision Trees as an Expert System for Clinical Decision Support for COVID-19.

COVID-19 has impacted billions of people and health care systems globally. However, there is currently no publicly available chatbot for patients and care providers to determine the potential severity of a COVID-19 infection or the possible biological system responses and comorbidities that can contribute to the development of severe cases of COVID-19. This preliminary investigation assesses this lack of a COVID-19 case-by-case chatbot into consideration when building a decision tree with binary classification that was stratified by age and body system, viral infection, comorbidities, and any manifestations. After reviewing the relevant literature, a decision tree was constructed using a suite of tools to build a stratified framework for a chatbot application and interaction with users. A total of 212 nodes were established that were stratified from lung to heart conditions along body systems, medical conditions, comorbidities, and relevant manifestations described in the literature. This resulted in a possible 63,360 scenarios, offering a method toward understanding the data needed to validate the decision tree and highlighting the complicated nature of severe cases of COVID-19. The decision tree confirms that stratification of the viral infection with the body system while incorporating comorbidities and manifestations strengthens the framework. Despite limitations of a viable clinical decision tree for COVID-19 cases, this prototype application provides insight into the type of data required for effective decision support.

A Hadoop/MapReduce Based Platform for Supporting Health Big Data Analytics.

In this paper, we report our practical experience in designing and implementing a platform with Hadoop/MapReduce framework for supporting health Big Data Analytics. Three billion of emulated health raw data was constructed and cross-referenced with data profiles and metadata based on existing health data at the Island Health Authority, BC, Canada. The patient data was stored over a Hadoop Distributed File System to simulate a presentation of an entire health authority's information system. Then, a High Performance Computing platform called WestGrid was used to benchmark the performance of the platform via several data query tests. The work is important as very few implementation studies existed that tested a BDA platform applied to patient data of a health authority system.

Using Distributed Data over HBase in Big Data Analytics Platform for Clinical Services.

Big data analytics (BDA) is important to reduce healthcare costs. However, there are many challenges of data aggregation, maintenance, integration, translation, analysis, and security/privacy. The study objective to establish an interactive BDA platform with simulated patient data using open-source software technologies was achieved by construction of a platform framework with Hadoop Distributed File System (HDFS) using HBase (key-value NoSQL database). Distributed data structures were generated from benchmarked hospital-specific metadata of nine billion patient records. At optimized iteration, HDFS ingestion of HFiles to HBase store files revealed sustained availability over hundreds of iterations; however, to complete MapReduce to HBase required a week (for 10 TB) and a month for three billion (30 TB) indexed patient records, respectively. Found inconsistencies of MapReduce limited the capacity to generate and replicate data efficiently. Apache Spark and Drill showed high performance with high usability for technical support but poor usability for clinical services. Hospital system based on patient-centric data was challenging in using HBase, whereby not all data profiles were fully integrated with the complex patient-to-hospital relationships. However, we recommend using HBase to achieve secured patient data while querying entire hospital volumes in a simplified clinical event model across clinical services.

Usability testing of Avoiding Diabetes Thru Action Plan Targeting (ADAPT) decision support for integrating care-based counseling of pre-diabetes in an electronic health record.

PURPOSE: Usability testing can be used to evaluate human-computer interaction (HCI) and communication in shared decision making (SDM) for patient-provider behavioral change and behavioral contracting. Traditional evaluations of usability using scripted or mock patient scenarios with think-aloud protocol analysis provide a way to identify HCI issues. In this paper we describe the application of these methods in the evaluation of the Avoiding Diabetes Thru Action Plan Targeting (ADAPT) tool, and test the usability of the tool to support the ADAPT framework for integrated care counseling of pre-diabetes. The think-aloud protocol analysis typically does not provide an assessment of how patient-provider interactions are effected in "live" clinical workflow or whether a tool is successful. Therefore, "Near-live" clinical simulations involving applied simulation methods were used to compliment the think-aloud results. This complementary usability technique was used to test the end-user HCI and tool performance by more closely mimicking the clinical workflow and capturing interaction sequences along with assessing the functionality of computer module prototypes on clinician workflow. We expected this method to further complement and provide different usability findings as compared to think-aloud analysis. Together, this mixed method evaluation provided comprehensive and realistic feedback for iterative refinement of the ADAPT system prior to implementation. METHODS: The study employed two phases of testing of a new interactive ADAPT tool that embedded an evidence-based shared goal setting component into primary care workflow for dealing with pre-diabetes counseling within a commercial physician office electronic health record (EHR). Phase I applied usability testing that involved "think-aloud" protocol analysis of eight primary care providers interacting with several scripted clinical scenarios. Phase II used "near-live" clinical simulations of five providers interacting with standardized trained patient actors enacting the clinical scenario of counseling for pre-diabetes, each of whom had a pedometer that recorded the number of steps taken over a week. In both phases, all sessions were audio-taped and motion screen-capture software was activated for onscreen recordings. Transcripts were coded using iterative qualitative content analysis methods. RESULTS: In Phase I, the impact of the components and layout of ADAPT on user's Navigation, Understandability, and Workflow were associated with the largest volume of negative comments (i.e. approximately 80% of end-user commentary), while Usability and Content of ADAPT were representative of more positive than negative user commentary. The heuristic category of Usability had a positive-to-negative comment ratio of 2.1, reflecting positive perception of the usability of the tool, its functionality, and overall co-productive utilization of ADAPT. However, there were mixed perceptions about content (i.e., how the information was displayed, organized and described in the tool). In Phase II, the duration of patient encounters was approximately 10 min with all of the Patient Instructions (prescriptions) and behavioral contracting being activated at the end of each visit. Upon activation, providers accepted the pathway prescribed by the tool 100% of the time and completed all the fields in the tool in the simulation cases. Only 14% of encounter time was spent using the functionality of the ADAPT tool in terms of keystrokes and entering relevant data. The rest of the time was spent on communication and dialog to populate the patient instructions. In all cases, the interaction sequence of reviewing and discussing exercise and diet of the patient was linked to the functionality of the ADAPT tool in terms of monitoring, response-efficacy, self-efficacy, and negotiation in the patient-provider dialog. There was a change from one-way dialog to two-way dialog and negotiation that ended in a behavioral contract. This change demonstrated the tool's sequence, which supported recording current exercise and diet followed by a diet and exercise goal setting procedure to reduce the risk of diabetes onset. CONCLUSIONS: This study demonstrated that "think-aloud" protocol analysis with "near-live" clinical simulations provided a successful usability evaluation of a new primary care pre-diabetes shared goal setting tool. Each phase of the study provided complementary observations on problems with the new onscreen tool and was used to show the influence of the ADAPT framework on the usability, workflow integration, and communication between the patient and provider. The think-aloud tests with the provider showed the tool can be used according to the ADAPT framework (exercise-to-diet behavior change and tool utilization), while the clinical simulations revealed the ADAPT framework to realistically support patient-provider communication to obtain behavioral change contract. SDM interactions and mechanisms affecting protocol-based care can be more completely captured by combining "near-live" clinical simulations with traditional "think-aloud analysis" which augments clinician utilization. More analysis is required to verify if the rich communication actions found in Phase II compliment clinical workflows.

Integrating usability testing and think-aloud protocol analysis with "near-live" clinical simulations in evaluating clinical decision support.

PURPOSE: Usability evaluations can improve the usability and workflow integration of clinical decision support (CDS). Traditional usability testing using scripted scenarios with think-aloud protocol analysis provide a useful but incomplete assessment of how new CDS tools interact with users and clinical workflow. "Near-live" clinical simulations are a newer usability evaluation tool that more closely mimics clinical workflow and that allows for a complementary evaluation of CDS usability as well as impact on workflow. METHODS: This study employed two phases of testing a new CDS tool that embedded clinical prediction rules (an evidence-based medicine tool) into primary care workflow within a commercial electronic health record. Phase I applied usability testing involving "think-aloud" protocol analysis of 8 primary care providers encountering several scripted clinical scenarios. Phase II used "near-live" clinical simulations of 8 providers interacting with video clips of standardized trained patient actors enacting the clinical scenario. In both phases, all sessions were audiotaped and had screen-capture software activated for onscreen recordings. Transcripts were coded using qualitative analysis methods. RESULTS: In Phase I, the impact of the CDS on navigation and workflow were associated with the largest volume of negative comments (accounting for over 90% of user raised issues) while the overall usability and the content of the CDS were associated with the most positive comments. However, usability had a positive-to-negative comment ratio of only 0.93 reflecting mixed perceptions about the usability of the CDS. In Phase II, the duration of encounters with simulated patients was approximately 12 min with 71% of the clinical prediction rules being activated after half of the visit had already elapsed. Upon activation, providers accepted the CDS tool pathway 82% of times offered and completed all of its elements in 53% of all simulation cases. Only 12.2% of encounter time was spent using the CDS tool. Two predominant clinical workflows, accounting for 75% of all cases simulations, were identified that characterized the sequence of provider interactions with the CDS. These workflows demonstrated a significant variation in temporal sequence of potential activation of the CDS. CONCLUSIONS: This study successfully combined "think-aloud" protocol analysis with "near-live" clinical simulations in a usability evaluation of a new primary care CDS tool. Each phase of the study provided complementary observations on problems with the new onscreen tool and was used to refine both its usability and workflow integration. Synergistic use of "think-aloud" protocol analysis and "near-live" clinical simulations provide a robust assessment of how CDS tools would interact in live clinical environments and allows for enhanced early redesign to augment clinician utilization. The findings suggest the importance of using complementary testing methods before releasing CDS for live use.

Plant community responses to 5 years of simulated climate change in meadow and heath ecosystems at a subarctic-alpine site.

Climate change was simulated by increasing temperature and nutrient availability in an alpine landscape. We conducted a field experiment of BACI-design (before/after control/impact) running for five seasons in two alpine communities (heath and meadow) with the factors temperature (increase of ca. 1.5-3.0 degrees C) and nutrients (5 g N, 5 g P per m(2)) in a fully factorial design in northern Swedish Lapland. The response variables were abundances of plant species and functional types. Plant community responses to the experimental perturbations were investigated, and the responses of plant functional types were examined in comparison to responses at the species level. Nutrient addition, exclusively and in combination with enhanced temperature increase, exerted the most pronounced responses at the species-specific and community levels. The main responses to nutrient addition were increases in graminoids and forbs, whereas deciduous shrubs, evergreen shrubs, bryophytes, and lichens decreased. The two plant communities of heath or meadow showed different vegetation responses to the environmental treatments despite the fact that both communities were located on the same subarctic-alpine site. Furthermore, we showed that the abundance of forbs increased in response to the combined treatment of temperature and nutrient addition in the meadow plant community. Within a single-plant functional type, most species responded similarly to the enhanced treatments although there were exceptions, particularly in the moss and lichen functional types. Plant community structure showed BACI responses in that vegetation dominance relationships in the existing plant functional types changed to varying degrees in all plots, including control plots. Betula nana and lichens increased in the temperature-increased enhancements and in control plots in the heath plant community during the treatment period. The increases in control plots were probably a response to the observed warming during the treatment period in the region.