Simon Lin

Being able to predict motor vehicle crashes which are a major public health concern would greatly improve traffic safety. The prevalence of mobile sensing platforms now allows spatially and temporally rich driving data to be collected relatively easily. Research efforts have been devoted to predicting crashes from such data. This paper seeks in particular to assess the feasibility and performance of using aggregated fine geo-resolution vehicle telemetric data for crash risk prediction. We acquired vehicle telemetric data from Geotab Inc., which recorded the frequency of hard acceleration, hard braking, harsh cornering, and the average magnitude of those harsh events among its registered commercial vehicles for every 150 × 150 m² roadway segment within Columbus, Ohio between January and April 2018. We aggregated the data, obtained the crash history from the Ohio Police Accident Report, and leveraged three machine learning–based algorithms to predict the crash risk. The results suggest that aggregated vehicle telemetric data could provide acceptable predictions for crash risk at a roadway segment level. Our models’ predictive performances were further improved and maximized by including in the models both vehicle telemetric data and roadway geometric characteristics. Several factors, such as the aggregated count of hard accelerations and the presence of an intersection, were shown to be the factors that potentially made the greatest contribution to crash occurrence. We concluded that vehicle telemetric data could provide complementary and valuable information about crash likelihood monitoring, which may enable the police and city planners to implement proactive safety interventions. Yet the nature of traffic crashes is still complex and multi-dimensional.

BACKGROUND Background: Documenting clinical notes in Electronic Health Record (EHR) systems is a crucial job for physicians, yet characterizing provider documentation behavior is still difficult because of the lack of applications to do so. OBJECTIVE Objective: This study aims to analyze physician documentation behavior phenotypes at a multi-specialty clinic, as well as develop open-source software to make such profiling more easily available for other clinics to replicate. METHODS Methods: For this observational retrospective study, we analyzed data on physician note-taking behavior from EPIC Signal and applied the k-means clustering algorithm to identify and analyze different phenotypes in the data set. RESULTS Results: We developed an efficient tool and easy-to-follow protocol to profile provider documentation behavior, which identified four physician phenotype clusters within our data. Of these four, two were characterized by physicians with high note-taking efficiency, with one cluster having low documentation length and above-average efficiency, and the other having average documentation length and extremely high efficiency. Of the two remaining clusters, one was characterized by physicians with low note-taking efficiency, and the last consisted of low-volume providers. CONCLUSIONS Conclusions: Cluster analysis is a useful method to characterize different physician note-taking behaviors, which can be used to efficiently identify both physician champions and physicians in need of note-taking training.

Introduction: The COVID-19 pandemic has hastened the adoption of telehealth and the drastic shift to an unfamiliar process may impose significant impact to the quality-of-care delivery. Many providers are interested in understanding the quality of their telehealth services from the patients' experience. Materials and Methods: A telehealth patient satisfaction survey (TPSS) was developed by using an iterative stakeholder-centered design approach, incorporating elements from validated telemedicine and customer service survey instruments, and meeting the operational needs and constraints. A cross-sectional study design was employed to collect survey responses from patients and families of a large pediatric hospital. Finally, we performed exploratory factor analysis (EFA) to extract latent constructs and factor loadings of the survey items to further explain relationships. Results: A 22-item TPSS closely matched the existing in-person patient satisfaction survey and mapped to a revised SERVPERF conceptual model that was proposed by the interdisciplinary committee. Survey was implemented in the HIPAA-compliant online platform REDCap ^® with survey link embedded in an automated Epic MyChart (Verona, WI) visit follow-up message. In total, 2,394 survey responses were collected between July 7, 2020, and September 2, 2020. EFA revealed three constructs (with factor loadings >0.30): admission process, perceived quality of services, and telehealth satisfaction. Conclusions: We reported the development of TPSS that met the operational needs of compatibility with existing data and possible comparison to in-person survey. The survey is short and yet covers both the clinical experience and telehealth usability, with acceptable survey validity.

Introduction: The COVID-19 pandemic accelerated the adoption of telehealth as an alternative to in-person hospital visits. To understand the factors impacting the quality of telehealth services, there is a need for validated survey instruments and conceptual frameworks. The objective of this study is to validate a telehealth patient satisfaction survey by structural equation modeling (SEM) and determine the relationship between the factors in the proposed telehealth patient satisfaction model (TPSM). Methods: We conducted a cross-sectional survey of pediatric patients and families receiving care from a comprehensive pediatric hospital in the Midwest between September 2020 and January 2021. In total, 2,039 usable responses were collected. We used an SEM approach by performing confirmatory factor analysis with Diagonally Weighted Least Squares modeling and Partial Least Squares—Path Modeling to establish the structural validity and examined the relationships among the constructs of “Admission Process” (AP), “Perceived Quality of Service” (PQS), and “Telehealth Satisfaction” (TS). Results: Participants were predominantly White (75%) and English-speaking (95%) parents (85%) of patients (mean age of patients was 10.2 years old). The survey responses were collected from patients visiting 43 department specialties, whereas 50% were behavioral and occupational therapy patients. The structural model showed that the admission process (AP) had a strong positive impact on perceived quality of service (PQS) ( p = 0.67, t = 36.1, p < 0.001). The PQS had a strong positive impact on telehealth satisfaction (TS) ( p = 0.66, t = 31.8, p < 0.001). The AP had a low positive direct impact on TS ( p = 0.16, t = 7.46, p < 0.05). Overall, AP and PQS explained 61% variances ( R ² ) of TS. Conclusions: We validated a newly proposed TS assessment model by using SEM. The TPSM will inform researchers to better understand the influencing factors in TS and help health care systems to improve telehealth patient satisfaction through a validated model.

The adoption of electronic health records (EHR) has become universal during the past decade, which has afforded in-depth data-based research. By learning from the large amount of healthcare data, various data-driven models have been built to predict future events for different medical tasks, such as auto diagnosis and heart-attack prediction. Although EHR is abundant, the population that satisfies specific criteria for learning population-specific tasks is scarce, making it challenging to train data-hungry deep learning models. This study presents the Claim Pre-Training (Claim-PT) framework, a generic pre-training model that first trains on the entire pediatric claims dataset, followed by a discriminative fine-tuning on each population-specific task. The semantic meaning of medical events can be captured in the pre-training stage, and the effective knowledge transfer is completed through the task-aware fine-tuning stage. The fine-tuning process requires minimal parameter modification without changing the model architecture, which mitigates the data scarcity issue and helps train the deep learning model adequately on small patient cohorts. We conducted experiments on a real-world pediatric dataset with more than one million patient records. Experimental results on two downstream tasks demonstrated the effectiveness of our method: our general task-agnostic pre-training framework outperformed tailored task-specific models, achieving more than 10% higher in model performance as compared to baselines. In addition, our framework showed a potential to transfer learned knowledge from one institution to another, which may pave the way for future healthcare model pre-training across institutions.

Clinical question answering (QA) aims to automatically answer questions from medical professionals based on clinical texts. Studies show that neural QA models trained on one corpus may not generalize well to new clinical texts from a different institute or a different patient group, where large-scale QA pairs are not readily available for model retraining. To address this challenge, we propose a simple yet effective framework, CliniQG4QA, which leverages question generation (QG) to synthesize QA pairs on new clinical contexts and boosts QA models without requiring manual annotations. In order to generate diverse types of questions that are essential for training QA models, we further introduce a seq2seq-based question phrase prediction (QPP) module that can be used together with most existing QG models to diversify the generation. Our comprehensive experiment results show that the QA corpus generated by our framework can improve QA models on the new contexts (up to 8% absolute gain in terms of Exact Match), and that the QPP module plays a crucial role in achieving the gain.

We present a large, challenging dataset, COUGH, for COVID-19 FAQ retrieval. Similar to a standard FAQ dataset, COUGH consists of three parts: FAQ Bank, Query Bank and Relevance Set. The FAQ Bank contains ~16K FAQ items scraped from 55 credible websites (e.g., CDC and WHO). For evaluation, we introduce Query Bank and Relevance Set, where the former contains 1,236 human-paraphrased queries while the latter contains ~32 human-annotated FAQ items for each query. We analyze COUGH by testing different FAQ retrieval models built on top of BM25 and BERT, among which the best model achieves 48.8 under P@5, indicating a great challenge presented by COUGH and encouraging future research for further improvement. Our COUGH dataset is available at https://github.com/sunlab-osu/covid-faq.

Cardiovascular disease (CVD) is a major public health concern in the United States. In response to the federally sponsored Million Hearts Risk Check Challenge, a team of programmers, software developers, health-information technologists, and clinicians in an integrated healthcare system in Wisconsin collaborated to develop Heart Health Mobile^TM (HHM), designed to improve awareness of cardiovascular disease risk and promote risk factor control among users. This paper outlines the development processes and highlights key lessons learned for mobile health applications. An agile project management methodology was used to dedicate adequate resources and employ adaptive planning and iterative development processes with a self-organized, cross-functional team. The initial HHM iOS app was developed and tested, and after additional modifications, gamified and HTML 5 versions of the app were released. The development of an iOS app is low in cost and sustainable by a healthcare system. Future app modifications to enhance data security and link self-reported cardiovascular risk assessment data to patient medical records may improve performance, patient relevance, and clinician acceptance of HHM in the primary-care setting. Legal and institutional barriers regarding the capture and analyses of protected health information must be mitigated to fully capture, analyze, and report patient health outcomes for future studies.

Background: Children with special health care needs (CSHCN) require more than the usual care management and coordination efforts from caregivers and health care providers (HCPs). Health information and communication technologies can potentially facilitate these efforts to increase the quality of care received by CSHCN. Objective: In this study, we aim to assess the feasibility of a voice-enabled medical diary app (SpeakHealth) by investigating its potential use among caregivers and HCPs. Methods: Following a mixed methods approach, caregivers of CSHCN were interviewed (n=10) and surveyed (n=86) about their care management and communication technology use. Only interviewed participants were introduced to the SpeakHealth app prototype, and they tested the app during the interview session. In addition, we interviewed complex care HCPs (n=15) to understand their perception of the value of a home medical diary such as the SpeakHealth app. Quantitative data were analyzed using descriptive statistics and correlational analyses. Theoretical thematic analysis was used to analyze qualitative data. Results: The survey results indicated a positive attitude toward voice-enabled technology and features; however, there was no strong correlation among the measured items. The caregivers identified communication, information sharing, tracking medication, and appointments as fairly and highly important features of the app. Qualitative analysis revealed the following two overarching themes: enablers and barriers in care communication and enablers and barriers in communication technologies. The subthemes included parent roles, care communication technologies, and challenges. HCPs found the SpeakHealth app to be a promising tool for timely information collection that could be available for sharing information with the health system. Overall, the findings demonstrated a variety of needs and challenges for caregivers of CSHCN and opportunities for voice-enabled, interactive medical diary apps in care management and coordination. Caregivers fundamentally look for better information sharing and communication with HCPs. Health care and communication technologies can potentially improve care communication and coordination in addressing the patient and caregiver needs. Conclusions: The perspectives of caregivers and providers suggested both benefits and challenges in using the SpeakHealth app for medical note-taking and tracking health events at home. Our findings could inform researchers and developers about the potential development and use of a voice-enabled medical diary app.

Background: Perinatal health care is critically important for maternal health outcomes in infants. The United States fares considerably worse than comparable countries for maternal and infant mortality rates. As such, alternative models of care or engagement are warranted. Ubiquitous digital devices and increased use of digital health tools have the potential to extend the reach to women and infants in their everyday lives and make a positive impact on their health outcomes. As voice technology becomes more mainstream, research is prudent to establish evidence-based practice on how to best leverage voice technology to promote maternal-infant health. Objective: The aim of this study is to assess the feasibility of using voice technology to support perinatal health and infant care practices. Methods: Perinatal women were recruited from a large Midwest Children’s Hospital via hospital email announcements and word of mouth. Owing to the technical aspects of the intervention, participants were required to speak English and use an iPhone. Demographics, patterns of technology use, and technology use specific to perinatal health or self-care practices were assessed at baseline. Next, participants were onboarded and asked to use the intervention, Self-Management Intervention–Life Essentials (SMILE), over the course of 2 weeks. SMILE provided users with perinatal health content delivered through mini podcasts (ranging from 3 to 8 minutes in duration). After each podcast, SMILE prompted users to provide immediate verbal feedback to the content. An exit interview was conducted with participants to gather feedback on the intervention and further explore participants’ perceptions of voice technology as a means to support perinatal health in the future. Results: In total, 19 pregnant women (17 to 36 weeks pregnant) were consented. Themes identified as important for perinatal health information include establishing routines, expected norms, and realistic expectations and providing key takeaways. Themes identified as important for voice interaction include customization and user preferences, privacy, family and friends, and context and convenience. Qualitative analysis suggested that perinatal health promotion content delivered by voice should be accurate and succinctly delivered and highlight key takeaways. Perinatal health interventions that use voice should provide users with the ability to customize the intervention but also provide opportunities to engage family members, particularly spouses. As a number of women multitasked while the intervention was being deployed, future interventions should leverage the convenience of voice technology while also balancing the influence of user context (eg, timing or ability to listen or talk versus nonvoice interaction with the system). Conclusions: Our findings demonstrate the short-term feasibility of disseminating evidence-based perinatal support via podcasts and curate voice-captured data from perinatal women. However, key areas of improvement have been identified specifically for perinatal interventions leveraging voice technology. Findings contribute to future content, design, and delivery considerations of perinatal digital health interventions.