Initial Experience with a COVID-19 Screening Chatbot Before Radiology Appointments.

Guidelines for COVID-19 issued by the Centers for Disease Control and Prevention prompted state and local governments to mandate safety measures for screening high-risk patient populations and for institutions to look for ways to limit human contact when possible. The aim of this study was to determine the feasibility of an automated communication system (chatbot) for COVID-19 screening before patients' radiology appointments and to describe patient experiences with the chatbot. We developed a chatbot for COVID-19 screening before outpatient radiology examination appointments and tested it in a pilot study from July 6 to August 31, 2020. The chatbot assessed the presence of any symptoms, exposure, and recent testing. User experience was assessed via a questionnaire based on a 5-point Likert scale. Multivariable logistic regression was performed to predict response rate. The chatbot COVID-19 screening SMS message was sent to 4687 patients. Of these patients, 2722 (58.1%) responded. Of the respondents, 46 (1.7%) reported COVID-19 symptoms; 34 (1.2%) had COVID-19 tests scheduled or pending. Of the 1965 nonresponders, authentication failed for 174 (8.8%), 1496 (76.1%) did not engage with the SMS message, and 251 (12.8%) timed out of the chatbot. The mean rating for the chatbot experience was 4.6. In a multivariable logistic regression model predicting response rate, English written-language preference independently predicted response (odds ratio, 2.71 [95% CI, 1.77-2.77]; P = .007). Age (P = 0.57) and sex (P = 0.51) did not predict response rate. SMS-based COVID-19 screening before scheduled radiology appointments was feasible. English written-language preference (not age or sex) was associated with higher response rate.

The impact of blood pressure on the risk of major bleeding complication after renal transplant biopsy.

PURPOSE: To assess the impact of elevated blood pressure on the rate of major hemorrhagic complication after renal transplant biopsy. METHODS: Pre-procedural systolic (SBP), diastolic (SBP), and mean arterial (MAP) blood pressure for consecutive patients undergoing US-guided renal transplant biopsies from 08/01/2015 to 7/31/2017 were retrospectively recorded. Patients who had a major bleeding complication were identified. The risk of complication as a function of SBP, DBP, and MAP was statistically analyzed, with significance set at p < 0.05. RESULTS: Of 1689 biopsies, there were 10 bleeding complications (10/1689, 0.59%). There was no statistically significant difference between biopsies with complication compared to those without complication based on SBP (p = 0.351), DBP (p = 0.088), or MAP (p = 0.132). Using risk dichotomization criteria, the odds ratio for hemorrhagic complication when the patient had SBP ≥ 180 mmHg and DBP ≥ 95 mmHg was 75.63 (95% CI 6.87-516.8, p = 0.002). CONCLUSION: The rate of hemorrhagic complication from renal transplant biopsy is low, and there is no statistically significant threshold for increased biopsy risk based on SBP, DBP, or MAP alone. The risk of complication was significantly higher only when both the SBP is ≥ 180 mmHg and DBP is ≥ 95 mmHg.

Clinical longitudinal evaluation of COVID-19 patients and prediction of organ-specific recovery using artificial intelligence.

Within COVID-19 there is an urgent unmet need to predict at the time of hospital admission which COVID-19 patients will recover from the disease, and how fast they recover to deliver personalized treatments and to properly allocate hospital resources so that healthcare systems do not become overwhelmed. To this end, we have combined clinically salient CT imaging data synergistically with laboratory testing data in an integrative machine learning model to predict organ-specific recovery of patients from COVID-19. We trained and validated our model in 285 patients on each separate major organ system impacted by COVID-19 including the renal, pulmonary, immune, cardiac, and hepatic systems. To greatly enhance the speed and utility of our model, we applied an artificial intelligence method to segment and classify regions on CT imaging, from which interpretable data could be directly fed into the predictive machine learning model for overall recovery. Across all organ systems we achieved validation set area under the receiver operator characteristic curve (AUC) values for organ-specific recovery ranging from 0.80 to 0.89, and significant overall recovery prediction in Kaplan-Meier analyses. This demonstrates that the synergistic use of an artificial intelligence (AI) framework applied to CT lung imaging and a machine learning model that integrates laboratory test data with imaging data can accurately predict the overall recovery of COVID-19 patients from baseline characteristics.