Albert T. Young

Artificial intelligence (AI) promises to change health care, with some studies showing proof of concept of a provider-level performance in various medical specialties. However, there are many barriers to implementing AI, including patient acceptance and understanding of AI. Patients' attitudes toward AI are not well understood. We systematically reviewed the literature on patient and general public attitudes toward clinical AI (either hypothetical or realised), including quantitative, qualitative, and mixed methods original research articles. We searched biomedical and computational databases from Jan 1, 2000, to Sept 28, 2020, and screened 2590 articles, 23 of which met our inclusion criteria. Studies were heterogeneous regarding the study population, study design, and the field and type of AI under study. Six (26%) studies assessed currently available or soon-to-be available AI tools, whereas 17 (74%) assessed hypothetical or broadly defined AI. The quality of the methods of these studies was mixed, with a frequent issue of selection bias. Overall, patients and the general public conveyed positive attitudes toward AI but had many reservations and preferred human supervision. We summarise our findings in six themes: AI concept, AI acceptability, AI relationship with humans, AI development and implementation, AI strengths and benefits, and AI weaknesses and risks. We suggest guidance for future studies, with the goal of supporting the safe, equitable, and patient-centred implementation of clinical AI.

Importance: Current methods for identifying hospitalized patients at increased risk of delirium require nurse-administered questionnaires with moderate accuracy. Objective: To develop and validate a machine learning model that predicts incident delirium risk based on electronic health data available on admission. Design, Setting, and Participants: Retrospective cohort study evaluating 5 machine learning algorithms to predict delirium using 796 clinical variables identified by an expert panel as relevant to delirium prediction and consistently available in electronic health records within 24 hours of admission. The training set comprised 14 227 adult patients with non-intensive care unit hospital stays and no delirium on admission who were discharged between January 1, 2016, and August 31, 2017, from UCSF Health, a large academic health institution. The test set comprised 3996 patients with hospital stays who were discharged between August 1, 2017, and November 30, 2017. Exposures: Patient demographic characteristics, diagnoses, nursing records, laboratory results, and medications available in electronic health records during hospitalization. Main Outcomes and Measures: Delirium was defined as a positive Nursing Delirium Screening Scale or Confusion Assessment Method for the Intensive Care Unit score. Models were assessed using the area under the receiver operating characteristic curve (AUC) and compared against the 4-point scoring system AWOL (age >79 years, failure to spell world backward, disorientation to place, and higher nurse-rated illness severity), a validated delirium risk-assessment tool routinely administered in this cohort. Results: The training set included 14 227 patients (5113 [35.9%] aged >64 years; 7335 [51.6%] female; 687 [4.8%] with delirium), and the test set included 3996 patients (1491 [37.3%] aged >64 years; 1966 [49.2%] female; 191 [4.8%] with delirium). In total, the analysis included 18 223 hospital admissions (6604 [36.2%] aged >64 years; 9301 [51.0%] female; 878 [4.8%] with delirium). The AWOL system achieved a baseline AUC of 0.678. The gradient boosting machine model performed best, with an AUC of 0.855. Setting specificity at 90%, the model had a 59.7% (95% CI, 52.4%-66.7%) sensitivity, 23.1% (95% CI, 20.5%-25.9%) positive predictive value, 97.8% (95% CI, 97.4%-98.1%) negative predictive value, and a number needed to screen of 4.8. Penalized logistic regression and random forest models also performed well, with AUCs of 0.854 and 0.848, respectively. Conclusions and Relevance: Machine learning can be used to estimate hospital-acquired delirium risk using electronic health record data available within 24 hours of hospital admission. Such a model may allow more precise targeting of delirium prevention resources without increasing the burden on health care professionals.

The transforming protein of polyoma virus, middle T antigen, associates with two cellular enzymes, pp60c-src, a protein tyrosine kinase, and a phosphatidylinositol kinase that forms phosphatidylinositol 3-phosphate. The formation of a ternary complex of these proteins is essential for complete transformation and maximal tumor induction by the virus. A mutant virus encoding an altered middle T protein that activates pp60c-src but fails to bind phosphatidylinositol kinase is partially defective in transformation. We have confirmed, using an enzymological method, that the product of the in vitro reaction catalyzed by middle T-pp60c-src-phosphatidylinositol kinase complexes is phosphatidylinositol 3-phosphate (PtdIns(3)P), as previously reported (Whitman, M., Downes, C. P., Keeler, M., Keller, T., and Cantley, L. (1988) Nature 332, 644-646). PtdIns(3)P is present in normal as well as virus-infected and transformed cells at levels ranging from 0.6 to 2.6% of the major phosphatidylinositol phosphate isomer, phosphatidylinositol 4-phosphate (PtdIns(4)P). Steady-state levels of PtdIns(3)P do not appear to be affected by the expression of middle T in cells. PtdIns(3)P is not hydrolyzed by bovine brain phospholipase C II, which readily cleaves PtdIns(4)P and other phosphatidylinositols. This result underscores the likelihood that the metabolism of PtdIns(3)P is distinct from that of PtdIns(4)P and raises further questions regarding a possible role of PtdIns(3)P in normal and neoplastic cell growth.

IMPORTANCE: Air pollution is a worldwide public health issue that has been exacerbated by recent wildfires, but the relationship between wildfire-associated air pollution and inflammatory skin diseases is unknown. OBJECTIVE: To assess the associations between wildfire-associated air pollution and clinic visits for atopic dermatitis (AD) or itch and prescribed medications for AD management. DESIGN, SETTING, AND PARTICIPANTS: This cross-sectional time-series study assessed the associations of air pollution resulting from the California Camp Fire in November 2018 and 8049 dermatology clinic visits (4147 patients) at an academic tertiary care hospital system in San Francisco, 175 miles from the wildfire source. Participants included pediatric and adult patients with AD or itch from before, during, and after the time of the fire (October 2018 through February 2019), compared with those with visits in the same time frame of 2015 and 2016, when no large wildfires were near San Francisco. Data analysis was conducted from November 1, 2019, to May 30, 2020. EXPOSURES: Wildfire-associated air pollution was characterized using 3 metrics: fire status, concentration of particulate matter less than 2.5 μm in diameter (PM2.5), and satellite-based smoke plume density scores. MAIN OUTCOMES AND MEASURES: Weekly clinic visit counts for AD or itch were the primary outcomes. Secondary outcomes were weekly numbers of topical and systemic medications prescribed for AD in adults. RESULTS: Visits corresponding to a total of 4147 patients (mean [SD] age, 44.6 [21.1] years; 2322 [56%] female) were analyzed. The rates of visits for AD during the Camp Fire for pediatric patients were 1.49 (95% CI, 1.07-2.07) and for adult patients were 1.15 (95% CI, 1.02-1.30) times the rate for nonfire weeks at lag 0, adjusted for temperature, relative humidity, patient age, and total patient volume at the clinics for pediatric patients. The adjusted rate ratios for itch clinic visits during the wildfire weeks were 1.82 (95% CI, 1.20-2.78) for the pediatric patients and 1.29 (95% CI, 0.96-1.75) for adult patients. A 10-μg/m3 increase in weekly mean PM2.5 concentration was associated with a 7.7% (95% CI, 1.9%-13.7%) increase in weekly pediatric itch clinic visits. The adjusted rate ratio for prescribed systemic medications in adults during the Camp Fire at lag 0 was 1.45 (95% CI, 1.03-2.05). CONCLUSIONS AND RELEVANCE: This cross-sectional study found that short-term exposure to air pollution due to the wildfire was associated with increased health care use for patients with AD and itch. These results may provide a better understanding of the association between poor air quality and skin health and guide health care professionals' counseling of patients with skin disease and public health practice.