Artificial intelligence (AI) has driven innovative transformation in healthcare service patterns, despite a lack of understanding of its performance in clinical practice. We conducted a cross-sectional analysis of AI-related trials in healthcare based on ClinicalTrials.gov, intending to investigate the trial characteristics and AI’s development status. Additionally, the Neo4j graph database and visualization technology were employed to construct an AI technology application graph, achieving a visual representation and analysis of research hotspots in healthcare AI. A total of 1725 eligible trials that were registered in ClinicalTrials.gov up to 31 March 2022 were included in this study. The number of trial registrations has dramatically grown each year since 2016. However, the AI-related trials had some design drawbacks and problems with poor-quality result reporting. The proportion of trials with prospective and randomized designs was insufficient, and most studies did not report results upon completion. Currently, most healthcare AI application studies are based on data-driven learning algorithms, covering various disease areas and healthcare scenarios. As few studies have publicly reported results on ClinicalTrials.gov, there is not enough evidence to support an assessment of AI’s actual performance. The widespread implementation of AI technology in healthcare still faces many challenges and requires more high-quality prospective clinical validation.

BACKGROUND The period of the rapid spread of the coronavirus disease 2019 (COVID-19) pandemic in the United States made people uncertain about their perception of the threat from COVID-19 and the response measures. To mount an effective response to this epidemic, it is necessary to understand the public's perceptions, behaviors and attitudes. OBJECTIVE To test hypotheses that the perception of a threat from COVID-19 influences attitudes and behaviors. METHODS This study used an open dataset of online questionnaires about COVID-19 provided by Nexoid, and selected the results of a questionnaire on behaviors, attitudes and perceptions related to COVID-19 among the US public from March 29 to April 20, 2020. In the end, a total of 466497 people living in the United States took part in the survey. RESULTS The average self-assessed probability of contracting COVID-19 in this study was 36.48%, and 43.5% of the respondents thought their chance of getting COVID-19 was less than 30%. The predicted mean self-assessed probability of contracting COVID-19 among males was 96.97% lower than that among females. Furthermore, compared with those who had close contact with less than 10 people, those who were in close contact with more than 10 people had a 67019.89% higher predicted mean self-assessed probability of contracting COVID-19 (b=6.51, P<0.001). Those who were engaged in critical work (b=7.90, P<0.001) had a mean self-assessed probability of contracting COVID-19 that was 270653.14% higher than that of those who worked at home. The odds of reporting disagreement with taking measures to reduce their personal risk were 14%, 17% and 29% lower for participants from states reporting 10001 to 20000 cases (OR=0.86, 95% CI: 0.79-0.94), 20001 to 40000 cases (OR=0.83, 95% CI: 0.75-0.91) and over 40001 COVID-19 cases (OR=0.71, 95% CI: 0.59-0.84), respectively, compared to those living in states with 1001 to 5000 cases. CONCLUSIONS This survey is the first attempt to describe on a large scale the determinants of the US public's perception of the threat from COVID-19. The self-assessed probability of contracting COVID-19 differed significantly based on the respondents’ gender, state of residence, age, BMI, smoking habit, alcohol consumption, drug use, disease, environment and behaviors. These findings have certain value as a reference for public health policy makers and healthcare workers seeking to identify target populations for COVID-19 prevention and health education.