Journal of Medical Internet Research最新文献

Background: The rising prevalence and swift spread of multidrug-resistant gram-negative bacteria (MDR-GNB), especially Klebsiella pneumoniae (KP), present a critical global health threat highlighted by the World Health Organization, with mortality rates soaring approximately 50% with inappropriate antimicrobial treatment.

Objective: This study aims to advance a novel strategy to develop an artificial intelligence-clinical decision support system (AI-CDSS) that combines machine learning (ML) with matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS), aiming to significantly improve the accuracy and speed of diagnosing antibiotic resistance, directly addressing the grave health risks posed by the widespread dissemination of pan drug-resistant gram-negative bacteria across numerous countries.

Methods: A comprehensive dataset comprising 165,299 bacterial specimens and 11,996 KP isolates was meticulously analyzed using MALDI-TOF MS technology. Advanced ML algorithms were harnessed to sculpt predictive models that ascertain resistance to quintessential antibiotics, particularly levofloxacin and ciprofloxacin, by using the amassed spectral data.

Results: Our ML models revealed remarkable proficiency in forecasting antibiotic resistance, with the random forest classifier emerging as particularly effective in predicting resistance to both levofloxacin and ciprofloxacin, achieving the highest area under the curve of 0.95. Performance metrics across different models, including accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and F₁-score, were detailed, underlining the potential of these algorithms in aiding the development of precision treatment strategies.

Conclusions: This investigation highlights the synergy between MALDI-TOF MS and ML as a beacon of hope against the escalating threat of antibiotic resistance. The advent of AI-CDSS heralds a new era in clinical diagnostics, promising a future in which rapid and accurate resistance prediction becomes a cornerstone in combating infectious diseases. Through this innovative approach, we answered the challenge posed by KP and other multidrug-resistant pathogens, marking a significant milestone in our journey toward global health security.

Background: Evaluation research increasingly needs alternatives to target or gross rating points to comprehensively measure total exposure to modern multichannel public education campaigns that use multiple channels, including TV, radio, digital video, and paid social media, among others. Ratings data typically only capture delivery of broadcast media (TV and radio) and excludes other channels. Studies are needed to validate objective cross-channel metrics such as impressions against self-reported exposure to campaign messages.

Objective: This study aimed to examine whether higher a volume of total media campaign impressions is predictive of individual-level self-reported campaign exposure in California.

Methods: We analyzed over 3 years of advertisement impressions from the California Tobacco Prevention Program's statewide tobacco education campaigns from August 2019 through December 2022. Impressions data varied across designated market areas (DMAs) and across time. These data were merged to individual respondents from 45 waves of panel survey data of Californians aged 18-55 years (N=151,649). Impressions were merged to respondents based on respondents' DMAs and time of survey completion. We used logistic regression to estimate the odds of respondents' campaign recall as a function of cumulative and past 3-month impressions delivered to each respondent's DMA.

Results: Cumulative impressions were positively and significantly associated with recall of each of the Flavors Hook Kids (odds ratio [OR] 1.15, P<.001), Dark Balloons and Apartment (OR 1.20, P<.001), We Are Not Profit (OR 1.36, P<.001), Tell Your Story (E-cigarette, or Vaping, product use Associated Lung Injury; OR 1.06, P<.05), and Thrown Away and Little Big Lies (OR 1.05, P<.01) campaigns. Impressions delivered in the past 3 months were associated with recall of the Flavors Hook Kids (OR 1.13, P<.001), Dark Balloons and Apartment (OR 1.08, P<.001), We Are Not Profit (OR 1.14, P<.001), and Thrown Away and Little Big Lies (OR 1.04, P<.001) campaigns. Past 3-month impressions were not significantly associated with Tell Your Story campaign recall. Overall, magnitudes of these associations were greater for cumulative impressions. We visualize recall based on postestimation predicted values from our multivariate logistic regression models.

Conclusions: Variation in cumulative impressions for California Tobacco Prevention Program's long-term multichannel tobacco education campaign is predictive of increased self-reported campaign recall, suggesting that impressions may be a valid proxy for potential campaign exposure. The use of impressions for purposes of evaluating public education campaigns may help address current methodological limitations arising from the fragmented nature of modern multichannel media campaigns.

Background: Worldwide, cardiovascular diseases are the primary cause of death, with hypertension as a key contributor. In 2019, cardiovascular diseases led to 17.9 million deaths, predicted to reach 23 million by 2030.

Objective: This study presents a new method to predict hypertension using demographic data, using 6 machine learning models for enhanced reliability and applicability. The goal is to harness artificial intelligence for early and accurate hypertension diagnosis across diverse populations.

Methods: Data from 2 national cohort studies, National Health Insurance Service-National Sample Cohort (South Korea, n=244,814), conducted between 2002 and 2013 were used to train and test machine learning models designed to anticipate incident hypertension within 5 years of a health checkup involving those aged ≥20 years, and Japanese Medical Data Center cohort (Japan, n=1,296,649) were used for extra validation. An ensemble from 6 diverse machine learning models was used to identify the 5 most salient features contributing to hypertension by presenting a feature importance analysis to confirm the contribution of each future.

Results: The Adaptive Boosting and logistic regression ensemble showed superior balanced accuracy (0.812, sensitivity 0.806, specificity 0.818, and area under the receiver operating characteristic curve 0.901). The 5 key hypertension indicators were age, diastolic blood pressure, BMI, systolic blood pressure, and fasting blood glucose. The Japanese Medical Data Center cohort dataset (extra validation set) corroborated these findings (balanced accuracy 0.741 and area under the receiver operating characteristic curve 0.824). The ensemble model was integrated into a public web portal for predicting hypertension onset based on health checkup data.

Conclusions: Comparative evaluation of our machine learning models against classical statistical models across 2 distinct studies emphasized the former's enhanced stability, generalizability, and reproducibility in predicting hypertension onset.

Background: Despite the popularity of JUUL e-cigarettes, other brands (eg, Elf Bar) may be gaining digital attention.

Objective: This study compared Google searches for Elf Bar and JUUL from 2022 to 2023 using Google Health Trends Application Programming Interface data.

Methods: Using an infodemiology approach, we examined weekly trends in Google searches (per 10 million) for "Elf Bar" and "JUUL" at the US national and state levels from January 1, 2022, to December 31, 2023. Joinpoint regression was used to assess statistically significant trends in the search probabilities for "Elf Bar" and "JUUL" during the study period.

Results: Elf Bar had less online interest than JUUL at the beginning of 2022. When the US Food and Drug Administration denied JUUL marketing authority on June 23, 2022, JUUL searches peaked at 2609.3 × 10⁷ and fell to 83.9 × 10⁷ on September 3, 2023. Elf Bar searches surpassed JUUL on July 10, 2022, and steadily increased, reaching 523.2 × 10⁷ on December 4, 2022. Overall, Elf Bar's weekly search probability increased by 1.6% (95% CI 1.5%-1.7%; P=.05) from January 2022 to December 2023, with the greatest increase between May 29 and June 19, 2022 (87.7%, 95% CI 35.9%-123.9%; P=.001). Elf Bar searches increased after JUUL's suspension in Pennsylvania (1010%), Minnesota (872.5%), Connecticut (803.5%), New York (738.1%), and New Jersey (702.9%).

Conclusions: Increasing trends in Google searches for Elf Bar indicate that there was a growing online interest in this brand in the United States in 2022.

Background: Large language models (LLMs) have dominated public interest due to their apparent capability to accurately replicate learned knowledge in narrative text. However, there is a lack of clarity about the accuracy and capability standards of LLMs in health care examinations.

Objective: We conducted a systematic review of LLM accuracy, as tested under health care examination conditions, as compared to known human performance standards.

Methods: We quantified the accuracy of LLMs in responding to health care examination questions and evaluated the consistency and quality of study reporting. The search included all papers up until September 10, 2023, with all LLMs published in English journals that report clear LLM accuracy standards. The exclusion criteria were as follows: the assessment was not a health care exam, there was no LLM, there was no evaluation of comparable success accuracy, and the literature was not original research.The literature search included the following Medical Subject Headings (MeSH) terms used in all possible combinations: "artificial intelligence," "ChatGPT," "GPT," "LLM," "large language model," "machine learning," "neural network," "Generative Pre-trained Transformer," "Generative Transformer," "Generative Language Model," "Generative Model," "medical exam," "healthcare exam," and "clinical exam." Sensitivity, accuracy, and precision data were extracted, including relevant CIs.

Results: The search identified 1673 relevant citations. After removing duplicate results, 1268 (75.8%) papers were screened for titles and abstracts, and 32 (2.5%) studies were included for full-text review. Our meta-analysis suggested that LLMs are able to perform with an overall medical examination accuracy of 0.61 (CI 0.58-0.64) and a United States Medical Licensing Examination (USMLE) accuracy of 0.51 (CI 0.46-0.56), while Chat Generative Pretrained Transformer (ChatGPT) can perform with an overall medical examination accuracy of 0.64 (CI 0.6-0.67).

Conclusions: LLMs offer promise to remediate health care demand and staffing challenges by providing accurate and efficient context-specific information to critical decision makers. For policy and deployment decisions about LLMs to advance health care, we proposed a new framework called RUBRICC (Regulatory, Usability, Bias, Reliability [Evidence and Safety], Interoperability, Cost, and Codesign-Patient and Public Involvement and Engagement [PPIE]). This presents a valuable opportunity to direct the clinical commissioning of new LLM capabilities into health services, while respecting patient safety considerations.

Trial registration: OSF Registries osf.io/xqzkw; https://osf.io/xqzkw.

Video abstracts can be useful in health research. A video abstract provides key messages about a research article and can increase public engagement, spark conversations, and may increase academic attention. A growing number of open source software programs make it easier to develop a video abstract. This viewpoint provides practical tips for creating a video abstract for health research.