Dataset Information

Development and Assessment of an Artificial Intelligence-Based Tool for Skin Condition Diagnosis by Primary Care Physicians and Nurse Practitioners in Teledermatology Practices.

ABSTRACT:

Importance

Most dermatologic cases are initially evaluated by nondermatologists such as primary care physicians (PCPs) or nurse practitioners (NPs).

Objective

To evaluate an artificial intelligence (AI)-based tool that assists with diagnoses of dermatologic conditions.

Design, setting, and participants

This multiple-reader, multiple-case diagnostic study developed an AI-based tool and evaluated its utility. Primary care physicians and NPs retrospectively reviewed an enriched set of cases representing 120 different skin conditions. Randomization was used to ensure each clinician reviewed each case either with or without AI assistance; each clinician alternated between batches of 50 cases in each modality. The reviews occurred from February 21 to April 28, 2020. Data were analyzed from May 26, 2020, to January 27, 2021.

Exposures

An AI-based assistive tool for interpreting clinical images and associated medical history.

Main outcomes and measures

The primary analysis evaluated agreement with reference diagnoses provided by a panel of 3 dermatologists for PCPs and NPs. Secondary analyses included diagnostic accuracy for biopsy-confirmed cases, biopsy and referral rates, review time, and diagnostic confidence.

Results

Forty board-certified clinicians, including 20 PCPs (14 women [70.0%]; mean experience, 11.3 [range, 2-32] years) and 20 NPs (18 women [90.0%]; mean experience, 13.1 [range, 2-34] years) reviewed 1048 retrospective cases (672 female [64.2%]; median age, 43 [interquartile range, 30-56] years; 41 920 total reviews) from a teledermatology practice serving 11 sites and provided 0 to 5 differential diagnoses per case (mean [SD], 1.6 [0.7]). The PCPs were located across 12 states, and the NPs practiced in primary care without physician supervision across 9 states. The NPs had a mean of 13.1 (range, 2-34) years of experience and practiced in primary care without physician supervision across 9 states. Artificial intelligence assistance was significantly associated with higher agreement with reference diagnoses. For PCPs, the increase in diagnostic agreement was 10% (95% CI, 8%-11%; P < .001), from 48% to 58%; for NPs, the increase was 12% (95% CI, 10%-14%; P < .001), from 46% to 58%. In secondary analyses, agreement with biopsy-obtained diagnosis categories of maglignant, precancerous, or benign increased by 3% (95% CI, -1% to 7%) for PCPs and by 8% (95% CI, 3%-13%) for NPs. Rates of desire for biopsies decreased by 1% (95% CI, 0-3%) for PCPs and 2% (95% CI, 1%-3%) for NPs; the rate of desire for referrals decreased by 3% (95% CI, 1%-4%) for PCPs and NPs. Diagnostic agreement on cases not indicated for a dermatologist referral increased by 10% (95% CI, 8%-12%) for PCPs and 12% (95% CI, 10%-14%) for NPs, and median review time increased slightly by 5 (95% CI, 0-8) seconds for PCPs and 7 (95% CI, 5-10) seconds for NPs per case.

Conclusions and relevance

Artificial intelligence assistance was associated with improved diagnoses by PCPs and NPs for 1 in every 8 to 10 cases, indicating potential for improving the quality of dermatologic care.

SUBMITTER: Jain A

PROVIDER: S-EPMC8082316 | biostudies-literature | 2021 Apr

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Development and Assessment of an Artificial Intelligence-Based Tool for Skin Condition Diagnosis by Primary Care Physicians and Nurse Practitioners in Teledermatology Practices.

Jain Ayush A Way David D Gupta Vishakha V Gao Yi Y de Oliveira Marinho Guilherme G Hartford Jay J Sayres Rory R Kanada Kimberly K Eng Clara C Nagpal Kunal K DeSalvo Karen B KB Corrado Greg S GS Peng Lily L Webster Dale R DR Dunn R Carter RC Coz David D Huang Susan J SJ Liu Yun Y Bui Peggy P Liu Yuan Y

JAMA network open 20210401 4

<h4>Importance</h4>Most dermatologic cases are initially evaluated by nondermatologists such as primary care physicians (PCPs) or nurse practitioners (NPs).<h4>Objective</h4>To evaluate an artificial intelligence (AI)-based tool that assists with diagnoses of dermatologic conditions.<h4>Design, setting, and participants</h4>This multiple-reader, multiple-case diagnostic study developed an AI-based tool and evaluated its utility. Primary care physicians and NPs retrospectively reviewed an enriche ...[more]

PMID: 33909055

Similar Datasets

Project description:IntroductionAn artificial intelligence as a medical device (AIaMD), built on convolutional neural networks, has demonstrated high sensitivity for melanoma. To be of clinical value, it needs to safely reduce referral rates. The primary objective of this study was to demonstrate that the AIaMD had a higher rate of correctly classifying lesions that did not need to be referred for biopsy or urgent face-to-face dermatologist review, compared to teledermatology standard of care (SoC), while achieving the same sensitivity to detect malignancy. Secondary endpoints included the sensitivity, specificity, positive and negative predictive values, and number needed to biopsy to identify one case of melanoma or squamous cell carcinoma (SCC) by both the AIaMD and SoC.MethodsThis prospective, single-centre, single-arm, masked, non-inferiority, adaptive, group sequential design trial recruited patients referred to a teledermatology cancer pathway (clinicaltrials.gov NCT04123678). Additional dermoscopic images of each suspicious lesion were taken using a smartphone with a dermoscopic lens attachment. The images were assessed independently by a consultant dermatologist and the AIaMD. The outputs were compared with the final histological or clinical diagnosis.ResultsA total of 700 patients with 867 lesions were recruited, of which 622 participants with 789 lesions were included in the per-protocol (PP) population. In total, 63.3% of PP participants were female; 89.0% identified as white, and the median age was 51 (range 18-95); and all Fitzpatrick skin types were represented including 25/622 (4.0%) type IV-VI skin. A total of 67 malignant lesions were identified, including 8 diagnosed as melanoma. The AIaMD sensitivity was set at 91 and 92.5%, to match the literature-defined clinician sensitivity (91.46%) as closely as possible. In both settings, the AIaMD identified had a significantly higher rate of identifying lesions that did not need a biopsy or urgent referral compared to SoC (p-value = 0.001) with comparable sensitivity for skin cancer.DiscussionThe AIaMD identified significantly more lesions that did not need to be referred for biopsy or urgent face-to-face dermatologist review, compared to teledermatologists. This has the potential to reduce the burden of unnecessary referrals when used as part of a teledermatology service.

Project description:ImportanceThe rapidly growing population of older adults and their concomitant high prevalence of chronic health conditions require an increased supply in the specialized geriatric workforce to meet increasing health care demands. Understanding trends and geographic disparities in the supply of the geriatric workforce is essential for developing effective policies.ObjectivesTo examine temporal and geographic trends in the supply of geriatric physicians (GMDs) and geriatric nurse practitioners (GNPs) from 2010 to 2020 and to assess potential disparities between metropolitan and nonmetropolitan counties.Design, setting, and participantsThis repeated cross-sectional study used annual county-level data from 2010 to 2020, encompassing all counties in the 50 US states and Washington, DC. Statistical analysis was performed from June 2023 to March 2024.Main outcomes and measuresThe primary outcomes were the numbers of GMDs, GNPs, and the combined number of GMDs and GNPs per 100 000 older adults. The secondary outcome included the proportion of counties with or without any GMDs or GNPs.ResultsFrom 2010 to 2020, the national per capita supply of GMDs decreased by 12.7%, from 13.4 per 100 000 older adults in 2010 to 11.7 per 100 000 older adults in 2020, while GNPs increased by 125.0%, from 4.4 per 100 000 older adults in 2010 to 9.9 per 100 000 older adults in 2020. The combined GMD and GNP workforce increased by 21.3%, from 17.8 per 100 000 older adults in 2010 to 21.6 per 100 000 older adults in 2020. The distributions of older adults, GMDs, and GNPs closely resembled the distribution of metropolitan and nonmetropolitan counties, with GMDs and GNPs highly concentrated in metropolitan counties where the number of older adults was greatest. Throughout the study period, 63.9% of counties (2008 of 3142 in 2010-2019; 2009 of 3143 in 2020), predominantly small and nonmetropolitan counties, had no GMDs or GNPs, which was associated with the disparities between metropolitan and nonmetropolitan counties.Conclusions and relevanceThis repeated cross-sectional study found that from 2010 to 2020, the overall national supply of GMDs and GNPs kept pace with the growth of the older population, largely due to the rapid growth in the number of GNPs. However, significant geographic disparities persisted, particularly in small and nonmetropolitan counties. Future efforts should focus on increasing the availability of GMDs and GNPs in underserved small and nonmetropolitan counties to ensure equitable access to geriatric care.

Project description:IntroductionClozapine (CLZ) is the only proven effective therapy for treatment-resistant schizophrenia, but it is underutilized across the globe. Previous findings suggest a lack of experience with CLZ prescription and concerns about CLZ's pharmacological characteristics are the prime reasons for CLZ underutilization. To our knowledge, it is currently unknown whether the reasons for underutilization and suggested solutions differ between physicians and nurse practitioners. Such differences are important as nurse practitioners are becoming increasingly involved in prescribing CLZ.MethodsTo examine to what degree physicians and nurse practitioners differ with regard to their take on reasons for CLZ underutilization and suggested solutions, an online questionnaire was distributed to physicians and nurse practitioners. The primary outcome was to compare the patient-related and prescriber-related reasons for CLZ underprescription between physicians and nurse practitioners, while secondary outcome measures included the potential solutions to prevent this underprescription.ResultsPhysicians (N = 112) and nurse practitioners (N = 41) agreed that the two most common reasons for underprescription (patient-related and prescriber-related) were refusal to undergo regular blood tests and side-effect concerns. They also agreed that the third most common prescriber-related reason was medical complications. Physicians rated patients' unwillingness to switch medication as the third most common reason for CLZ underprescription, whereas nurse practitioners rated refusal to undergo baseline bloodtests as the third most common reason. The solutions to reduce underprescription largely corresponded between both groups.ConclusionsWe conclude that slight differences exist between physicians' and nurse practitioners' viewpoints on patient-related and prescriber-related reasons for CLZ underprescription. Future research projects should involve patients to elucidate whether the patient-related factors put forward by prescribers align with the patients' opinions.

Project description:ImportanceGiven the disparities in access to primary care and the growing nurse practitioner (NP) workforce, it is important to understand the distribution of primary care practices with NPs across communities of varying socioeconomic characteristics.ObjectiveTo compare the socioeconomic characteristics of communities that have primary care practices with or without NPs.Design, setting, and participantsThis secondary, cross-sectional analysis of 79 743 primary care practices used 4 merged data sources: the IQVIA 2023 OneKey database, the 2020 US biennial Census, the 2017-2022 American Community Survey, and the 2021 Area Deprivation Index (ADI). US Census Tracts and Divisions from 2023 were used to examine differences in socioeconomic characteristics across communities.ExposurePrimary care practices with NPs vs without NPs.Main outcomes and measuresSocioeconomic characteristics of Census Tracts and Block Groups, including racial and ethnic composition, median household income, percentage below the federal poverty level, educational attainment, and the ADI (an ordinal percentile ranking of Census Block Groups from 1 to 100, with 1 being least disadvantaged and 100 being most disadvantaged).ResultsOf 79 743 primary care practices, 42 601 (53.4%) employed NPs in 2023. Practices with NPs, compared with those without, were significantly more likely to be in communities classified as low income (23.3% vs 17.0%; P < .001) and rural (11.9% vs 5.5%; P < .001). On average, these communities had a higher proportion of the population living below the federal poverty level (14.4% [95% CI, 14.3%-14.5%] vs 12.8% [95% CI, 12.7%-12.9%]; P < .001) and without a high school diploma (19.8% [95% CI, 19.7%-19.9%] vs 18.5% [95% CI, 18.4%-18.6%]; P < .001). Communities with practices with NPs also had significantly higher mean ADI percentiles than communities with practices without NPs (53.3% [95% CI, 53.1%-53.6%] vs 42.5% [95% CI, 42.2%-42.7%]; P < .001). As the number of primary care practices decreased in disadvantaged areas, the proportion of practices with NPs increased. In most US Census Divisions, there were more primary care practices with NPs than without, a difference that was marked in low-income communities.Conclusions and relevanceThis cross-sectional study of primary care practices in the US found that primary care practices with NPs, compared with those without NPs, were more likely to be located in communities with lower income and educational attainment and greater levels of overall socioeconomic disadvantage. This finding suggests that NPs are key to ensuring access to primary care in communities with socioeconomic disadvantage.

Project description:BackgroundGeneral practitioners (GPs) care for a large number of patients with various diseases in very short timeframes under high uncertainty. Thus, systems enabled by artificial intelligence (AI) are promising and time-saving solutions that may increase the quality of care.ObjectiveThis study aims to understand GPs' attitudes toward AI-enabled systems in medical diagnosis.MethodsWe interviewed 18 GPs from Germany between March 2020 and May 2020 to identify determinants of GPs' attitudes toward AI-based systems in diagnosis. By analyzing the interview transcripts, we identified 307 open codes, which we then further structured to derive relevant attitude determinants.ResultsWe merged the open codes into 21 concepts and finally into five categories: concerns, expectations, environmental influences, individual characteristics, and minimum requirements of AI-enabled systems. Concerns included all doubts and fears of the participants regarding AI-enabled systems. Expectations reflected GPs' thoughts and beliefs about expected benefits and limitations of AI-enabled systems in terms of GP care. Environmental influences included influences resulting from an evolving working environment, key stakeholders' perspectives and opinions, the available information technology hardware and software resources, and the media environment. Individual characteristics were determinants that describe a physician as a person, including character traits, demographic characteristics, and knowledge. In addition, the interviews also revealed the minimum requirements of AI-enabled systems, which were preconditions that must be met for GPs to contemplate using AI-enabled systems. Moreover, we identified relationships among these categories, which we conflate in our proposed model.ConclusionsThis study provides a thorough understanding of the perspective of future users of AI-enabled systems in primary care and lays the foundation for successful market penetration. We contribute to the research stream of analyzing and designing AI-enabled systems and the literature on attitudes toward technology and practice by fostering the understanding of GPs and their attitudes toward such systems. Our findings provide relevant information to technology developers, policymakers, and stakeholder institutions of GP care.

Dataset Information

Development and Assessment of an Artificial Intelligence-Based Tool for Skin Condition Diagnosis by Primary Care Physicians and Nurse Practitioners in Teledermatology Practices.

Importance

Objective

Design, setting, and participants

Exposures

Main outcomes and measures

Results

Conclusions and relevance

Publications

Development and Assessment of an Artificial Intelligence-Based Tool for Skin Condition Diagnosis by Primary Care Physicians and Nurse Practitioners in Teledermatology Practices.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets