scholarly journals Medical Specialty Recommendations by an Artificial Intelligence Chatbot on a Smartphone: Development and Deployment (Preprint)

2021 ◽  
Author(s):  
Hyeonhoon Lee ◽  
Jaehyun Kang ◽  
Jonghyeon Yeo
Keyword(s):  
Artificial Intelligence ◽  
Primary Care ◽  
Deep Learning ◽  
Language Processing ◽  
Web Application ◽  
Medical Specialty ◽  
Primary Care Providers ◽  
Contactless Method ◽  
Care Providers ◽  
Medical Specialist

BACKGROUND The current coronavirus disease 2019 (COVID-19) pandemic limits daily activities, even contact between patients and primary care providers. This makes it more difficult to provide adequate primary care services, which include connecting patients to an appropriate medical specialist. A smartphone-compatible AI chatbot that classifies patients’ symptoms and recommends the appropriate medical specialty could provide a valuable solution. OBJECTIVE In order to establish a contactless method of recommending the appropriate medical specialty, this study aims to construct a deep learning-based natural language processing (NLP) pipeline and to develop an artificial intelligence (AI) chatbot that can be used on a smartphone. METHODS We collected 118,008 sentences containing information on symptoms with labels (medical specialty), conducted data cleansing, and finally constructed a pipeline of 51,134 sentences for this study. Several deep learning models, including four different Long Short-Term Memory (LSTM) models with or without attention and with or without a pretrained FastText embedding layer as well as Bidirectional Encoder Representations from Transformers (BERT) for NLP, were trained and validated using a randomly selected test dataset. The performance of the models was evaluated by the precision, recall, F1 score and area under the receiver operating characteristic curve (AUC). An AI chatbot was also designed to make it easy for patients to use this recommendation system. We used an open-source framework called Alpha to develop our AI chatbot. This takes the form of a web application with a frontend chat interface capable of conversing in text and a backend cloud-based server application to handle data collection, process the data with a deep learning model, and offer the medical specialty recommendation in a responsive web which is compatible with both desktops and smartphones. RESULTS The BERT model yielded the best performance, with an AUC of 0.964 and F1 score of 0.768, followed by LSTM with embedding vectors, with an AUC of 0.965 and F1 score of 0.739. Considering the limitations of computing resources and the wide availability of smartphones, the LSTM model with embedding vectors trained on our dataset was adopted for our AI chatbot service. We also deployed an Alpha version of the AI chatbot to be executed on both desktops and smartphones. CONCLUSIONS With the increasing need for telemedicine during the current COVID-19 pandemic, an AI chatbot based on a deep learning-based NLP model that can recommend a medical specialty to patients using their smartphones would be exceedingly useful. The chatbot allows patients to quickly and contactlessly identify the proper medical specialist based on their symptoms, and so may support both patients and primary care providers.

scholarly journals Artificial Intelligence Support for Skin Lesion Triage in Primary Care and Dermatology (Preprint)

2021 ◽  
Author(s):  
Harmony Thompson ◽  
Amanda Oakley ◽  
Michael B Jameson ◽  
Adrian Bowling
Keyword(s):  
Artificial Intelligence ◽  
Primary Care ◽  
Health Care ◽  
Skin Cancer ◽  
Sensitivity And Specificity ◽  
Health Care Access ◽  
Real Life ◽  
Primary Care Providers ◽  
Skin Lesions ◽  
Care Providers

BACKGROUND Primary care providers, dermatology specialists, and health care access are key components of primary prevention, early diagnosis, and treatment of skin cancer. Artificial intelligence (AI) offers the promise of diagnostic support for nonspecialists, but real-world clinical validation of AI in primary care is lacking. OBJECTIVE We aimed to (1) assess the reliability of an AI-based clinical triage algorithm in classifying benign and malignant skin lesions and (2) evaluate the quality of images obtained in primary care using the study camera (3Gen DermLite Cam v4 or similar). METHODS This was a single-center, prospective, double-blinded observational study with a predetermined study design. We recruited participants with suspected skin cancer in 20 primary care practices who were referred for assessment via teledermatology. A second set of photographs taken using a standardized camera was processed by the AI algorithm. We evaluated the image quality and compared two teledermatologists’ diagnoses by consensus (the “gold standard”) with AI and histology where applicable. RESULTS Our primary outcome assessment stratified 391 skin lesions by management as benign, uncertain, or malignant. Uncertain lesions were not included in the sensitivity and specificity analyses. Uncertain lesions included lesions that had either diagnostic or management uncertainties. For the remaining 242 lesions, the sensitivity was 97.26% (95% CI 93.13%-99.25%) and the specificity was 97.92% (95% CI 92.68%-99.75%). The AI algorithm was compared with the histological diagnoses for 123 lesions. The sensitivity was 100% (95% CI 95.85%-100%) and the specificity was 72.22% (95% CI 54.81%-85.80%). CONCLUSIONS The AI algorithm demonstrates encouraging results, with high sensitivity and specificity, concordant with previous AI studies. It shows potential as a triage tool in conjunction with teledermatology to augment health care and improve access to dermatology. Further real-life studies need to be conducted on a larger scale to assess the reliability, usability, and cost-effectiveness of the algorithm in primary care.

The National Residency Exchange: A Proposal to Restore Primary Care in an Age of Microspecialization

2012 ◽  
Vol 38 (1) ◽  
pp. 158-195 ◽  
Author(s):  
Glen Cheng
Keyword(s):  
Primary Care ◽  
Health Insurance ◽  
Primary Care Physicians ◽  
Healthcare Access ◽  
Medical Specialty ◽  
Primary Care Providers ◽  
The United States ◽  
Care Providers ◽  
Loan Repayment ◽  
Patient Protection

Healthcare deficiencies in the United States have long been perpetuated by a shortage of primary care providers. A core purpose of the Patient Protection and Affordable Care Act (PPACA) is to provide health insurance for America's approximately fifty million uninsured. Implementation of universal health insurance, however, does not mean sufficient healthcare access for all, since the supply of physicians does not and will not meet demand. For reasons reviewed in this Article, the current physician shortage mainly impacts primary care providers. This shortage is particularly troubling because increased provision of primary care relative to specialty care has been associated with improvement in health outcomes, disease prevention, cost effectiveness, and coordination of care. This Article highlights provisions in the PPACA that impact primary care physicians. Finally, this Article proposes the creation of a universal primary care loan repayment program and a national residency exchange designed to alleviate the U.S. primary care crisis by facilitating optimal distribution of resident physicians in each medical specialty based on community need.

scholarly journals Artificial intelligence in breast cancer screening: primary care provider preferences

2020 ◽  
Author(s):  
Nathaniel Hendrix ◽  
Brett Hauber ◽  
Christoph I Lee ◽  
Aasthaa Bansal ◽  
David L Veenstra
Keyword(s):  
Breast Cancer ◽  
Artificial Intelligence ◽  
Primary Care ◽  
Decision Making ◽  
Cancer Screening ◽  
Breast Cancer Screening ◽  
Primary Care Providers ◽  
Latent Class Models ◽  
Supporting Evidence ◽  
Care Providers

Abstract Background Artificial intelligence (AI) is increasingly being proposed for use in medicine, including breast cancer screening (BCS). Little is known, however, about referring primary care providers’ (PCPs’) preferences for this technology. Methods We identified the most important attributes of AI BCS for ordering PCPs using qualitative interviews: sensitivity, specificity, radiologist involvement, understandability of AI decision-making, supporting evidence, and diversity of training data. We invited US-based PCPs to participate in an internet-based experiment designed to force participants to trade off among the attributes of hypothetical AI BCS products. Responses were analyzed with random parameters logit and latent class models to assess how different attributes affect the choice to recommend AI-enhanced screening. Results Ninety-one PCPs participated. Sensitivity was most important, and most PCPs viewed radiologist participation in mammography interpretation as important. Other important attributes were specificity, understandability of AI decision-making, and diversity of data. We identified 3 classes of respondents: “Sensitivity First” (41%) found sensitivity to be more than twice as important as other attributes; “Against AI Autonomy” (24%) wanted radiologists to confirm every image; “Uncertain Trade-Offs” (35%) viewed most attributes as having similar importance. A majority (76%) accepted the use of AI in a “triage” role that would allow it to filter out likely negatives without radiologist confirmation. Conclusions and Relevance Sensitivity was the most important attribute overall, but other key attributes should be addressed to produce clinically acceptable products. We also found that most PCPs accept the use of AI to make determinations about likely negative mammograms without radiologist confirmation.

scholarly journals Artificial Intelligence Support for Skin Lesion Triage in Primary Care and Dermatology

Iproceedings ◽  
10.2196/35395 ◽  
2021 ◽  
Vol 6 (1) ◽  
pp. e35395
Author(s):  
Harmony Thompson ◽  
Amanda Oakley ◽  
Michael B Jameson ◽  
Adrian Bowling
Keyword(s):  
Artificial Intelligence ◽  
Primary Care ◽  
Health Care ◽  
Skin Cancer ◽  
Sensitivity And Specificity ◽  
Conflicts Of Interest ◽  
Real Life ◽  
Primary Care Providers ◽  
Skin Lesions ◽  
Care Providers

Background Primary care providers, dermatology specialists, and health care access are key components of primary prevention, early diagnosis, and treatment of skin cancer. Artificial intelligence (AI) offers the promise of diagnostic support for nonspecialists, but real-world clinical validation of AI in primary care is lacking. Objective We aimed to (1) assess the reliability of an AI-based clinical triage algorithm in classifying benign and malignant skin lesions and (2) evaluate the quality of images obtained in primary care using the study camera (3Gen DermLite Cam v4 or similar). Methods This was a single-center, prospective, double-blinded observational study with a predetermined study design. We recruited participants with suspected skin cancer in 20 primary care practices who were referred for assessment via teledermatology. A second set of photographs taken using a standardized camera was processed by the AI algorithm. We evaluated the image quality and compared two teledermatologists’ diagnoses by consensus (the “gold standard”) with AI and histology where applicable. Results Our primary outcome assessment stratified 391 skin lesions by management as benign, uncertain, or malignant. Uncertain lesions were not included in the sensitivity and specificity analyses. Uncertain lesions included lesions that had either diagnostic or management uncertainties. For the remaining 242 lesions, the sensitivity was 97.26% (95% CI 93.13%-99.25%) and the specificity was 97.92% (95% CI 92.68%-99.75%). The AI algorithm was compared with the histological diagnoses for 123 lesions. The sensitivity was 100% (95% CI 95.85%-100%) and the specificity was 72.22% (95% CI 54.81%-85.80%). Conclusions The AI algorithm demonstrates encouraging results, with high sensitivity and specificity, concordant with previous AI studies. It shows potential as a triage tool in conjunction with teledermatology to augment health care and improve access to dermatology. Further real-life studies need to be conducted on a larger scale to assess the reliability, usability, and cost-effectiveness of the algorithm in primary care. Acknowledgments MoleMap NZ, who developed the AI algorithm, provided some funding for this study. HT's salary was partially sponsored by MoleMap NZ, who developed the AI algorithm. AB is a shareholder and consultant to Molemap Ltd provider of the AI algorithm. Conflicts of Interest None declared.

The Perspective of Older Men With Depression on Suicide and Its Prevention in Primary Care

Crisis ◽  
2018 ◽  
Vol 39 (5) ◽  
pp. 397-405 ◽  
Author(s):  
Steven Vannoy ◽  
Mijung Park ◽  
Meredith R. Maroney ◽  
Jürgen Unützer ◽  
Ester Carolina Apesoa-Varano ◽  
...  
Keyword(s):  
Primary Care ◽  
Low Socioeconomic Status ◽  
Thematic Analysis ◽  
Older Men ◽  
Primary Care Providers ◽  
Depression Treatment ◽  
Care Providers ◽  
Low Socioeconomic ◽  
Self Image ◽  
The Impact

Abstract. Background: Suicide rates in older men are higher than in the general population, yet their utilization of mental health services is lower. Aims: This study aimed to describe: (a) what primary care providers (PCPs) can do to prevent late-life suicide, and (b) older men's attitudes toward discussing suicide with a PCP. Method: Thematic analysis of interviews focused on depression and suicide with 77 depressed, low-socioeconomic status, older men of Mexican origin, or US-born non-Hispanic whites recruited from primary care. Results: Several themes inhibiting suicide emerged: it is a problematic solution, due to religious prohibition, conflicts with self-image, the impact on others; and, lack of means/capacity. Three approaches to preventing suicide emerged: talking with them about depression, talking about the impact of their suicide on others, and encouraging them to be active. The vast majority, 98%, were open to such conversations. An unexpected theme spontaneously arose: "What prevents men from acting on suicidal thoughts?" Conclusion: Suicide is rarely discussed in primary care encounters in the context of depression treatment. Our study suggests that older men are likely to be open to discussing suicide with their PCP. We have identified several pragmatic approaches to assist clinicians in reducing older men's distress and preventing suicide.

Sign in / Sign up

Export Citation Format

Share Document