Background: Physicians invest hours creating patient notes, which are rich in information but difficult for computers to analyze due to their unstructured format. GPT-4 reshaped our ability to process text, yet it is ...
详细信息
Background: Physicians invest hours creating patient notes, which are rich in information but difficult for computers to analyze due to their unstructured format. GPT-4 reshaped our ability to process text, yet it is unknown how well this model can handle medical notes. This project aims to compare GPT-4’s ability to annotate medical notes against experienced physicians across three different languages at multiple institutions and countries. Methods: This study included eight sites from four countries - the United States, Colombia, Singapore, and Italy. Each site contributed seven de-identified notes (admission, progress, or consult) from hospitalized patients. GPT-4 assessed each note by answering 14 questions, including demographic information, clinical judgments, data quality, and patients’ eligibility for a hypothetical study enrollment. For validation, two physicians from each site independently evaluated GPT-4's responses. Findings: Overall, 56 medical notes, written in English, Italian, and Spanish, were analyzed. A total of 784 responses from GPT-4 were generated. Both physicians agreed with GPT-4’s response 79% of the time (622/784, 95%CI 76-82%). Only one of the two physicians agreed with GPT-4’s response 10% of the time (82/784, 95%CI 8-13%). Neither physician agreed with GPT-4’s response 10% of the time (80/784, 95%CI 8-13%). Both physicians agreed with GPT-4 more often in notes written in Spanish and Italian than in English, with agreement rates of 88% (86/98, 95%CI 79-93%), 84% (82/98, 95%CI 75-90%), and 77% (454/588, 95%CI 74-80%), respectively. Hallucinations were rare (10/784, 95%CI 0-2%). GPT-4 correctly selected patients for a hypothetical study enrollment based on three criteria 90% of the time (95%CI 81-98%). Interpretation: The findings indicate that GPT-4 annotations demonstrated a high agreement rate with physicians across all languages. We also demonstrate GPT-4's potential to assist in patient selection for studies. Funding: None. Declarati
作者:
Abhishek KumarR. DhanuskodiR. KaliappanK. NandakumarAbhishek Kumar teaches design philosophies at Anant National University
Ahmedabad. He earned his Ph.D in Management from Pondicherry University. He is an Economics graduate from Calcutta University and MBA from BIM Trichy. He has published more than 20 articles in reputed international journals has authored two books written articles and columns for newspapers and is quoted on issues related to leadership and marketing by various media platforms. His research work comprises construction of brand personality scale for media aesthetics and phenomenological design. His recent publications are on philosophy of a photograph hermeneutic reality of product and on philosophy of intimate spaces. R. Dhanuskodi has nearly 40 years of R&D experience at BHEL
India in technical areas applicable for thermal power plants. He is a life member of The Institution of Engineers (India) and The Combustion Institute. He has won two BHEL’s Excel awards under the best author category for technical papers. He has visited France Netherlands and Germany under Indo-Europe Clean Coal Development Program. He has guided 42 UG PG and PhD project works. He holds 11 patents 40 copyrights and 2 design registrations. He has presented papers in 20 conferences and published in 10 national and international journals. R. Kaliappan completed his bachelors in electrical and electronics engineering and Masters in Computer Science. He has 36 years of research experience in different fields of power generation and power plant subsystems such as heat transfer studies on boiler circulation
efficiency improvements of boiler subsystems product improvements/ enhancements and setting up test facilities for research studies. He has published a number of technical papers on MHD power generation and heat transfer studies in various national and international journals. He has more than 25 patents and copyrights on products development related to power boilers. He has won BHEL’S gold medal for product development for Smart Wall Blowing system. K. Nandakumar
The field of heliophysics encompasses a diverse range of research areas and rich plasma environments, from the solar dynamo and solar wind turbulence, to the complex terrestrial thermosphere-ionosphere-magnetosphere (...
The field of heliophysics encompasses a diverse range of research areas and rich plasma environments, from the solar dynamo and solar wind turbulence, to the complex terrestrial thermosphere-ionosphere-magnetosphere (TIM) system, to investigations of both induced and intrinsic magnetospheres across the solar system. This perspective paper delves into the experiences and outcomes of a summer school held in person in Boulder, Colorado, USA, from 7–11 August, which offered a platform for interdisciplinary learning and collaboration. The event, supported by NASA's Living with a Star program and hosted by the University Corporation for Atmospheric Research and Cooperative programs for the Advancement of Earth System Science, assembled a cohort of 26 graduate students and postdoctoral researchers from around the world, all engaged in heliophysics research. The summer school included a series of lectures, interactive activities, and a culminating capstone project. The participants' diverse backgrounds enriched discussions and encouraged novel approaches to traditional problems. The capstone projects spanned an array of topics, including investigating the origins of solar wind switchbacks, dissecting the sequence of events from solar eruptions and their corresponding terrestrial consequences, investigating the energy transfer from solar coronal mass ejections to Earth's magnetosphere, and advocating for the exploration of Coulomb collisions in understanding large-scale global systems. Through this perspective, we shed light on the value of international and interdisciplinary collaboration in advancing heliophysics research. This perspective paper encapsulates the ethos of the summer school, serving as a testament to the continuing and collective pursuit of unraveling the mysteries of the heliosphere. Cooperative programs for the Advancement of Earth System Science/University Corporation for Atmospheric Research held the seventeenth annual NASA Heliophysics Summer School bot
Eosinophilic esophagitis (EoE) is a food allergy-associated disease characterized by esophageal eosinophilia and associated inflammation. EoE has become increasingly common in industrialized countries. However, curren...
Eosinophilic esophagitis (EoE) is a food allergy-associated disease characterized by esophageal eosinophilia and associated inflammation. EoE has become increasingly common in industrialized countries. However, current management strategies, such as treatment with swallowed steroids and dietary restrictions, are nonspecific or negatively impact the quality of life of EoE patients. In addition, use of more specific therapies that target immunoglobulin E (IgE) and interleukin (IL)-5 have been largely unsuccessful in ameliorating EoE. Thus, there is an urgent need to identify novel, specific immunological pathways that underlie the pathogenesis of EoE that could be targeted to treat this disease. Recently, a genome-wide association study identified that EoE is associated with a gain-of-function polymorphism in the gene that encodes thymic stromal lymphopoietin (TSLP). TSLP is an IL-7 family member cytokine that promotes allergic inflammation by eliciting T helper type 2 cytokine responses, IgE production, and the expansion of a distinct population of basophils . However, whether TSLP directly promotes allergic inflammation in the esophagus and the mechanisms by which TSLP might contribute to the pathogenesis of EoE remain unknown. Here, we describe a new murine model of experimental EoE-like disease that was used to investigate the role of TSLP in allergic inflammation of the esophagus. Murine experimental EoE-like disease developed independently of IgE but was dependent on TSLP-elicited basophils. Critically, therapeutic antibody-mediated neutralization of TSLP or depletion of basophil populations ameliorated established EoE-like disease in mice. Finally, elevated TSLP levels and exaggerated basophil responses were observed in esophageal biopsies from EoE patients, and a gain-of-function polymorphism in TSLP correlated with increased basophil responses in patients with EoE. Together, these data indicate that TSLP-elicited basophil responses may play a key role in medi
Scientometrics and bibliometrics, the subfields of library and information science, deal with the quantity and quality of research outputs. Currently, various scientometric indices are being used to quantify and compa...
详细信息
Although machine learning (ML) has shown promise in numerous domains, there are concerns about generalizability to out-of-sample data. This is currently addressed by centrally sharing ample, and importantly diverse, d...
详细信息
暂无评论