Article
Oncology
Andrea Cocci, Marta Pezzoli, Mattia Lo Re, Giorgio Ivan Russo, Maria Giovanna Asmundo, Mikkel Fode, Giovanni Cacciamani, Sebastiano Cimino, Andrea Minervini, Emil Durukan
Summary: This study analyzes the quality and appropriateness of ChatGPT's responses to Urology case studies compared to those of a urologist. The findings show that 52% of the responses generated by ChatGPT were deemed appropriate. However, there is room for improvement in both the appropriateness and quality of ChatGPT's responses.
PROSTATE CANCER AND PROSTATIC DISEASES
(2023)
Article
Urology & Nephrology
Juliusz Jan Szczesniewski, Carlos Tellez Fouz, Alejandra Ramos Alba, Francisco Javier Diaz Goizueta, Ana Garcia Tello, Luis Llanes Gonzalez
Summary: The study aimed to evaluate the quality of information provided by the artificial intelligence software ChatGPT and determine if it is a secure source for patients. Two urologists analysed the responses provided by ChatGPT using the DISCERN questionnaire and a brief instrument for evaluating the quality of informed consent documents. The results showed that while the overall information provided by ChatGPT was well-balanced, it should be used with caution due to the lack of disclosure of information sources and potential bias.
WORLD JOURNAL OF UROLOGY
(2023)
Editorial Material
Multidisciplinary Sciences
Jeffrey M. Perkel
Summary: Computer scientists warn that although the tools are powerful, they are not as intelligent as they may appear, and should be used with caution.
Article
Computer Science, Information Systems
Mi-Na Chu
Summary: Companies aim to develop an innovative and creative organizational culture for improved performance. This study examined the impact of ChatGPT's system quality, information quality, and service quality on user satisfaction and benefits. The findings showed that all three quality components had positive effects, with service quality being the most influential. However, the moderating effect of a flexible organizational culture was not significant.
Article
Radiology, Nuclear Medicine & Medical Imaging
Alexander Rau, Stephan Rau, Daniela Zoeller, Anna Fink, Hien Tran, Caroline Wilpert, Johanna Nattenmuller, Jakob Neubauer, Fabian Bamberg, Marco Reisert, Maximilian F. Russe
Summary: This study explores the use of an interactive chatbot to provide personalized imaging recommendations based on radiologic guidelines. The chatbot's performance is compared to that of general radiologists and generic chatbots, and it is found that the chatbot performs at least as well as humans, with the context-based algorithm showing superior performance. The chatbot also provides substantial time and cost savings.
Review
Surgery
Ali Talyshinskii, Nithesh Naik, B. M. Zeeshan Hameed, Ulanbek Zhanbyrbekuly, Gafur Khairli, Bakhman Guliev, Patrick Juilebo-Jones, Lazaros Tzelves, Bhaskar Kumar Somani
Summary: ChatGPT has proven to be a valuable tool in urology for clinical documentation, patient education, and research. However, concerns remain about web crawling, plagiarism risks, and patient data privacy. Further improvement and studies are needed to optimize its use in urology.
FRONTIERS IN SURGERY
(2023)
Article
Obstetrics & Gynecology
Bengu Mutlu Sutcuoglu, Melike Guler
Summary: The study finds that ChatGPT provides mostly accurate information about premature ovarian insufficiency (POI), but there are also instances of significantly inaccurate information. Further research is needed to validate the use of AI chatbots in reproductive health.
MENOPAUSE-THE JOURNAL OF THE NORTH AMERICAN MENOPAUSE SOCIETY
(2023)
Review
Public, Environmental & Occupational Health
Thomas Gadsden, Sikhumbuzo A. Mabunda, Anna Palagyi, Asri Maharani, Sujarwoto Sujarwoto, Michelle Baddeley, Stephen Jan
Summary: The review of evidence on the impact of performance-based incentives for community health workers found that publicly recognizing CHWs' efforts can improve service delivery outcomes, while large financial incentives may lead to neglect of non-incentivized tasks. Further research exploring the interactions among different forms of incentives, contextual factors, and sustainability is needed as calls to professionalize CHW programs continue to grow.
BULLETIN OF THE WORLD HEALTH ORGANIZATION
(2021)
Article
Chemistry, Multidisciplinary
Ping Feng, Hang Wu, Ziqian Yang, Yunyi Wang, Dantong Ouyang
Summary: This article introduces a method for relation extraction task that utilizes prompt information and Top-K prediction sets for data augmentation, and incorporates ChatGPT to improve the performance of the relational classification model. Experimental results demonstrate that the proposed method achieves high results on the evaluation dataset.
APPLIED SCIENCES-BASEL
(2023)
Article
Environmental Sciences
Desneiges S. Murray, Edom Moges, Laurel Larsen, Michelle D. Shattuck, William H. Mcdowell, Adam S. Wymore
Summary: This study used an information theory metric to investigate the (a)synchrony between nitrogen wet deposition and watershed nitrogen outputs. The results showed that wet deposition nitrogen has a lagged nonlinear impact on river nitrogen, and wet deposition organic nitrogen and ammonium ions have the most significant influence on river nitrogen. In addition, the study found that assimilation is a positive lagged function of increasing wet deposition nitrogen.
WATER RESOURCES RESEARCH
(2023)
Article
Environmental Sciences
Desneiges S. Murray, Edom Moges, Laurel Larsen, Michelle D. Shattuck, William H. Mcdowell, Adam S. Wymore
Summary: This study investigates the impact of nitrogen wet deposition on watershed biogeochemical cycling using the transfer entropy method. The results show that wet deposition N has a lagged nonlinear effect on river N, with river DON being mostly influenced by wet deposition N. Additionally, the processes of mineralization and assimilation display stronger synchrony than nitrification.
WATER RESOURCES RESEARCH
(2023)
Article
Computer Science, Information Systems
Ji Young Yun, Dong Jin Kim, Nara Lee, Eun Key Kim
Summary: This study evaluated the quality of the answers provided by ChatGPT during hypothetical breast augmentation consultations and compared the scores between plastic surgeons and laypersons. The results showed that plastic surgeons generally scored lower than laypersons, particularly in the consultation category for DISCERN reliability, information quality, and score. Additionally, plastic surgeons rated the overall quality of the procedure category lower compared to other question categories.
INTERNATIONAL JOURNAL OF MEDICAL INFORMATICS
(2023)
Article
Health Care Sciences & Services
Harriet Louise Walker, Shahi Ghani, Christoph Kuemmerli, Christian Andreas Nebiker, Beat Peter Muller, Dimitri Aristotle Raptis, Sebastian Manuel Staubli
Summary: ChatGPT-4 is an AI chatbot that can answer complex questions and may become the new standard for accessing medical information. This study assessed the reliability of medical information provided by ChatGPT and found that it is comparable in quality to existing internet information.
JOURNAL OF MEDICAL INTERNET RESEARCH
(2023)
Article
Quantum Science & Technology
Krzysztof Domino, Matyas Koniorczyk, Zbigniew Puchala
Summary: The ability to evaluate the outcomes of quantum annealers is crucial for their use in complex computational tasks. This study introduces a statistical test based on the data alone to assess the quality of Ising-based annealers' output, specifically examining the probability of sampling the ground state. A higher probability value indicates that the lower part of the spectrum is more likely to be included in the sample. Using a plausible model of the univariate energy distribution of the sample, the ground-state energy and temperature are expressed as a function of cumulants up to the third order. By evaluating this multiple times using Bootstrap resampling, an estimated histogram of ground-state energies is obtained, allowing for the deduction of the desired parameter. This approach provides an easily implementable method for validating the output of Ising-based annealers, as demonstrated through experiments with actual samples from quantum annealer devices.
QUANTUM INFORMATION PROCESSING
(2022)
Article
Health Care Sciences & Services
Vagelis Hristidis, Nicole Ruggiano, Ellen L. Brown, Sai Rithesh Reddy Ganta, Selena Stewart
Summary: This study compared the quality of results between ChatGPT and Google for queries related to people living with dementia or other cognitive decline and their caregivers (PLWD). The results showed that Google had higher currency and reliability, while ChatGPT provided more objective and relevant responses. However, ChatGPT had lower readability and rarely included the source of a result.
JOURNAL OF MEDICAL INTERNET RESEARCH
(2023)