Fiona McGowan Martha Morrison, Nima Rezaei, Amanuel Godana Arero, Vasko Graklanov, Sevan Iritsyan, Mariya Ivanovska, Rangariari Makuku, Leander Penaso Marquez, Kseniia Minakova, Lindelwa Phakamile Mmema, Piotr Rzymski, Ganna Zavolodko
Abstract: Artificial intelligence (AI) technologies have already played a revolutionary role in scientific research, from diagnostics to text-generative AI used in scientific writing. The use of AI in the scientific field needs transparent regulation, especially with a longstanding history of use—the first AI technologies in science were developed in the 1950s. Since then, AI has gone from being able to alter texts to producing them using billions of parameters to generate accurate and natural texts. However, scientific work requires high ethical and professional standards, and the rise of AI use in the field has led to many institutions and journals releasing statements and restrictions on its use. AI, being reliant on its users can exacerbate and increase existing biases in the field without being able to take accountability. AI responses can also often lack specificity and depth. However, it is important not to condemn the use of AI in scientific work as a whole. This article has partial use of an AI large language model (LLM), specifically Chatbot Generative Pre-Trained Transformer (ChatGPT), to demonstrate the theories with clear examples. Several recommendations on both a strategic and regulatory level have been formulated in this paper to enable the complementary use of AI alongside ethically-conducted scientific research or for educational purposes, where it shows great potential as a transformative force in interactive work. Policymakers should create wide-reaching, clear guidelines and legal frameworks for using AI to remove the burden of consideration from educators and senior researchers. Caution in the scientific community is advised, though further understanding and work to improve AI use is encouraged.
{"title":"Maintaining scientific integrity and high research standards against the backdrop of rising artificial intelligence use across fields","authors":"Fiona McGowan Martha Morrison, Nima Rezaei, Amanuel Godana Arero, Vasko Graklanov, Sevan Iritsyan, Mariya Ivanovska, Rangariari Makuku, Leander Penaso Marquez, Kseniia Minakova, Lindelwa Phakamile Mmema, Piotr Rzymski, Ganna Zavolodko","doi":"10.21037/jmai-23-63","DOIUrl":"https://doi.org/10.21037/jmai-23-63","url":null,"abstract":"Abstract: Artificial intelligence (AI) technologies have already played a revolutionary role in scientific research, from diagnostics to text-generative AI used in scientific writing. The use of AI in the scientific field needs transparent regulation, especially with a longstanding history of use—the first AI technologies in science were developed in the 1950s. Since then, AI has gone from being able to alter texts to producing them using billions of parameters to generate accurate and natural texts. However, scientific work requires high ethical and professional standards, and the rise of AI use in the field has led to many institutions and journals releasing statements and restrictions on its use. AI, being reliant on its users can exacerbate and increase existing biases in the field without being able to take accountability. AI responses can also often lack specificity and depth. However, it is important not to condemn the use of AI in scientific work as a whole. This article has partial use of an AI large language model (LLM), specifically Chatbot Generative Pre-Trained Transformer (ChatGPT), to demonstrate the theories with clear examples. Several recommendations on both a strategic and regulatory level have been formulated in this paper to enable the complementary use of AI alongside ethically-conducted scientific research or for educational purposes, where it shows great potential as a transformative force in interactive work. Policymakers should create wide-reaching, clear guidelines and legal frameworks for using AI to remove the burden of consideration from educators and senior researchers. Caution in the scientific community is advised, though further understanding and work to improve AI use is encouraged.","PeriodicalId":73815,"journal":{"name":"Journal of medical artificial intelligence","volume":"101 3-4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135714328","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Background: Google Play Store is a popular Android app store where users’ reviews and ratings provide valuable insights. As part of application development, clients and app designers have a significant impact on the market. Accurately predicting market trends is critical to the success of applications, and this is where information mining comes in. By evaluating various factors such as application name, pricing, reviews, and category, we can predict which types of apps are most likely to be successful.
背景:Google Play Store是一个受欢迎的Android应用商店，用户的评论和评级可以提供有价值的见解。作为应用程序开发的一部分，客户和应用程序设计人员对市场有重大影响。准确预测市场趋势对于应用程序的成功至关重要，而这正是信息挖掘的用武之地。通过评估应用名称、定价、评论和类别等各种因素，我们可以预测哪种类型的应用最有可能获得成功。
{"title":"Comparison of machine learning algorithms used to catalog Google Appstore","authors":"Priyadarshini Pattanaik, Dimple Nagpal","doi":"10.21037/jmai-23-58","DOIUrl":"https://doi.org/10.21037/jmai-23-58","url":null,"abstract":"Background: Google Play Store is a popular Android app store where users’ reviews and ratings provide valuable insights. As part of application development, clients and app designers have a significant impact on the market. Accurately predicting market trends is critical to the success of applications, and this is where information mining comes in. By evaluating various factors such as application name, pricing, reviews, and category, we can predict which types of apps are most likely to be successful.","PeriodicalId":73815,"journal":{"name":"Journal of medical artificial intelligence","volume":"50 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135372774","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Background: Large language models (LLMs) such as ChatGPT have emerged as a potentially powerful application in medicine. One of these strengths is the ability for ChatGPT to analyze text and to perform certain tasks. International Classification of Diseases (ICD) codes are universally utilized in medicine and have served as a uniform platform for insurance and billing. However, the task of coding ICDs after each patient encounter is time-consuming on physicians, particularly in fast paced clinics such as retina clinics. Additionally, searching for the most specific, correct ICD code may add additional time, resulting in providers electing for more general ICD codes. LLMs may help to relieve this burden by analyzing notes written by a provider and automatically generate an ICD code that can be used for the encounter.
{"title":"Applying large language model artificial intelligence for retina International Classification of Diseases (ICD) coding","authors":"Joshua Ong, Nikita Kedia, Sanjana Harihar, Sharat Chandra Vupparaboina, Sumit Randhir Singh, Ramesh Venkatesh, Kiran Vupparaboina, Sandeep Chandra Bollepalli, Jay Chhablani","doi":"10.21037/jmai-23-106","DOIUrl":"https://doi.org/10.21037/jmai-23-106","url":null,"abstract":"Background: Large language models (LLMs) such as ChatGPT have emerged as a potentially powerful application in medicine. One of these strengths is the ability for ChatGPT to analyze text and to perform certain tasks. International Classification of Diseases (ICD) codes are universally utilized in medicine and have served as a uniform platform for insurance and billing. However, the task of coding ICDs after each patient encounter is time-consuming on physicians, particularly in fast paced clinics such as retina clinics. Additionally, searching for the most specific, correct ICD code may add additional time, resulting in providers electing for more general ICD codes. LLMs may help to relieve this burden by analyzing notes written by a provider and automatically generate an ICD code that can be used for the encounter.","PeriodicalId":73815,"journal":{"name":"Journal of medical artificial intelligence","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136128165","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Lucas Lacerda de Souza, Felipe Paiva Fonseca, Manoela Domingues Martins, Oslei Paes de Almeida, Helder Antônio Rebelo Pontes, Fábio Luiz Coracin, Márcio Ajudarte Lopes, Syed Ali Khurram, Alan Roger Santos-Silva, Ahmed Hagag, Pablo Agustin Vargas
{"title":"ChatGPT and medicine: a potential threat to science or a step towards the future?","authors":"Lucas Lacerda de Souza, Felipe Paiva Fonseca, Manoela Domingues Martins, Oslei Paes de Almeida, Helder Antônio Rebelo Pontes, Fábio Luiz Coracin, Márcio Ajudarte Lopes, Syed Ali Khurram, Alan Roger Santos-Silva, Ahmed Hagag, Pablo Agustin Vargas","doi":"10.21037/jmai-23-70","DOIUrl":"https://doi.org/10.21037/jmai-23-70","url":null,"abstract":"","PeriodicalId":73815,"journal":{"name":"Journal of medical artificial intelligence","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136153099","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Simon E. Skalicky, Robert N. Weinreb, Nahum Goldmann, Ricardo Augusto Paletta Guedes, Christophe Baudouin, Xiulan Zhang, Aukje van Gestel, Eytan Z. Blumenthal, Paul L. Kaufman, Robert Rothman, Ana Maria Vasquez, Paul Harasymowycz, Derek S. Welsbie, Ivan Goldberg
{"title":"Implementing an artificial intelligence system to comprehensively manage people with glaucoma: a blueprint","authors":"Simon E. Skalicky, Robert N. Weinreb, Nahum Goldmann, Ricardo Augusto Paletta Guedes, Christophe Baudouin, Xiulan Zhang, Aukje van Gestel, Eytan Z. Blumenthal, Paul L. Kaufman, Robert Rothman, Ana Maria Vasquez, Paul Harasymowycz, Derek S. Welsbie, Ivan Goldberg","doi":"10.21037/jmai-23-91","DOIUrl":"https://doi.org/10.21037/jmai-23-91","url":null,"abstract":"","PeriodicalId":73815,"journal":{"name":"Journal of medical artificial intelligence","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135662883","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"The impact of prompt engineering in large language model performance: a psychiatric example","authors":"Declan Grabb","doi":"10.21037/jmai-23-71","DOIUrl":"https://doi.org/10.21037/jmai-23-71","url":null,"abstract":"","PeriodicalId":73815,"journal":{"name":"Journal of medical artificial intelligence","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136153108","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yasanthi Malika Hirimutugoda, Thusari P. Silva, Nimalka M. Wagarachchi
: Certainty is a significant part of disease detection, involving various kinds of imaging and machine learning (ML) methodologies. More precisely than other ML methods, a convolutional neural network (CNN) can classify images. As its parameters are deterministic, it cannot indicate the level of uncertainty in its predictions. Predictions made by predetermined CNNs may yield inaccurate findings, and there is no evaluation of confidence in these results. These outcomes may have harmful effects and lack trustworthiness. Uncertainty quantification (UQ) is critical to evaluating confidence in prediction. The noise, illumination, segmentation, and edge issues common to medical images also impact pre-trained CNN algorithms and lead to uncertain outcomes. This review aims to investigate the main inherent uncertainty issue in CNN and what form of UQ method can be applied with CNN to the task of medical image classification. This research proposes a novel approach by combining the superior properties of the Bayesian approach and fusion methods to reduce the uncertainty in CNN models. This study concludes that, despite a number of unresolved technical and scientific issues, various types of fusion approaches have improved the clinical validity for diagnosing and analytical purposes, and it is a field of study that has the capacity to grow significantly in the years to come.
{"title":"Handling the predictive uncertainty of convolutional neural network in medical image analysis: a review","authors":"Yasanthi Malika Hirimutugoda, Thusari P. Silva, Nimalka M. Wagarachchi","doi":"10.21037/jmai-23-40","DOIUrl":"https://doi.org/10.21037/jmai-23-40","url":null,"abstract":": Certainty is a significant part of disease detection, involving various kinds of imaging and machine learning (ML) methodologies. More precisely than other ML methods, a convolutional neural network (CNN) can classify images. As its parameters are deterministic, it cannot indicate the level of uncertainty in its predictions. Predictions made by predetermined CNNs may yield inaccurate findings, and there is no evaluation of confidence in these results. These outcomes may have harmful effects and lack trustworthiness. Uncertainty quantification (UQ) is critical to evaluating confidence in prediction. The noise, illumination, segmentation, and edge issues common to medical images also impact pre-trained CNN algorithms and lead to uncertain outcomes. This review aims to investigate the main inherent uncertainty issue in CNN and what form of UQ method can be applied with CNN to the task of medical image classification. This research proposes a novel approach by combining the superior properties of the Bayesian approach and fusion methods to reduce the uncertainty in CNN models. This study concludes that, despite a number of unresolved technical and scientific issues, various types of fusion approaches have improved the clinical validity for diagnosing and analytical purposes, and it is a field of study that has the capacity to grow significantly in the years to come.","PeriodicalId":73815,"journal":{"name":"Journal of medical artificial intelligence","volume":"206 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136055906","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Mohammed Sedki, Nathan Vidal, Paul Roux, Caroline Barry, Mario Speranza, Bruno Falissard, Eric Brunet-Gouet
Background: This paper proposes a proof of concept of using natural language processing (NLP) techniques to categorize valence of family relationships described in free texts written by French teenagers. The proposed study traces the evolution of techniques for word embedding.
{"title":"Using a self-attention architecture to automate valence categorization of French teenagers’ free descriptions of their family relationships: a proof of concept","authors":"Mohammed Sedki, Nathan Vidal, Paul Roux, Caroline Barry, Mario Speranza, Bruno Falissard, Eric Brunet-Gouet","doi":"10.21037/jmai-23-8","DOIUrl":"https://doi.org/10.21037/jmai-23-8","url":null,"abstract":"Background: This paper proposes a proof of concept of using natural language processing (NLP) techniques to categorize valence of family relationships described in free texts written by French teenagers. The proposed study traces the evolution of techniques for word embedding.","PeriodicalId":73815,"journal":{"name":"Journal of medical artificial intelligence","volume":"97 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135427677","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
James Behrmann, Ellen M. Hong, Shannon Meledathu, Aliza Leiter, Michael Povelaitis, Mariela Mitre
Background: Large language models (LLMs) like chat generative pre-trained transformer (ChatGPT) have gained popularity in healthcare by performing at or near the passing threshold for the United States Medical Licensing Exam (USMLE), but some limitations should be considered. Dermatology is a specialized medical field that relies heavily on visual recognition and images for diagnosis. This paper aimed to measure ChatGPT’s abilities to answer dermatology questions and compare this sub-specialty accuracy to its overall scores on USMLE Step exams.
{"title":"Chat generative pre-trained transformer’s performance on dermatology-specific questions and its implications in medical education","authors":"James Behrmann, Ellen M. Hong, Shannon Meledathu, Aliza Leiter, Michael Povelaitis, Mariela Mitre","doi":"10.21037/jmai-23-47","DOIUrl":"https://doi.org/10.21037/jmai-23-47","url":null,"abstract":"Background: Large language models (LLMs) like chat generative pre-trained transformer (ChatGPT) have gained popularity in healthcare by performing at or near the passing threshold for the United States Medical Licensing Exam (USMLE), but some limitations should be considered. Dermatology is a specialized medical field that relies heavily on visual recognition and images for diagnosis. This paper aimed to measure ChatGPT’s abilities to answer dermatology questions and compare this sub-specialty accuracy to its overall scores on USMLE Step exams.","PeriodicalId":73815,"journal":{"name":"Journal of medical artificial intelligence","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134914311","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Ansh Roge, Patrick Ting, Andrew Chern, William Ting
Background: Skin lesion classification through dermatoscopic images is the most common method for non-invasive diagnostics of dermatologic conditions. Feature extraction through deep learning (DL) based convolutional neural networks (CNNs) provides insight into differential attributes of skin lesions that may pertain to its malignancy. In this study, we sought to improve the performance of standard CNN architectures in skin lesion classification by providing a machine learning (ML)-derived risk score from patient demographic data.
{"title":"Deep ensemble learning using a demographic machine learning risk stratifier for binary classification of skin lesions using dermatoscopic images","authors":"Ansh Roge, Patrick Ting, Andrew Chern, William Ting","doi":"10.21037/jmai-23-38","DOIUrl":"https://doi.org/10.21037/jmai-23-38","url":null,"abstract":"Background: Skin lesion classification through dermatoscopic images is the most common method for non-invasive diagnostics of dermatologic conditions. Feature extraction through deep learning (DL) based convolutional neural networks (CNNs) provides insight into differential attributes of skin lesions that may pertain to its malignancy. In this study, we sought to improve the performance of standard CNN architectures in skin lesion classification by providing a machine learning (ML)-derived risk score from patient demographic data.","PeriodicalId":73815,"journal":{"name":"Journal of medical artificial intelligence","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135427672","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：