Letter to the Editor
Can ChatGPT be a new educational tool in medicine?
¿Puede ser ChatGPT una nueva herramienta educativa en medicina?
Read
Not available
Timeswas read the article
Total PDF
Total HTML
Share statistics
array:23 [ "pii" => "S2387020623004011" "issn" => "23870206" "doi" => "10.1016/j.medcle.2023.05.023" "estado" => "S300" "fechaPublicacion" => "2023-10-27" "aid" => "6311" "copyright" => "Elsevier España, S.L.U.. All rights reserved" "copyrightAnyo" => "2023" "documento" => "simple-article" "crossmark" => 1 "subdocumento" => "cor" "cita" => "Med Clin. 2023;161:363-4" "abierto" => array:3 [ "ES" => false "ES2" => false "LATM" => false ] "gratuito" => false "lecturas" => array:1 [ "total" => 0 ] "itemSiguiente" => array:18 [ "pii" => "S2387020623004138" "issn" => "23870206" "doi" => "10.1016/j.medcle.2023.05.024" "estado" => "S300" "fechaPublicacion" => "2023-10-27" "aid" => "6312" "copyright" => "Elsevier España, S.L.U." "documento" => "simple-article" "crossmark" => 1 "subdocumento" => "cor" "cita" => "Med Clin. 2023;161:364-5" "abierto" => array:3 [ "ES" => false "ES2" => false "LATM" => false ] "gratuito" => false "lecturas" => array:1 [ "total" => 0 ] "en" => array:10 [ "idiomaDefecto" => true "cabecera" => "<span class="elsevierStyleTextfn">Letter to the Editor</span>" "titulo" => "Non-Islet Cell Tumour Hypoglycemia" "tienePdf" => "en" "tieneTextoCompleto" => "en" "paginas" => array:1 [ 0 => array:2 [ "paginaInicial" => "364" "paginaFinal" => "365" ] ] "titulosAlternativos" => array:1 [ "es" => array:1 [ "titulo" => "Hipoglucemia tumoral no relacionada con células de islotes" ] ] "contieneTextoCompleto" => array:1 [ "en" => true ] "contienePdf" => array:1 [ "en" => true ] "autores" => array:1 [ 0 => array:2 [ "autoresLista" => "Gema López-Gallardo, Bothayna Oulad Ahmed, Alfonso Soto Moreno" "autores" => array:3 [ 0 => array:2 [ "nombre" => "Gema" "apellidos" => "López-Gallardo" ] 1 => array:2 [ "nombre" => "Bothayna" "apellidos" => "Oulad Ahmed" ] 2 => array:2 [ "nombre" => "Alfonso" "apellidos" => "Soto Moreno" ] ] ] ] ] "idiomaDefecto" => "en" "EPUB" => "https://multimedia.elsevier.es/PublicationsMultimediaV1/item/epub/S2387020623004138?idApp=UINPBA00004N" "url" => "/23870206/0000016100000008/v1_202310190859/S2387020623004138/v1_202310190859/en/main.assets" ] "itemAnterior" => array:19 [ "pii" => "S2387020623003959" "issn" => "23870206" "doi" => "10.1016/j.medcle.2023.05.020" "estado" => "S300" "fechaPublicacion" => "2023-10-27" "aid" => "6309" "copyright" => "Elsevier España, S.L.U." "documento" => "simple-article" "crossmark" => 1 "subdocumento" => "cor" "cita" => "Med Clin. 2023;161:362-3" "abierto" => array:3 [ "ES" => false "ES2" => false "LATM" => false ] "gratuito" => false "lecturas" => array:1 [ "total" => 0 ] "en" => array:11 [ "idiomaDefecto" => true "cabecera" => "<span class="elsevierStyleTextfn">Letter to the Editor</span>" "titulo" => "Isolated hypoglossal nerve palsy as an initial manifestation of a lung adenocarcinoma (occipital condyle syndrome)" "tienePdf" => "en" "tieneTextoCompleto" => "en" "paginas" => array:1 [ 0 => array:2 [ "paginaInicial" => "362" "paginaFinal" => "363" ] ] "titulosAlternativos" => array:1 [ "es" => array:1 [ "titulo" => "Parálisis aislada del nervio hipogloso como manifestación inicial de un adenocarcinoma pulmonar (síndrome del cóndilo occipital)" ] ] "contieneTextoCompleto" => array:1 [ "en" => true ] "contienePdf" => array:1 [ "en" => true ] "resumenGrafico" => array:2 [ "original" => 0 "multimedia" => array:8 [ "identificador" => "fig0005" "etiqueta" => "Figure 1" "tipo" => "MULTIMEDIAFIGURA" "mostrarFloat" => true "mostrarDisplay" => false "figura" => array:1 [ 0 => array:4 [ "imagen" => "gr1.jpeg" "Alto" => 1614 "Ancho" => 2925 "Tamanyo" => 528093 ] ] "detalles" => array:1 [ 0 => array:3 [ "identificador" => "at0005" "detalle" => "Figure " "rol" => "short" ] ] "descripcion" => array:1 [ "en" => "<p id="spar0005" class="elsevierStyleSimplePara elsevierViewall">(A) Detail of the neurological examination showing tongue deviation to the left and ipsilateral atrophy. (B) Cross section of the cranial tomography showing an osteolytic lesion at the level of the left condyle (arrow).</p>" ] ] ] "autores" => array:1 [ 0 => array:2 [ "autoresLista" => "Antonio Bustos-Merlo, Antonio Rosales-Castillo, Jessica Ramírez Taboada" "autores" => array:3 [ 0 => array:2 [ "nombre" => "Antonio" "apellidos" => "Bustos-Merlo" ] 1 => array:2 [ "nombre" => "Antonio" "apellidos" => "Rosales-Castillo" ] 2 => array:2 [ "nombre" => "Jessica" "apellidos" => "Ramírez Taboada" ] ] ] ] ] "idiomaDefecto" => "en" "Traduccion" => array:1 [ "es" => array:9 [ "pii" => "S0025775323003147" "doi" => "10.1016/j.medcli.2023.05.016" "estado" => "S300" "subdocumento" => "" "abierto" => array:3 [ "ES" => false "ES2" => false "LATM" => false ] "gratuito" => false "lecturas" => array:1 [ "total" => 0 ] "idiomaDefecto" => "es" "EPUB" => "https://multimedia.elsevier.es/PublicationsMultimediaV1/item/epub/S0025775323003147?idApp=UINPBA00004N" ] ] "EPUB" => "https://multimedia.elsevier.es/PublicationsMultimediaV1/item/epub/S2387020623003959?idApp=UINPBA00004N" "url" => "/23870206/0000016100000008/v1_202310190859/S2387020623003959/v1_202310190859/en/main.assets" ] "en" => array:13 [ "idiomaDefecto" => true "cabecera" => "<span class="elsevierStyleTextfn">Letter to the Editor</span>" "titulo" => "Can ChatGPT be a new educational tool in medicine?" "tieneTextoCompleto" => true "saludo" => "Dear Editor," "paginas" => array:1 [ 0 => array:2 [ "paginaInicial" => "363" "paginaFinal" => "364" ] ] "autores" => array:1 [ 0 => array:4 [ "autoresLista" => "Yunpeng Luo, Na Hu" "autores" => array:2 [ 0 => array:3 [ "nombre" => "Yunpeng" "apellidos" => "Luo" "referencia" => array:1 [ 0 => array:2 [ "etiqueta" => "<span class="elsevierStyleSup">a</span>" "identificador" => "aff0005" ] ] ] 1 => array:4 [ "nombre" => "Na" "apellidos" => "Hu" "email" => array:1 [ 0 => "15519164340@163.com" ] "referencia" => array:2 [ 0 => array:2 [ "etiqueta" => "<span class="elsevierStyleSup">b</span>" "identificador" => "aff0010" ] 1 => array:2 [ "etiqueta" => "<span class="elsevierStyleSup">*</span>" "identificador" => "cor0005" ] ] ] ] "afiliaciones" => array:2 [ 0 => array:3 [ "entidad" => "Department of Anesthesiology, Guizhou Provincial People's Hospital, Guiyang, Guizhou Province, China" "etiqueta" => "a" "identificador" => "aff0005" ] 1 => array:3 [ "entidad" => "Department of Radiology, The Affiliated Hospital of Guizhou Medical University, Guiyang, Guizhou Province, China" "etiqueta" => "b" "identificador" => "aff0010" ] ] "correspondencia" => array:1 [ 0 => array:3 [ "identificador" => "cor0005" "etiqueta" => "⁎" "correspondencia" => "Corresponding author." ] ] ] ] "titulosAlternativos" => array:1 [ "es" => array:1 [ "titulo" => "¿Puede ser ChatGPT una nueva herramienta educativa en medicina?" ] ] "textoCompleto" => "<span class="elsevierStyleSections"><p id="par0005" class="elsevierStylePara elsevierViewall">It is already widespread in medicine to use artificial intelligence (AI) as a solution to a variety of problems. The medical domain has investigated language models as tools for individualized patient interactions and consumer education about health. Even though these models have shown potential, they have only had limited success when used for testing clinical knowledge via generative question answering (QA).<a class="elsevierStyleCrossRef" href="#bib0030"><span class="elsevierStyleSup">1</span></a></p><p id="par0010" class="elsevierStylePara elsevierViewall">In November 2022, OpenAI released ChatGPT (a chatbot powered by AI) in San Francisco, CA and it has captured the attention of the world.<a class="elsevierStyleCrossRef" href="#bib0035"><span class="elsevierStyleSup">2</span></a> Using both supervised and reinforcement learning, ChatGPT augments OpenAI's GPT-3.5 language models.<a class="elsevierStyleCrossRef" href="#bib0035"><span class="elsevierStyleSup">2</span></a> This AI-powered chatbot has a wide range of capabilities, including writing poems and essays, resolving coding problems, and explaining complex medical concepts, such as “What's the imaging findings of peripheral lung cancer?” Online search has been advanced to a whole new level by ChatGPT, according to some people.<a class="elsevierStyleCrossRef" href="#bib0040"><span class="elsevierStyleSup">3</span></a> With the programme, you will be able to understand complex issues rapidly, in-depth, and generate conversational responses, recalling your own past responses as you ask questions in real-time. Using ChatGPT may represent the first in a new line of clinical knowledge-based models that better capture dialogic interactivity.</p><p id="par0015" class="elsevierStylePara elsevierViewall">The ability of ChatGPT to function as a virtual medical tutor was assessed by Gilson A. et al.<a class="elsevierStyleCrossRef" href="#bib0045"><span class="elsevierStyleSup">4</span></a> They evaluated ChatGPT's responses qualitatively through using rational justification and extrinsic and intrinsic information. And their results demonstrated that ChatGPT can be used as a novel medical educational tool due to its wealth of medical knowledge. Even though the findings are promising, can we really use ChatGPT as a novel medical educational tool?</p><p id="par0020" class="elsevierStylePara elsevierViewall">As such, I experimented with this new AI chatbot. I asked a question: “Can we use ChatGPT as a novel medical educational tool?” in it, and the following was the response it gave:<span class="elsevierStyleDisplayedQuote" id="dsq0005"><p id="spar0005" class="elsevierStyleSimplePara elsevierViewall">Yes, ChatGPT can be used as a novel medical educational tool. By providing information on medical topics in a conversational format, it can help make learning about complex subjects more accessible and engaging for students and healthcare professionals. Additionally, as a language model, ChatGPT can also assist with answering specific questions and clarifying concepts in real-time, providing a convenient and efficient way for individuals to learn and stay informed about medical topics.</p></span><span class="elsevierStyleDisplayedQuote" id="dsq0010"><p id="spar0010" class="elsevierStyleSimplePara elsevierViewall">However, it's important to note that the information provided by ChatGPT is not a substitute for professional medical advice and should not be used to diagnose or treat medical conditions. Always consult with a qualified healthcare provider for personalized medical advice.</p></span></p><p id="par0025" class="elsevierStylePara elsevierViewall">This answer is very scientific, its role is affirmed, but the disadvantages are also emphasized. Mogali S.R.<a class="elsevierStyleCrossRef" href="#bib0050"><span class="elsevierStyleSup">5</span></a> also experimented with ChatGPT by asking questions about anatomy teaching and learning. However, it returned some errors answers. AI chatbots may require years of human supervision to produce accurate and reliable medical information due to the extensive, complex, and dynamic nature of medical information. As a result, when researching medical subjects, it is wise to use standard resources like peer-reviewed articles, lecture notes, and textbooks until it has been demonstrated that ChatGPT is risk-free for medical education.</p><p id="par0030" class="elsevierStylePara elsevierViewall">It seems that large language models (e.g. ChatGPT) could be useful for medical education as well as clinical decision-making in the future. Nonetheless, ChatGPT's medical knowledge based solely on rote memorization of disease mechanisms is not adequate for assessing the skills needed for modern medicine. Choosing the “right” treatment in the clinic requires doctors to discard bias, be creative, and think critically, as well as consider many practical factors. In spite of ChatGPT's significant promise, its deployment into clinical practice will face several challenges.</p><span id="sec0005" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0005">Authors’ contributions</span><p id="par0035" class="elsevierStylePara elsevierViewall">Yunpeng Luo contributed to the conception of article and drafted the manuscript. Na Hu critically analyzed the manuscript and provided valuable feedback. All authors contributed towards the acquisition of data, approved the final version of the manuscript, and agree to be responsible for the accuracy and integrity of the study.</p></span><span id="sec0010" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0010">Ethical considerations</span><p id="par0040" class="elsevierStylePara elsevierViewall">In my article there is no patient data and does not involve human or animal subjects. Ethical approval and informed consent are not applicable.</p></span><span id="sec0015" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0015">Funding</span><p id="par0045" class="elsevierStylePara elsevierViewall">This study was supported partly by <span class="elsevierStyleGrantSponsor" id="gs1">Science and Technology Projects of Guizhou Province</span> (grant no. <span class="elsevierStyleGrantNumber" refid="gs1">Qiankehe Basic-ZK[2022]</span> General 422).</p></span><span id="sec0020" class="elsevierStyleSection elsevierViewall"><span class="elsevierStyleSectionTitle" id="sect0020">Conflicts of interest</span><p id="par0050" class="elsevierStylePara elsevierViewall">None declared.</p></span></span>" "textoCompletoSecciones" => array:1 [ "secciones" => array:5 [ 0 => array:2 [ "identificador" => "sec0005" "titulo" => "Authors’ contributions" ] 1 => array:2 [ "identificador" => "sec0010" "titulo" => "Ethical considerations" ] 2 => array:2 [ "identificador" => "sec0015" "titulo" => "Funding" ] 3 => array:2 [ "identificador" => "sec0020" "titulo" => "Conflicts of interest" ] 4 => array:1 [ "titulo" => "References" ] ] ] "pdfFichero" => "main.pdf" "tienePdf" => true "bibliografia" => array:2 [ "titulo" => "References" "seccion" => array:1 [ 0 => array:2 [ "identificador" => "bibs0015" "bibliografiaReferencia" => array:5 [ 0 => array:3 [ "identificador" => "bib0030" "etiqueta" => "1" "referencia" => array:1 [ 0 => array:2 [ "contribucion" => array:1 [ 0 => array:2 [ "titulo" => "Thinking about GPT3 in-context learning for biomedical IE?" "autores" => array:1 [ 0 => array:2 [ "etal" => true "autores" => array:6 [ 0 => "B.J. Gutiérrez" 1 => "N. Mcneal" 2 => "C. Washington" 3 => "Y. Chen" 4 => "L. Li" 5 => "H. Sun" ] ] ] ] ] "host" => array:2 [ 0 => array:1 [ "Revista" => array:2 [ "tituloSerie" => "Think Again" "fecha" => "2022" ] ] 1 => array:1 [ "WWW" => array:1 [ "link" => "https://arxiv.org/abs/2203.08410v1" ] ] ] ] ] ] 1 => array:3 [ "identificador" => "bib0035" "etiqueta" => "2" "referencia" => array:1 [ 0 => array:2 [ "contribucion" => array:1 [ 0 => array:2 [ "titulo" => "Introducing ChatGPT" "autores" => array:1 [ 0 => array:2 [ "etal" => false "autores" => array:1 [ 0 => "A.I. Open" ] ] ] ] ] "host" => array:1 [ 0 => array:1 [ "Libro" => array:1 [ "fecha" => "2022" ] ] ] ] ] ] 2 => array:3 [ "identificador" => "bib0040" "etiqueta" => "3" "referencia" => array:1 [ 0 => array:2 [ "contribucion" => array:1 [ 0 => array:2 [ "titulo" => "A new chat bot is a ‘code red’ for Google's search business" "autores" => array:1 [ 0 => array:2 [ "colaboracion" => "The New York Times" "etal" => false ] ] ] ] "host" => array:1 [ 0 => array:1 [ "Libro" => array:1 [ "fecha" => "2022" ] ] ] ] ] ] 3 => array:3 [ "identificador" => "bib0045" "etiqueta" => "4" "referencia" => array:1 [ 0 => array:2 [ "contribucion" => array:1 [ 0 => array:2 [ "titulo" => "How does ChatGPT perform on the United States medical licensing examination? The implications of large language models for medical education and knowledge assessment" "autores" => array:1 [ 0 => array:2 [ "etal" => true "autores" => array:6 [ 0 => "A. Gilson" 1 => "C.W. Safranek" 2 => "T. Huang" 3 => "V. Socrates" 4 => "L. Chi" 5 => "R.A. Taylor" ] ] ] ] ] "host" => array:1 [ 0 => array:1 [ "Revista" => array:4 [ "tituloSerie" => "JMIR Med Educ" "fecha" => "2023" "volumen" => "8" "paginaInicial" => "e45312" ] ] ] ] ] ] 4 => array:3 [ "identificador" => "bib0050" "etiqueta" => "5" "referencia" => array:1 [ 0 => array:2 [ "contribucion" => array:1 [ 0 => array:2 [ "titulo" => "Initial impressions of ChatGPT for anatomy education" "autores" => array:1 [ 0 => array:2 [ "etal" => false "autores" => array:1 [ 0 => "S.R. Mogali" ] ] ] ] ] "host" => array:1 [ 0 => array:2 [ "doi" => "10.1002/ase.2261" "Revista" => array:2 [ "tituloSerie" => "Anat Sci Educ" "fecha" => "2023" ] ] ] ] ] ] ] ] ] ] ] "idiomaDefecto" => "en" "url" => "/23870206/0000016100000008/v1_202310190859/S2387020623004011/v1_202310190859/en/main.assets" "Apartado" => array:4 [ "identificador" => "43309" "tipo" => "SECCION" "en" => array:2 [ "titulo" => "Letters to the Editor" "idiomaDefecto" => true ] "idiomaDefecto" => "en" ] "PDF" => "https://static.elsevier.es/multimedia/23870206/0000016100000008/v1_202310190859/S2387020623004011/v1_202310190859/en/main.pdf?idApp=UINPBA00004N&text.app=https://www.elsevier.es/" "EPUB" => "https://multimedia.elsevier.es/PublicationsMultimediaV1/item/epub/S2387020623004011?idApp=UINPBA00004N" ]
