ChatGPT supera médicos para fazer diagnóstico em consulta, diz estudo

[{"selector":"#anim-77ddf9a0-c0c6-43b9-97d6-fcb18d4f715b","keyframes":{"opacity":[0,1]},"delay":0,"duration":600,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"both"}] [{"selector":"#anim-7b3a9b4f-2c84-4b05-a1b4-fcaeee8831a2","keyframes":{"opacity":[0,1]},"delay":0,"duration":1000,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"both"}] [{"selector":"#anim-fc53717f-9539-44a9-b03c-ea0ae7c85659","keyframes":{"opacity":[0,1]},"delay":0,"duration":500,"easing":"cubic-bezier(.3,0,.55,1)","fill":"both"}] [{"selector":"#anim-45974232-8929-4d73-8c16-f073ce9ff0c6","keyframes":{"transform":["scale(0.3333333333333333)","scale(1)"]},"delay":0,"duration":500,"easing":"cubic-bezier(.3,0,.55,1)","fill":"forwards"}] Matheus Bertelli/Pexels

[{"selector":"#anim-6b57d965-34af-4cb0-8d4c-ca261c1984bc","keyframes":{"opacity":[0,1]},"delay":0,"duration":600,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"both"}] [{"selector":"#anim-67824f17-2251-4cb6-b4e5-cbbcffd13907","keyframes":{"opacity":[0,1]},"delay":0,"duration":1500,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"both"}] [{"selector":"#anim-08d1f501-b0f1-4f6d-8af1-fc3b2626de0c","keyframes":{"opacity":[0,1]},"delay":0,"duration":500,"easing":"cubic-bezier(.3,0,.55,1)","fill":"both"}] [{"selector":"#anim-9067f4ce-3f34-4835-b049-bf45ec5b6e65","keyframes":{"transform":["scale(0.3333333333333333)","scale(1)"]},"delay":0,"duration":500,"easing":"cubic-bezier(.3,0,.55,1)","fill":"forwards"}] Cientistas do hospital afiliado da Universidade de Harvard conduziram um estudo que concluiu que o raciocínio clínico do ChatGPT-4 pode ser igual ou mais preciso do que dos médicos Freepik

[{"selector":"#anim-a2829652-6819-4e26-9cb0-5c6ebef3ba11","keyframes":{"opacity":[0,1]},"delay":0,"duration":600,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"both"}] [{"selector":"#anim-7712a6b8-8d38-422a-afb3-950d65262ab7","keyframes":{"opacity":[0,1]},"delay":0,"duration":1500,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"both"}] [{"selector":"#anim-2d8f42c7-d71b-4d4e-a67c-f7e6fb71d749","keyframes":{"opacity":[0,1]},"delay":0,"duration":500,"easing":"cubic-bezier(.3,0,.55,1)","fill":"both"}] [{"selector":"#anim-ba146403-ba9d-4548-aa3a-693a53f0ee24","keyframes":{"transform":["scale(0.3333333333333333)","scale(1)"]},"delay":0,"duration":500,"easing":"cubic-bezier(.3,0,.55,1)","fill":"forwards"}] Na pesquisa, foram comparadas as justificativas de diagnóstico de 20 casos escritas por médicos assistentes, por residentes e pela inteligência artificial (IA) Pressfoto/Freepik

[{"selector":"#anim-f16e8de7-d1d7-41a4-b84f-24450e098923","keyframes":{"opacity":[0,1]},"delay":0,"duration":600,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"both"}] [{"selector":"#anim-4647b3b9-1032-44c7-b4f3-15b7bcd8c834","keyframes":{"opacity":[0,1]},"delay":0,"duration":1500,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"both"}] [{"selector":"#anim-cc36702d-dc35-4512-98fe-7f6f9bcc91c8","keyframes":{"opacity":[0,1]},"delay":0,"duration":500,"easing":"cubic-bezier(.3,0,.55,1)","fill":"both"}] [{"selector":"#anim-ef5d571a-81ea-420f-b7bc-a38a8810a654","keyframes":{"transform":["scale(0.3333333333333333)","scale(1)"]},"delay":0,"duration":500,"easing":"cubic-bezier(.3,0,.55,1)","fill":"forwards"}] Para a condução do estudo foi utilizado o r-IDEA, ferramenta existente que atribui uma nota para o raciocínio clínico de profissionais de saúde Freepik

[{"selector":"#anim-f0f77207-ff19-4c48-b10f-564f17091e3e","keyframes":{"opacity":[0,1]},"delay":0,"duration":600,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"both"}] [{"selector":"#anim-6e8d03ff-5968-42b5-ba10-cf2db6112f08","keyframes":{"opacity":[0,1]},"delay":0,"duration":1500,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"both"}] [{"selector":"#anim-dd5b3f0f-30b9-45d7-99a1-ee9d5808db72","keyframes":{"opacity":[0,1]},"delay":0,"duration":500,"easing":"cubic-bezier(.3,0,.55,1)","fill":"both"}] [{"selector":"#anim-05c9a557-30a4-45e9-b645-0efcaff26167","keyframes":{"transform":["scale(0.3333333333333333)","scale(1)"]},"delay":0,"duration":500,"easing":"cubic-bezier(.3,0,.55,1)","fill":"forwards"}] rawpixel.com/Freepik Na pontuação média obtida pelo r-IDEA, o ChatGPT-4 superou todos os outros participantes, conseguindo 10, a nota máxima; já os médicos assistentes alcançaram nove e os residentes, oito

[{"selector":"#anim-df603e01-1230-4afe-9053-702a512ae5a4","keyframes":{"opacity":[0,1]},"delay":0,"duration":600,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"both"}] [{"selector":"#anim-312d4e25-1c1f-4309-aa53-183bdff41357","keyframes":{"opacity":[0,1]},"delay":0,"duration":1500,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"both"}] [{"selector":"#anim-54a1cc9b-7798-4cfc-a2f3-ac5cccf95e62","keyframes":{"opacity":[0,1]},"delay":0,"duration":500,"easing":"cubic-bezier(.3,0,.55,1)","fill":"both"}] [{"selector":"#anim-58c1111f-7e9f-444a-9790-0446c824d32b","keyframes":{"transform":["scale(0.3333333333333333)","scale(1)"]},"delay":0,"duration":500,"easing":"cubic-bezier(.3,0,.55,1)","fill":"forwards"}] jcomp/Freepik Sobre precisão de diagnóstico, o robô e os profissionais de saúde tiveram empate, mas a IA tinha mais informações incorretas no raciocínio clínico

[{"selector":"#anim-0b1eb544-72fe-47c7-a056-101567564526","keyframes":{"opacity":[0,1]},"delay":0,"duration":600,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"both"}] [{"selector":"#anim-41e78d21-d705-4149-a4a5-4e3c38e12d63","keyframes":{"opacity":[0,1]},"delay":0,"duration":1500,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"both"}] [{"selector":"#anim-ed690fd5-05e2-4c71-bdfb-6f6ca498f4d2","keyframes":{"opacity":[0,1]},"delay":0,"duration":500,"easing":"cubic-bezier(.3,0,.55,1)","fill":"both"}] [{"selector":"#anim-50cab706-d0c4-4be4-a18c-a273e4541118","keyframes":{"transform":["scale(0.3333333333333333)","scale(1)"]},"delay":0,"duration":500,"easing":"cubic-bezier(.3,0,.55,1)","fill":"forwards"}] Levart_Photographer/Unsplash Os pesquisadores concluíram que a ferramenta pode ser útil para auxiliar nos diagnósticos e prevenir erros, mas não substitui os médicos

CNNBRASIL.COM.BR

[{"selector":"#anim-435e2b0c-0e66-46f0-9c2e-237433757e5c","keyframes":{"opacity":[0,1]},"delay":0,"duration":600,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"both"}] [{"selector":"#anim-c73e1d18-aa11-470f-919b-c39139e4ed0d","keyframes":{"opacity":[0,1]},"delay":0,"duration":600,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"both"}] [{"selector":"#anim-53c70625-e597-4383-b378-26bdcd30105d","keyframes":{"opacity":[0,1]},"delay":0,"duration":1000,"easing":"cubic-bezier(0.2, 0.6, 0.0, 1)","fill":"both"}] [{"selector":"#anim-e5f7eb26-6783-44bf-ae58-8c549f8c3c41","keyframes":{"transform":["translate3d(0px, 2132.63615%, 0)","translate3d(0px, 0px, 0)"]},"delay":0,"duration":1000,"easing":"cubic-bezier(0.2, 0.6, 0.0, 1)","fill":"both"}] [{"selector":"#anim-0ec7bac1-9051-4912-8459-629099b03b6c","keyframes":{"opacity":[0,1]},"delay":0,"duration":600,"easing":"cubic-bezier(0.2, 0.6, 0.0, 1)","fill":"both"}] [{"selector":"#anim-1e45b02c-be11-4723-8b2c-7aa153d08f6f","keyframes":{"transform":["translate3d(-117.84387%, 0px, 0)","translate3d(0px, 0px, 0)"]},"delay":0,"duration":600,"easing":"cubic-bezier(0.2, 0.6, 0.0, 1)","fill":"both"}] [{"selector":"#anim-ec225866-b68f-49a5-9cfd-171c12108f54","keyframes":{"opacity":[0,1]},"delay":0,"duration":1000,"easing":"cubic-bezier(0.2, 0.6, 0.0, 1)","fill":"both"}] [{"selector":"#anim-5a7f01c2-6398-4a3f-9a86-49d64b549cf5","keyframes":{"transform":["translate3d(0px, 1223.93157%, 0)","translate3d(0px, 0px, 0)"]},"delay":0,"duration":1000,"easing":"cubic-bezier(0.2, 0.6, 0.0, 1)","fill":"both"}] leia mais em frimufilms/Freepik