{"id":25650,"date":"2024-12-25T18:00:40","date_gmt":"2024-12-25T10:00:40","guid":{"rendered":"https:\/\/www.1ai.net\/?p=25650"},"modified":"2024-12-25T18:00:40","modified_gmt":"2024-12-25T10:00:40","slug":"%e6%96%b0%e7%a0%94%e7%a9%b6%e5%8f%91%e7%8e%b0-openai-%e7%9a%84-o1-preview-ai-%e6%a8%a1%e5%9e%8b%e5%9c%a8%e8%af%8a%e6%96%ad%e6%a3%98%e6%89%8b%e5%8c%bb%e7%96%97%e6%a1%88%e4%be%8b%e6%96%b9%e9%9d%a2","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/25650.html","title":{"rendered":"New Study Finds OpenAI's o1-preview AI Model Outperforms Doctors in Diagnosing Tricky Medical Cases"},"content":{"rendered":"<p>Dec. 25, 2012 - A team of researchers from Harvard Medical School and Stanford University has conducted an in-depth evaluation of medical diagnostic <a href=\"https:\/\/www.1ai.net\/en\/tag\/openai\" title=\"[View articles tagged with [OpenAI]]\" target=\"_blank\" >OpenAI<\/a> of the o1-preview model.<strong>found to be better than human doctors at diagnosing tricky<a href=\"https:\/\/www.1ai.net\/en\/tag\/%e5%8c%bb%e7%96%97\" title=\"[View articles tagged with [medical]]\" target=\"_blank\" >Medical<\/a>Case in point.<\/strong><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-25651\" title=\"8cd83fd9j00sp1ns9000id000gm006xp\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2024\/12\/8cd83fd9j00sp1ns9000id000gm006xp.jpg\" alt=\"8cd83fd9j00sp1ns9000id000gm006xp\" width=\"598\" height=\"249\" \/><\/p>\n<p>According to the study, o1-preview correctly diagnosed 78.31 TP3T of test cases, and was even more accurate at 88.61 TP3T in 70 case-specific comparison tests, significantly better than its predecessor, GPT-4, at 72.91 TP3T.<\/p>\n<p>Using the Standardized Scale for the Assessment of the Quality of Medical Reasoning R-IDEA, o1-preview achieved 78 perfect scores out of 80 cases. In comparison, experienced physicians achieved perfect scores in only 28 cases, and residents in only 16 cases.<\/p>\n<p>In complex cases designed by 25 experts, o1-preview scored as high as 861 TP3T, more than twice as high as physicians using GPT-4 (411 TP3T) and those using traditional tools (341 TP3T).<\/p>\n<p>The researchers acknowledged the limitations of the test in that some of the test cases may have been included in o1-preview's training data and that the test focused primarily on the system working alone and did not adequately consider scenarios in which it would work in concert with a human physician; in addition, the diagnostic test suggested by o1-preview is costly and has limitations in practical application.<\/p>","protected":false},"excerpt":{"rendered":"<p>On December 25, a scientific team from Harvard Medical School and Stanford University conducted an in-depth assessment of OpenAI ' s O1-preview model in the field of medical diagnosis, which found that it was better at diagnosing difficult medical cases than human doctors. According to the study, o1-preview correctly diagnosed the test cases of 78.3%, and the accuracy rate was as high as 88.6% in the comparison of 70 specific cases, significantly better than the 72.9% of its predecessor GPT-4. R-IDEA, o1-preview, used the medical reasoning quality assessment standard, achieved 78 full points in 80 cases. In contrast, experienced doctors received full scores in 28 cases<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[146],"tags":[190,211],"collection":[],"class_list":["post-25650","post","type-post","status-publish","format-standard","hentry","category-news","tag-openai","tag-211"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/25650","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/comments?post=25650"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/25650\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=25650"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=25650"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=25650"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=25650"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}