[{"data":1,"prerenderedAt":220},["ShallowReactive",2],{"blog-\u002Fblog\u002Fzh-hant\u002Fhow-to-evaluate-skin":3,"blog-\u002Fblog\u002Fzh-hant\u002Fhow-to-evaluate-skin-surround":194,"blog-\u002Fblog\u002Fzh-hant\u002Fhow-to-evaluate-skin-translations":203},{"id":4,"title":5,"author":6,"body":9,"date":178,"description":179,"draft":180,"extension":181,"image":182,"meta":183,"navigation":185,"path":186,"seo":187,"stem":188,"tags":189,"__hash__":193},"blogZhHant\u002Fblog\u002Fzh-hant\u002Fhow-to-evaluate-skin.md","用 AI 評估肌膚這件事",{"name":7,"picture":8},"Soo Ick Cho","\u002Fimages\u002Fblog\u002Fauthor-sooickcho.jpg",{"type":10,"value":11,"toc":170},"minimark",[12,16,21,24,27,31,34,37,40,43,46,49,53,56,62,65,68,120,123],[13,14,15],"p",{},"上一篇文章裡，我談到測量肌膚的「測量」本身，比想像中浮動得多。這時自然會有這樣的想法：「如果儀器不可靠，那讓受過訓練的醫師親眼看、直接判斷不就好了？」我自己也曾長期這麼相信。然而在做 AI 研究的過程中，我卻不得不面對一件事：就連「醫師的判斷」，也無法乾淨俐落地收斂成單一的正確答案。",[17,18,20],"h2",{"id":19},"我的第一篇論文是鑑別唇部的皮膚癌","我的第一篇論文，是鑑別「唇部的皮膚癌」",[13,22,23],{},"在大學附設醫院時，我的第一個研究主題，是用 AI 鑑別長在嘴唇上的罕見皮膚癌。當時用 AI「診斷」疾病的研究才剛開始出現，這項研究因為連不常見的皮膚癌類型都能分辨出來而獲得肯定，得以刊登在皮膚科領域的重要期刊（《英國皮膚科學期刊》，BJD）上。",[13,25,26],{},"診斷相對容易處理，因為像「是不是皮膚癌」這樣的問題，一般被認為有正確答案。（嚴格說來，診斷其實也更接近機率。）不過這類診斷研究當時已經迅速趨於飽和，於是我把目光轉向下一個問題：「嚴重程度評估」。",[17,28,30],{"id":29},"嚴重程度沒有唯一的正確答案","「嚴重程度」沒有唯一的正確答案",[13,32,33],{},"接著，我進行了用 AI 評估痤瘡與異位性皮膚炎嚴重程度的研究。兩項研究都因為創新而獲得肯定，刊登在不錯的期刊上，但在研究過程中真正撞上的牆，卻是另一回事。",[13,35,36],{},"要訓練 AI，就需要「標註資料（ground truth）」。於是我們讓多位皮膚科醫師看同一張照片並評定嚴重程度，結果答案彼此分歧。大多數情況下大家的答案相近，完全南轅北轍的情況很少見（例如 4 級當中，有人評 3 級、有人評 2 級），但即便如此，「唯一的正確答案」並不存在。",[13,38,39],{},"和「是不是皮膚癌」不同，嚴重程度是一個多位專家的判斷會散成一個分布的問題。因此需要和診斷不同的做法，而我的研究，正是在解決這個環節。",[17,41,42],{"id":42},"這不只是皮膚科的問題",[13,44,45],{},"後來我在醫療 AI 公司（Lunit）開發病理分析 AI 時，也遇到了一模一樣的牆。我們用病理科醫師在組織影像上標記的答案來訓練模型，但就連那些答案，也因醫師而異。同一塊組織，有人判斷為癌前病變，有人判斷為惡性病變。",[13,47,48],{},"那時我才明白，我在皮膚科遇到的問題，並非皮膚科獨有，而是整個醫學領域底下普遍存在的問題。歷經皮膚科 AI 研究與病理 AI 研究，我從經驗中學會了該如何用 AI 處理這種「發散的正確答案」。",[17,50,52],{"id":51},"所以怎麼評估才是關鍵","所以，「怎麼評估」才是關鍵",[13,54,55],{},"關鍵不在於硬塞給 AI 一個嚴格的唯一正確答案，而是讓它學習實際醫療現場中存在的判斷分布，也就是所謂的「機率」。不是因為某一個人錯了就把他的意見刪掉，而是把多位專家判斷所構成的範圍，原封不動地當作訓練的素材。我在首爾大學醫院與 Lunit 都以這種方式開發 AI 模型，也確認了這種做法更適合嚴重程度評估。而當時累積的訣竅，如今正應用在 INSKINVIEW 的 AI 模型開發上。",[13,57,58],{},[59,60,61],"strong",{},"參考文獻",[13,63,64],{},"以下是本文提到的研究。",[13,66,67],{},"皮膚科 AI 研究",[69,70,71,90,105],"ul",{},[72,73,74,75,79,80],"li",{},"唇部皮膚癌鑑別 — ",[76,77,78],"em",{},"British Journal of Dermatology",", 2020;182:1388-1394.\n",[69,81,82],{},[72,83,84],{},[85,86,87],"a",{"href":87,"rel":88},"https:\u002F\u002Facademic.oup.com\u002Fbjd\u002Farticle-abstract\u002F182\u002F6\u002F1388\u002F6697632",[89],"nofollow",[72,91,92,93,96,97],{},"痤瘡嚴重程度評估 — ",[76,94,95],{},"American Journal of Clinical Dermatology",", 2023;24:649-659.\n",[69,98,99],{},[72,100,101],{},[85,102,103],{"href":103,"rel":104},"https:\u002F\u002Flink.springer.com\u002Farticle\u002F10.1007\u002Fs40257-023-00777-5",[89],[72,106,107,108,111,112],{},"異位性皮膚炎嚴重程度評估 — ",[76,109,110],{},"IEEE Journal of Biomedical and Health Informatics",", 2023;27:166-175.\n",[69,113,114],{},[72,115,116],{},[85,117,118],{"href":118,"rel":119},"https:\u002F\u002Fieeexplore.ieee.org\u002Fabstract\u002Fdocument\u002F9932872\u002F",[89],[13,121,122],{},"病理科 AI 研究",[69,124,125,140,155],{},[72,126,127,128,131,132],{},"非小細胞肺癌病理分析 AI（PD-L1 判讀） — ",[76,129,130],{},"European Journal of Cancer",", 2022;170:17-26.\n",[69,133,134],{},[72,135,136],{},[85,137,138],{"href":138,"rel":139},"https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fabs\u002Fpii\u002FS0959804922002155",[89],[72,141,142,143,146,147],{},"乳癌病理分析 AI（腫瘤浸潤淋巴球評估） — ",[76,144,145],{},"npj Breast Cancer",", 2023;9:71.\n",[69,148,149],{},[72,150,151],{},[85,152,153],{"href":153,"rel":154},"https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41523-023-00577-4",[89],[72,156,157,158,161,162],{},"乳癌病理分析 AI（HER2·ER·PR 判讀） — ",[76,159,160],{},"Breast Cancer Research",", 2024;26:31.\n",[69,163,164],{},[72,165,166],{},[85,167,168],{"href":168,"rel":169},"https:\u002F\u002Flink.springer.com\u002Farticle\u002F10.1186\u002Fs13058-024-01784-y",[89],{"title":171,"searchDepth":172,"depth":172,"links":173},"",2,[174,175,176,177],{"id":19,"depth":172,"text":20},{"id":29,"depth":172,"text":30},{"id":42,"depth":172,"text":42},{"id":51,"depth":172,"text":52},"2026-07-19","皮膚的嚴重程度，並沒有唯一的正確答案。一位皮膚科專科醫師歷經大學附設醫院與醫療 AI 公司，所面對的「醫師的判斷有多可信」這個問題，以及他如何用 AI 來解決的故事。",false,"md","\u002Fimages\u002Fblog\u002FChatGPT Image 2026년 7월 19일 오후 10_37_40.png",{"slug":184},"how-to-evaluate-skin",true,"\u002Fblog\u002Fzh-hant\u002Fhow-to-evaluate-skin",{"title":5,"description":179},"blog\u002Fzh-hant\u002Fhow-to-evaluate-skin",[190,191,192],"肌膚評估","醫療 AI","INSKINVIEW","Ox6KB1YOFecH2JKP721i17LpHBXrIBWJrRucmWB2ywk",[195,199],{"title":196,"path":197,"stem":198,"children":-1},"為什麼要用手機做肌膚分析","\u002Fblog\u002Fzh-hant\u002Fwhy-smartphone","blog\u002Fzh-hant\u002Fwhy-smartphone",{"title":200,"path":201,"stem":202,"children":-1},"把肌膚化為資料來測量：一位皮膚科醫師的記錄","\u002Fblog\u002Fzh-hant\u002Fmeasuring-skin-as-data","blog\u002Fzh-hant\u002Fmeasuring-skin-as-data",[204,208,212,216],{"code":205,"name":206,"path":207},"ko","한국어","\u002Fblog\u002Fhow-to-evaluate-skin",{"code":209,"name":210,"path":211},"en","English","\u002Fblog\u002Fen\u002Fhow-to-evaluate-skin",{"code":213,"name":214,"path":215},"ja","日本語","\u002Fblog\u002Fja\u002Fhow-to-evaluate-skin",{"code":217,"name":218,"path":219},"zh-hans","简体中文","\u002Fblog\u002Fzh-hans\u002Fhow-to-evaluate-skin",1791649183556]