17 Dec. 2021
Led by Dr. Chi-Long, Department of Pathology, Taipei Medical University Hospital (TMUH), our joint study, pathological classification of lung cancer types using annotation-free whole slide training approach deep learning method won the 18th National Innovation Award.
The National Innovation Award was created by the Institute for Biotechnology and Medicine Industry (IBMI) in Taiwan and has become the highest honor for corporations with biomedical technologies and research teams with great R&D potentials.
This awarded study has been published by《Nature Communications》early this year.
Our experiments conducted on a data set of 9,662 lung cancer WSIs reveal that the proposed method achieves areas under the receiver operating characteristic curve of 0.9594 and 0.9414 for adenocarcinoma and squamous cell carcinoma classification on the testing set, respectively. Furthermore, the method demonstrates higher classification performance than multiple-instance learning as well as strong localization results for small lesions through class activation mapping.
由臺北醫學大學附設醫院陳志榮醫師帶領與雲象科技合作的團隊,榮獲第18屆國家新創獎, 以 「AI輔助免人工標註全玻片判讀肺癌數位病理影像」拿下學研新創獎。
這是全新開發、一種無切割、免標註的訓練方法,運用全玻片影像,直接訓練深度學習模型。如此,可減輕病理 AI 訓練的人力標註負擔,進而可將醫院所累積的大量玻片資料,研發AI應用,為數位病理診斷加乘效益。
這項不同於醫業界常見的模型,分類效能高達96%,表現優於國外現有模型,技術領先歐美知名業者。成果已於《Nature Communications》發表,在國際亦獲注目與肯定。
全球數位病理市場預估在2025年將達十億5400萬美元,成長強勁。此案證明我們躋身世界水準的實力,而國家新創獎則更肯定了創新利基、商業化價值,為開拓市場注入更大信心。
相關連結
陳志榮醫師介紹本案 https://is.gd/JXt7VG
共同研究發表 https://is.gd/HAB1nF
Comments