aetherAI’s Machine Learning Approach for T-cell Lymphomas Diagnosis Support published in Cancers

Updated: Dec 21, 2021

Oct. 2021


aetherAI presented a machine learning approach for accurate quantification of nuclear morphometrics and differential diagnosis of primary intestinal T-cell lymphomas. The human interpretable machine learning approach can be easily applied to other lymphomas and potentially even broader disease categories. This approach not only brings deeper insights into lymphoma phenotypes but also paves the way for future discoveries concerning their relationship with disease classification and outcome.


This study has collaborated with Dr. Shih-Sung Chuang, Department of Pathology, Chi-Mei Medical Center, and 4 other departments of Pathology. The study, < Machine Learning Based on Morphological Features Enables Classification of Primary Intestinal T-Cell Lymphomas> is accepted and published by Cancers, a journal of oncology in Oct. 2021.


Full article: https://is.gd/8hc7rI


雲象科技與永康奇美醫院莊世松醫師合作,成功將AI應用在醫療影像,進行細胞核量化型態分析,協助醫師對淋巴瘤有更細緻、精準的診斷方法,施以更精確的治療。


T細胞淋巴瘤是少見的疾病,精確的診斷仰賴病理科醫師對於腫瘤組織切片的型態分析。其中有兩類在形態上難以區分的疾病:1、MEITL,單形性上皮腸T細胞淋巴瘤;2、ITCL-NOS,腸道T細胞淋巴瘤。透過細胞核大小變異度、形態分布來區分這兩類疾病,對病理科醫師挑戰甚鉅。


雲象AI科學家運用數千顆細胞核的標註資料,訓練出的AI模型能對淋巴瘤的細胞核精確地偵測,描繪輪廓,並且進一步計算出每個細胞核面積大小,長短軸比例等可量化的形態資訊,依此進一步訓練機器學習演算法,對MEITL以及ITCL-NOS進行分類,預測水準可高達AUC 96.6%。


藉此AI量化工具,醫師不再需要對T淋巴瘤的形態作粗糙的二元分類,而可用量化的形態分析數據,對淋巴瘤細胞進行精確的統計描述。


研究發表於國際知名期刊《Cancers》,全文請見 https://is.gd/8hc7rI