Weidong Xie, Anjia Han, Xunzhang Wang, Tiantian Zhen, Yan Yang, Huijuan Shi, Elyas Mamatkadir and Jinwen Wang
The First Affiliated Hospital, Sun Yat-sen University, China Sun Yat-sen University, China DM Intelligence Ltd., China
Posters & Accepted Abstracts: RRJMHS
Cancer is the second important morbidity and mortality factor among women and the most incident type is breast cancer. The diagnosis of biopsy tissue with hematoxylin and eosin (H&E) stained images is non-trivial and specialists often disagree on the final diagnosis. Actually, computer-aided diagnosis systems contribute to reduce the cost and increase the efficiency of this process. Therefore, we have established a diagnostic tool based on a deep-learning framework for the screening of patients with invasive ductal carcinoma. The dataset of tissue slides used in this project consists of 30,000 samples from eligible patients in our hospital. Available tissue samples above were split into a training set, for learning the CNN parameters, and test set, for evaluating its performance. An accuracy of 94% was obtained for non-cancer (i.e. normal or benign) vs. malignant (i.e. invasive carcinoma). This will be helping specialists identify cancerization which is not visible under a single microscope, and this is just the start of what we have planned.
Weidong Xie is a inventor, founder and CEO of DM Intelligence. Following graduation from Imperial College London with honor in Biological Medicine he took office as Associate Professor in Sun Yat-sen University and Director/ PI in St. Jude Children‘s Research Hospital, USA. His research results in regards to T-cell viral immunity which is listed as the remarkable scientific breakthroughs by famous journals. After a decade of experience in small molecule drug discovery, he leads technology startups successfully and AI in medical imaging & pathology diagnosis is the key point he focuses on.
E-mail: [email protected]