Attend the next MICCAI Industrial Talk - June 18, 2024

Tuesday 11th June 2024

Cancer detection and screening using non-contrast CT imaging

Join us for the next exciting talk in our MICCAI Industrial Talk Series. This webinar will be in two parts and features two wonderful speakers and topics.

Title 1: Large-scale pancreatic cancer detection via non-contrast CT and deep learning
Speaker: Dr. Yingda Xia, Alibaba DAMO Academy USA
Title 2: Esophageal cancer screening via Text-guided Supervision from Reports using non-contrast CT imaging
Speaker: Dr. Jiawen Yao, Alibaba DAMO Academy

Date: June 18, 2024
Time: 10:00 am - 11:00 am (ET) / 2:00 pm - 3:00 pm (UTC)

Registration is free (and required).

Title 1: Large-scale pancreatic cancer detection via non-contrast CT and deep learning

Abstract: Pancreatic ductal adenocarcinoma (PDAC), the most deadly solid malignancy, is typically detected late and at an inoperable stage. Early or incidental detection is associated with prolonged survival, but screening asymptomatic individuals for PDAC using a single test remains unfeasible due to the low prevalence and potential harms of false positives. Non-contrast computed tomography (CT), routinely performed for clinical indications, offers the potential for large-scale screening, however, identification of PDAC using non-contrast CT has long been considered impossible. Here, we develop a deep learning approach, pancreatic cancer detection with artificial intelligence (PANDA), that can detect and classify pancreatic lesions with high accuracy via non-contrast CT. PANDA is trained on a dataset of 3,208 patients from a single center. PANDA achieves an area under the receiver operating characteristic curve (AUC) of 0.986–0.996 for lesion detection in a multicenter validation involving 6,239 patients across 10 centers, outperforms the mean radiologist performance by 34.1% in sensitivity and 6.3% in specificity for PDAC identification, and achieves a sensitivity of 92.9% and specificity of 99.9% for lesion detection in a real-world multi-scenario validation consisting of 20,530 consecutive patients. Notably, PANDA utilized with non-contrast CT shows non-inferiority to radiology reports (using contrast-enhanced CT) in the differentiation of common pancreatic lesion subtypes. PANDA could potentially serve as a new tool for large-scale pancreatic cancer screening.

Bio: Dr. Yingda Xia is a staff algorithm engineer at Alibaba DAMO Academy USA. His research interest lies in biomedical artificial intelligence and computer vision. His current focus is on novel imaging-based algorithms for cancer screening and diagnosis. He received his PhD from Johns Hopkins University, advised by Bloomberg Distinguished Professor Alan Yuille. He has published 30+ peer-reviewed articles in top medicine and AI-related journals and conferences, such as Nature Medicine, Annals of Surgery, CVPR, ECCV, ICCV, NeurIPS, and MICCAI.

Title 2: Esophageal cancer screening via Text-guided Supervision from Reports using non-contrast CT imaging

Abstract: Esophageal cancer is the second most deadly cancer. Although some screening methods have been developed, they are expensive, and might be difficult to apply to the general population. We investigate the feasibility of esophageal tumor detection on the noncontrast CT scan, which could potentially be used for opportunistic cancer screening. Clinical reports can offer a "free lunch'' supervision information and provide tumor location as a type of weak label to cope with screening tasks, thus saving human labeling workloads, if properly leveraged. We propose a novel text-guided learning method to achieve highly accurate cancer detection results. Our quantitative experimental results validate that our approach can reduce human annotation efforts by at least 70% while maintaining comparable cancer detection accuracy to competing fully supervised methods.

Bio: Dr. Jiawen Yao is a senior algorithm expert at Alibaba DAMO Academy. His current research projects focus on developing algorithms to improve the screening, diagnosis, and treatment recommendations in oncology for precision medicine. He received his PhD from University of Texas at Arlington, advised by Jenkins Garrett Professor Junzhou Huang. Dr. Yao has published 50+ peer-reviewed articles in top medicine-related conferences and journals, such as Nature Medicine, Annals of Surgery, Radiology, Clinical Cancer Research, Medical Image Analysis, CVPR, ICCV, and MICCAI. His work DeepPrognosis has been selected for MICCAI-MedIA Special Issue of Best Papers in 2020, and DeepAttnMISL is one of MedIAMost Cited Articles.