MICCAI(International Conference on Medical Image Computing and Computer
Assisted Intervention, MICCAI)
是聚焦于医疗图像处理和辅助治疗技术方面的综合性学术顶级年会,由国际医学图像计算和计算机辅助干预协会主办。在2019年投稿量比2018年增加近70%的背景下,平安科技人工智能中心产出的5篇高分命中论文,展现了极高的研究质量。
RSNA(Radiology Society of North American,
RSNA)则是放射医学界的盛会和放射医学临床研究的风向标,每年11月底至12月初在美国芝加哥召开,是全球放射学临床研究成果和主要医疗器械厂商产品发布的重要场合,会议论文代表了放射医学界医学应用研究的最高水平。
平安科技人工智能中心和长庚医院著名的创伤急诊中心合作,提出新的检测髋关节和盆腔所有部位骨折的AI技术(Weakly Supervised
Universal Fracture Detection in Pelvic
X-ray),该技术对髋关节各部位骨折的敏感度和特异度均达到95%以上,而对于常见于老年人的股骨骨折则达到99%的高敏感度。该算法已经达到国际领先医院的创伤外科、急诊科医生的水平。
图一:髋部骨折自动检测系统,红色块为AI自动检出的骨折部位
Y. Wang, et al., "Weakly Supervised Universal Fracture Detection in Pelvic
X-ray", International Conference on Medical Image Computing and Computer
Assisted Intervention (MICCAI), Shenzhen, China, 2019 (early accept)
C. Cheng et al., Universal High Performance Pelvic/Hip Fracture Detection on
Pelvic Radiographs of Trauma Patients using Cascaded Deep Networks, RSNA,
Chicago, USA, 2019 (Scientific Oral Paper)
Y. Tang, Y. Tang, et al., "TUNA-Net: Task-oriented UNsupervised Adversarial
Network for Disease Recognition in Cross-Domain Chest X-rays". International
Conference on Medical Image Computing and Computer Assisted Intervention
(MICCAI), Shenzhen, China, 2019
Y. Tang, Y. Tang, et al., "Task-oriented Unsupervised Adversarial Network for
Disease Recognition in Cross-Domain Chest Radiographs", RSNA, Chicago, USA, 2019
(Scientific Oral Paper)
D. Jin, D. Guo, et al., "Deep Esophageal Clinical Target Volume Delineation
using Encoded 3D Spatial Context of Tumor, Lymph Nodes, and Organs At Risk",
International Conference on Medical Image Computing and Computer Assisted
Intervention (MICCAI), Shenzhen, China, 2019 (early accept)
D. Jin, D. Guo, et al., "Accurate Esophageal Gross Tumor Volume Segmentation
in PET/CT using Two-Stream Chained 3D Deep Network Fusion", International
Conference on Medical Image Computing and Computer Assisted Intervention
(MICCAI), Shenzhen, China, 2019 (early accept, Oral presentation)
K. Yan, Y. Tang, et al., MULAN: Multitask Universal Lesion Analysis Network
for Joint Lesion Detection, Tagging, and Segmentation", International Conference
on Medical Image Computing and Computer Assisted Intervention (MICCAI),
Shenzhen, China, 2019
K. Yan, Y. Tang, et al., MULAN: Multitask Universal Lesion Analysis Network
for Joint Lesion Detection, Tagging, and Segmentation", RSNA, Chicago, USA, 2019
(Scientific Oral Paper)