Author: Xuehai He; Xingyi Yang; Shanghang Zhang; Jinyu Zhao; Yichen Zhang; Eric Xing; Pengtao Xie
Title: Sample-Efficient Deep Learning for COVID-19 Diagnosis Based on CT Scans Document date: 2020_4_17
ID: l3f469ht_49
Snippet: In addition to our COVID-19 CT dataset, we also include images from the Lung Nodule Analysis (LUNA) [54] database as a source of additional unlabeled CT data. It is originally designed for lung nodule detection and segmentation. From the total 888 CT scans, we randomly select 500 subjects and from each we extract two CT slices that contain annotated Algorithm 1 Algorithm of Self-Trans Input: batch sizes N S , N L , temperature Ï„ , LUNA dataset D.....
Document: In addition to our COVID-19 CT dataset, we also include images from the Lung Nodule Analysis (LUNA) [54] database as a source of additional unlabeled CT data. It is originally designed for lung nodule detection and segmentation. From the total 888 CT scans, we randomly select 500 subjects and from each we extract two CT slices that contain annotated Algorithm 1 Algorithm of Self-Trans Input: batch sizes N S , N L , temperature Ï„ , LUNA dataset D L , COVID-CT dataset D C , model f pretrained on Ima-geNet dataset, dictionary Q as a queue of K keys, augmentation operator a, a from the same family of augmentations Initialize encoder networks for query f q and keys f k :
Search related documents:
Co phrase search for related documents- batch size and CT slice: 1
- CT dataset and CT scan: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25
- CT dataset and CT slice: 1, 2, 3, 4, 5, 6, 7, 8, 9
- lung nodule detection and nodule detection: 1, 2, 3
Co phrase search for related documents, hyperlinks ordered by date