CT datasets for Aortic dissection
Contrast-enhanced computed tomography (CE-CT) is the gold standard for diagnosing aortic dissection (AD). However, contrast agents can cause allergic reactions or renal failure in some patients. Moreover, AD diagnosis by radiologists using non- contrast-enhanced CT (NCE-CT) images has poor sensitivity. To address this issue, a novel deep learning methos was proposed for AD detection using NCE-CT volumes. It may have great potential to reduce the misdiagnosis of AD using NCE-CT in clinical practice.
All datasets were collected from two medical centers (i.e., Beijing Anzhen Hospital, Capital Medical University, Beijing, China; Fujian Provincial Hospital, Fuzhou, China) using three vendors (i.e., SIEMENS, TOSHIBA, and GE MEDICAL SYSTEMS) with kilovolt peak (KVP) of 100 ∼ 120. The datasets consisted of 207 subjects (i.e., 78 AD patients and 129 non-AD patients with other cardiovascular diseases). We sequentially performed Non-contrast enhanced CT and Contrast enhanced CT on each subject with the same scan conditions, including position, coverage, and parameters. To reduce motion artifacts and misregistration between NCE-CE and CE-CT, we acquired the datasets using electrocardiographic (ECG) triggering and breath-holding at the end of respiration.
In addition, we provide the true and false lumen masks for segmentation. True lumen was anotated with 1 and false lumen was anotated with 2.