Author: Cheng Jin; Weixiang Chen; Yukun Cao; Zhanwei Xu; Xin Zhang; Lei Deng; Chuansheng Zheng; Jie Zhou; Heshui Shi; Jianjiang Feng
Title: Development and Evaluation of an AI System for COVID-19 Document date: 2020_3_23
ID: k1lg8c7q_20
Snippet: The AI system performs slightly better than the average of five radiologists. The ROC curve had AUC of 0. 9805, sensitivity of 0. 9470, and specificity of 0. 9139 on the cohort of reader study (Figure 2 b, d). In 46% (6/13) of cases, when the AI system missed, the radiologist also missed (Table 2 b), indicating that the diagnosis of these missed cases is challenging. Among the five readers, one reader performed better than the AI system, one read.....
Document: The AI system performs slightly better than the average of five radiologists. The ROC curve had AUC of 0. 9805, sensitivity of 0. 9470, and specificity of 0. 9139 on the cohort of reader study (Figure 2 b, d). In 46% (6/13) of cases, when the AI system missed, the radiologist also missed (Table 2 b), indicating that the diagnosis of these missed cases is challenging. Among the five readers, one reader performed better than the AI system, one reader performed worse, and the rest three have similar performance as the AI system at different operating points. Performance of the AI system in COVID-19 diagnosis compared to five readers is shown in Figure 2 b and Table 2 c.
Search related documents:
Co phrase search for related documents- AI system and reader study: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14
- AI system and reader study cohort: 1, 2, 3
- AI system and ROC curve: 1, 2, 3, 4, 5, 6
- AI system and similar performance: 1, 2, 3, 4
- AI system performance and diagnosis AI system performance: 1, 2
- AI system performance and similar performance: 1, 2, 3
- operating point and ROC curve: 1, 2, 3, 4, 5, 6, 7, 8, 9
- reader study and ROC curve: 1
- reader study cohort and ROC curve: 1
Co phrase search for related documents, hyperlinks ordered by date