To Analyze the Value of Artificial Intelligence in Diagnosing Rib Fracture with Multi-Slice Spiral CT
Objective:To evaluate the effectiveness of artificial intelligence(AI)rib fracture detection system in diagnosing rib fracture with multi-slice spiral CT(MSCT).Methods:From May 2021 to May 2022,50 patients with rib fracture and 50 patients with rib fracture were by 64-slice spiral CT scan,then,using Al image analysis system(AI group),artificial group and artificial+AI group to compare and analyze the characteristics of rib,get the number of rib fracture,and then count the detection time of rib fracture in three groups,false positive rate,sensitivity and accuracy.Results:The average time for AI to detect rib fracture was(63±18)s,the false positive rate was 4.56％,the sensitivity was 87.75％,the accuracy was 95.13％,the average time for AI to detect rib fracture was(203±68)s,the false positive rate was 1.30％,the sensitivity was 84.69％,the average time of artificial rib fracture detection by AI was(136±28)s,the false positive rate was 0.96％,the sensitivity was 92.85％,the accuracy was 98.79％.Conclusion:AI technique has a high sensitivity and accuracy in diagnosis of RIB fracture.Although there is a certain false positive rate,it can be used as an assistant in manual operation and reduce the time of reading films,it has remarkable effect in improving the quality of diagnosis and working efficiency.
multi-slice spiral CTrib fractureartificial intelligencediagnostic value
万方数据
维普
