The value of artificial intelligence model based on CT imaging in the differential diagnosis of kidney stones in civil aviation pilots
10.3760/cma.j.cn113854-20231224-00136
- VernacularTitle:基于CT成像的人工智能模型在民航飞行员肾结石鉴别诊断中的价值
- Author:
Bin LIU
1
;
Jianfei LI
;
Xiaokun ZHANG
;
Ying ZHANG
;
Haifeng ZHU
Author Information
1. 民航总医院放射科，北京 100123
- Publication Type:Journal Article
- Keywords:
Kidney calculi;
Renal papillary calcification;
Artificial intelligence;
Civil pilots
- From:
Chinese Journal of Aerospace Medicine
2024;35(4):286-289
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To accurately identify kidney stones and renal papillary calcification in civil aviation pilots by using an artificial intelligence model based on CT imaging.Methods:The imaging data of 135 civil pilots with kidney stones or renal papillary calcification confirmed by flexible ureteroscope who underwent urinary CT examination in Civil Aviation General Hospital from August 2015 to May 2021 were retrospectively analyzed. Preoperative CT images of all patients were analyzed blindly by 2 radiologists and recall and accuracy rates were calculated. Pilots with kidney stones and renal papillary calcification were randomly assigned as the training set and test set in a ratio of 4∶1 respectively. An artificial intelligence model was constructed based on preoperative CT plain scan images of training set. The test set was used to detect the model and calculate the recall rate and accuracy rate of the final model in the diagnosis of the 2 types of lesions.Results:Data from 86 pilots with kidney stones and 49 pilots with renal papillary calcification were used to build the model and the training iteration of model establishment was stopped at the 50th time. The final results of the model interpretation test set showed that the recall rate and accuracy rate diagnosing kidney stones was 87.58% and 97.17%, respectively, and the recall rate and accuracy rate diagnosing renal sinus calcification was 96.07% and 92.83%, respectively, and the accuracy rate of the model in diagnosing kidney stones and renal sinus calcification was 96.57%.Conclusions:The artificial intelligence model based on CT imaging has a certain clinical value for the rapid and accurate determination of kidney stones and renal papillary calcification by air crew examiners in civil aviation.
