Automatic Measurement Method for Spatial Resolution of MRI Based on the ACR Phantom
10.3969/j.issn.1005-5185.2025.06.005
- VernacularTitle:基于ACR模体的MRI设备空间分辨力自动测量方法
- Author:
Yu ZHANG
1
;
Hongxia YIN
;
Yawen LIU
;
Pengling REN
;
Yanjun HU
;
Tianxin CHENG
;
Zhenghan YANG
;
Zhenchang WANG
;
Hui XU
Author Information
1. 首都医科大学附属北京友谊医院医学工程处,北京 100050
- Publication Type:Journal Article
- Keywords:
Magnetic resonance imaging;
Quality control;
Support vector machine;
Spatial resolution;
Automatic measurement
- From:
Chinese Journal of Medical Imaging
2025;33(6):595-600,606
- CountryChina
- Language:Chinese
-
Abstract:
Purpose To measure the spatial resolution in MRI quality control testing automatically based on the American College of Radiology(ACR)phantom using the support vector machine(SVM)method,and the feasibility,accuracy and measurement speed of this method are explored.Materials and Methods Quality control tests were performed using eight MRI devices at Beijing Friendship Hospital of Capital Medical University.A retrospective study was conducted on 71 MRI quality control test images collected based on ACR phantoms between 2017 and 2019.The images were preprocessed by binarization,extraction region of interest and so on.An SVM-based classification model was constructed for analyzing the spatial resolution of dot arrays in row and column directions.The dataset was randomly split into a training set and a test set.The generalization performance of the classification model in this study was evaluated through accuracy,precision,recall and F1 score on the test set.Comparing the results of spatial resolution measurements obtained by both manual and automatic method,we demonstrated the feasibility and accuracy of the method.Additionally,the time taken for the automatic spatial resolution measurement was recorded.Results In this study,the proposed method of automatically measuring the spatial resolution of ACR phantom test images using SVM was feasible,high accuracy and short time.In classification performance test,the accuracy of the spatial resolution of the row directional latices was 95%,the precision was 100%.The accuracy of the spatial resolution of the column directional latices was 97%,the precision was 100%.Among the test cases,the results of automatic measurements matched those of manual measurements in 13 out of 14 cases.On average,automatic spatial resolution measurement took 0.158 seconds per case.Conclusion This study achieves automatic measurement of spatial resolution in MRI quality control based on the ACR phantom using SVM method.The method demonstrates high accuracy and fast measurement speeds,holding significant implications for future rapid MRI quality control stability testing.
