http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
K-공간 기반 디노이징 딥 러닝 기법을 이용한 요추 자기공명영상 검사의 유용성에 관한 연구
이선경,조용근 대한자기공명기술학회 2022 대한자기공명기술학회지 Vol.32 No.1
Magnetic resonance imaging (MRI) is one of the main diagnostic tools used to assess spinal pathologies as it helps diagnose the source of lower back pain by producing high contrast, high resolution, multiplanar, detailed images of the lower spine and surrounding tissues. However, MRI can cause motion artifacts in patients who are unable to cooperate due to long scan times; thus, various techniques have been developed to shorten MRI examination times. Deep learning (DL) techniques that significantly reduce scan times while maintaining high image quality are attracting the attention of researchers. In this study, to determine the usefulness of DL techniques for lumbar spine MRIs, sagittal and axial T2 emphasis images were obtained before and after a DL technique was applied, and signal-to-noise ratio, contrast-to-noise ratio, scan time, overall image quality, and consistency of lesion diagnosis were compared and analyzed. It was found that the quality of the image improved and the artifacts caused by movement or breathing were reduced owing to the shorter MRI examination time. Hence, it is expected that MRI examination using DL techniques will provide images with higher diagnostic value while increasing patient satisfaction, making it a useful method in clinical practice.