핵의학 갑상샘 팬텀 영상에서 보간법이 초고분해능 ResNet 모델성능에 미치는 영향에 관한 연구
Study on Impact of Interpolation Methods on Performance of Super-Resolution ResNet Model in Nuclear Medicine Thyroid Phantom Imaging
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
The nuclear medicine imaging can improve the image quality through the application of various interpolation techniques. Additionally, deep learning algorithm, which perform feature extraction between input and label datasets, is widely utilized to improve image quality in nuclear medicine. Thus, the purpose of this study was to confirm the performance of deep learning algorithm according to applied various interpolation methods as input data using the thyroid phantom images. A total of 200 thyroid phantom images, each sized 256 × 256, were acquired at an activity of 37 M㏃ for 1 minute to generate the label images. Interpolation methods including nearest neighbor, bilinear, biquadratic, bicubic, biquartic, and biquintic were applied during both downsampling and upsampling processes. The super-resolution residual network (SRResNet) architecture was implemented with a learning rate of 0.0001 and 300 epochs, using an 8:1:1 ratio for train, validation, and test sets, respectively. The generated output images analyzed using coefficient of variation (COV) and contrast to noise ratio (CNR). Consequently, the SRResNet algorithm, which used low-resolution images generated with the bicubic interpolation method, showed the highest performance. This study demonstrates the importance of selecting appropriate interpolation methods for generating input images to improve the accuracy of the SRResNet algorithm in nuclear medicine thyroid imaging and even in other medical fields for diagnosis.
Abstract
The nuclear medicine imaging can improve the image quality through the application of various interpolation techniques. Additionally, deep learning algorithm, which perform feature extraction between input and label datasets, is widely utilized to improve image quality in nuclear medicine. Thus, the purpose of this study was to confirm the performance of deep learning algorithm according to applied various interpolation methods as input data using the thyroid phantom images. A total of 200 thyroid phantom images, each sized 256 × 256, were acquired at an activity of 37 M㏃ for 1 minute to generate the label images. Interpolation methods including nearest neighbor, bilinear, biquadratic, bicubic, biquartic, and biquintic were applied during both downsampling and upsampling processes. The super-resolution residual network (SRResNet) architecture was implemented with a learning rate of 0.0001 and 300 epochs, using an 8:1:1 ratio for train, validation, and test sets, respectively. The generated output images analyzed using coefficient of variation (COV) and contrast to noise ratio (CNR). Consequently, the SRResNet algorithm, which used low-resolution images generated with the bicubic interpolation method, showed the highest performance. This study demonstrates the importance of selecting appropriate interpolation methods for generating input images to improve the accuracy of the SRResNet algorithm in nuclear medicine thyroid imaging and even in other medical fields for diagnosis.
참고문헌 (17)
- Park Y, "Korean thyroid association guidelines on the management of differentiated thyroid cancers; overview and summary 2024", International Journal of Thyroidology, 17(1), pp.1~20 (2024)[DOI]
- Ahn J, "Application of artificial intelligence to evaluate thyroid nodules", Journal of Clinical Otolaryngology Head and Neck Surgery, 34(2), pp.17~22 (2023)[DOI]
- Kim TY, "Current status and future perspectives in differentiated thyroid cancer", Endocrinology And Metabolism, 29(3), pp.217~225 (2014)[DOI]
- Gnanasegaran G, "Molecular imaging agents for SPECT(and SPECT", CT), 2014, pp.41(Suppl 1), S26~S35
- Kim JS, "Study of simultaneous counting of thyroid uptake with quantitative analysis of thyroid scans", Journal of Radiological Science and Technology, 46(5), pp.401~408 (2023)[DOI]
- Pithani MB, "Bilinear and bicubic interpolations for image presentation of mechanical stress and temperature distribution", Power Engineering and Engineering Thermophysics, 1(1), pp.8~18 (2022)[DOI]
- Fry J, "A comparison of bicubic and biquintic interpolators suitable for real-time hardware implementation", 2009 IEEE Applied Imagery Pattern Recognition Workshop (AIPR 2009), pp.1~5 (2009)
- Kim K, "Improvement of signal and noise performance using single image super-resolution based on deep learning in single photon-emission computed tomography imaging system", Nuclear Engineering and Technology, 53(7), pp.2341~2347 (2021)[DOI]
- Jeong E, "Deep learning application of gamma camera quality control in nuclear medicine", Journal of Radiological Science and Technology, 43(6), pp.461~467 (2020)[DOI]
- Jeong EH, "Application of deep learning-based nuclear medicine lung study classification model", Journal of Radiological Science and Technology, 45(1), pp.41~47 (2022)[DOI]
- Kang H, "Evaluation of denoising performance of resnet deep learning model for ultrasound images corresponding to two frequency parameters", Bioengineering, 11(7), pp.723~ (2024)[DOI]
- Gilman A, "Interpolation models for image super-resolution", 4th IEEE International Symposium on Electronic Design, Test and Applications (delta 2008), pp.55~60 (2008)
- Lee Y, "Simulation and experimental studies of super resolution convolutional neural network algorithm in ultrasound image", Journal of The Korean Society of Radiology, 17(5), pp.693~699 (2023)[DOI]
- Suzuki K, "Overview of deep learning in medical imaging", Radiological Physics and Technology, 10, pp.257~273 (2017)[DOI]
- Becker D, "Procedure guideline for thyroid uptake measurement : 1. 0. Society of Nuclear Medicine", Journal of Nuclear Medicine, 37(7), pp.1266~1268 (1996)
- Ledig C, "Photo-realistic single image super-resolution using a generative adversarial network", 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp.105~114 (2017)
- Shi W, "Real-time single image and video super-resolution using an efficient sub-pixel convolutional neural network", 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp.1874~1883 (2016)