{"title":"使用基于深度学习的模型，从3D头部MRI定位器（AutoAlign head）图像生成类似mprage的高分辨率图像。","authors":"Hiroshi Tagawa, Yasutaka Fushimi, Koji Fujimoto, Satoshi Nakajima, Sachi Okuchi, Akihiko Sakata, Sayo Otani, Krishna Pandu Wicaksono, Yang Wang, Satoshi Ikeda, Shuichi Ito, Masaki Umehana, Akihiro Shimotake, Akira Kuzuya, Yuji Nakamoto","doi":"10.1007/s11604-024-01728-8","DOIUrl":null,"url":null,"abstract":"<p><strong>Purpose: </strong>Magnetization prepared rapid gradient echo (MPRAGE) is a useful three-dimensional (3D) T1-weighted sequence, but is not a priority in routine brain examinations. We hypothesized that converting 3D MRI localizer (AutoAlign Head) images to MPRAGE-like images with deep learning (DL) would be beneficial for diagnosing and researching dementia and neurodegenerative diseases. We aimed to establish and evaluate a DL-based model for generating MPRAGE-like images from MRI localizers.</p><p><strong>Materials and methods: </strong>Brain MRI examinations including MPRAGE taken at a single institution for investigation of mild cognitive impairment, dementia and epilepsy between January 2020 and December 2022 were included retrospectively. Images taken in 2020 or 2021 were assigned to training and validation datasets, and images from 2022 were used for the test dataset. Using the training and validation set, we determined one model using visual evaluation by radiologists with reference to image quality metrics of peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), and Learned Perceptual Image Patch Similarity (LPIPS). The test dataset was evaluated by visual assessment and quality metrics. Voxel-based morphometric analysis was also performed, and we evaluated Dice score and volume differences between generated and original images of major structures were calculated as absolute symmetrized percent change.</p><p><strong>Results: </strong>Training, validation, and test datasets comprised 340 patients (mean age, 56.1 ± 24.4 years; 195 women), 36 patients (67.3 ± 18.3 years, 20 women), and 193 patients (59.5 ± 24.4 years; 111 women), respectively. The test dataset showed: PSNR, 35.4 ± 4.91; SSIM, 0.871 ± 0.058; and LPIPS 0.045 ± 0.017. No overfitting was observed. Dice scores for the segmentation of main structures ranged from 0.788 (left amygdala) to 0.926 (left ventricle). Quadratic weighted Cohen kappa values of visual score for medial temporal lobe between original and generated images were 0.80-0.88.</p><p><strong>Conclusion: </strong>Images generated using our DL-based model can be used for post-processing and visual evaluation of medial temporal lobe atrophy.</p>","PeriodicalId":14691,"journal":{"name":"Japanese Journal of Radiology","volume":" ","pages":""},"PeriodicalIF":2.1000,"publicationDate":"2025-01-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Generation of high-resolution MPRAGE-like images from 3D head MRI localizer (AutoAlign Head) images using a deep learning-based model.\",\"authors\":\"Hiroshi Tagawa, Yasutaka Fushimi, Koji Fujimoto, Satoshi Nakajima, Sachi Okuchi, Akihiko Sakata, Sayo Otani, Krishna Pandu Wicaksono, Yang Wang, Satoshi Ikeda, Shuichi Ito, Masaki Umehana, Akihiro Shimotake, Akira Kuzuya, Yuji Nakamoto\",\"doi\":\"10.1007/s11604-024-01728-8\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><strong>Purpose: </strong>Magnetization prepared rapid gradient echo (MPRAGE) is a useful three-dimensional (3D) T1-weighted sequence, but is not a priority in routine brain examinations. We hypothesized that converting 3D MRI localizer (AutoAlign Head) images to MPRAGE-like images with deep learning (DL) would be beneficial for diagnosing and researching dementia and neurodegenerative diseases. We aimed to establish and evaluate a DL-based model for generating MPRAGE-like images from MRI localizers.</p><p><strong>Materials and methods: </strong>Brain MRI examinations including MPRAGE taken at a single institution for investigation of mild cognitive impairment, dementia and epilepsy between January 2020 and December 2022 were included retrospectively. Images taken in 2020 or 2021 were assigned to training and validation datasets, and images from 2022 were used for the test dataset. Using the training and validation set, we determined one model using visual evaluation by radiologists with reference to image quality metrics of peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), and Learned Perceptual Image Patch Similarity (LPIPS). The test dataset was evaluated by visual assessment and quality metrics. Voxel-based morphometric analysis was also performed, and we evaluated Dice score and volume differences between generated and original images of major structures were calculated as absolute symmetrized percent change.</p><p><strong>Results: </strong>Training, validation, and test datasets comprised 340 patients (mean age, 56.1 ± 24.4 years; 195 women), 36 patients (67.3 ± 18.3 years, 20 women), and 193 patients (59.5 ± 24.4 years; 111 women), respectively. The test dataset showed: PSNR, 35.4 ± 4.91; SSIM, 0.871 ± 0.058; and LPIPS 0.045 ± 0.017. No overfitting was observed. Dice scores for the segmentation of main structures ranged from 0.788 (left amygdala) to 0.926 (left ventricle). Quadratic weighted Cohen kappa values of visual score for medial temporal lobe between original and generated images were 0.80-0.88.</p><p><strong>Conclusion: </strong>Images generated using our DL-based model can be used for post-processing and visual evaluation of medial temporal lobe atrophy.</p>\",\"PeriodicalId\":14691,\"journal\":{\"name\":\"Japanese Journal of Radiology\",\"volume\":\" \",\"pages\":\"\"},\"PeriodicalIF\":2.1000,\"publicationDate\":\"2025-01-11\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Japanese Journal of Radiology\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1007/s11604-024-01728-8\",\"RegionNum\":4,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Japanese Journal of Radiology","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1007/s11604-024-01728-8","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Generation of high-resolution MPRAGE-like images from 3D head MRI localizer (AutoAlign Head) images using a deep learning-based model.
Purpose: Magnetization prepared rapid gradient echo (MPRAGE) is a useful three-dimensional (3D) T1-weighted sequence, but is not a priority in routine brain examinations. We hypothesized that converting 3D MRI localizer (AutoAlign Head) images to MPRAGE-like images with deep learning (DL) would be beneficial for diagnosing and researching dementia and neurodegenerative diseases. We aimed to establish and evaluate a DL-based model for generating MPRAGE-like images from MRI localizers.
Materials and methods: Brain MRI examinations including MPRAGE taken at a single institution for investigation of mild cognitive impairment, dementia and epilepsy between January 2020 and December 2022 were included retrospectively. Images taken in 2020 or 2021 were assigned to training and validation datasets, and images from 2022 were used for the test dataset. Using the training and validation set, we determined one model using visual evaluation by radiologists with reference to image quality metrics of peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), and Learned Perceptual Image Patch Similarity (LPIPS). The test dataset was evaluated by visual assessment and quality metrics. Voxel-based morphometric analysis was also performed, and we evaluated Dice score and volume differences between generated and original images of major structures were calculated as absolute symmetrized percent change.
Results: Training, validation, and test datasets comprised 340 patients (mean age, 56.1 ± 24.4 years; 195 women), 36 patients (67.3 ± 18.3 years, 20 women), and 193 patients (59.5 ± 24.4 years; 111 women), respectively. The test dataset showed: PSNR, 35.4 ± 4.91; SSIM, 0.871 ± 0.058; and LPIPS 0.045 ± 0.017. No overfitting was observed. Dice scores for the segmentation of main structures ranged from 0.788 (left amygdala) to 0.926 (left ventricle). Quadratic weighted Cohen kappa values of visual score for medial temporal lobe between original and generated images were 0.80-0.88.
Conclusion: Images generated using our DL-based model can be used for post-processing and visual evaluation of medial temporal lobe atrophy.
期刊介绍：
Japanese Journal of Radiology is a peer-reviewed journal, officially published by the Japan Radiological Society. The main purpose of the journal is to provide a forum for the publication of papers documenting recent advances and new developments in the field of radiology in medicine and biology. The scope of Japanese Journal of Radiology encompasses but is not restricted to diagnostic radiology, interventional radiology, radiation oncology, nuclear medicine, radiation physics, and radiation biology. Additionally, the journal covers technical and industrial innovations. The journal welcomes original articles, technical notes, review articles, pictorial essays and letters to the editor. The journal also provides announcements from the boards and the committees of the society. Membership in the Japan Radiological Society is not a prerequisite for submission. Contributions are welcomed from all parts of the world.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
