Niigata University of Health and Welfare: Reducing the Burden of 'Rescans'! AI Automatically Corrects MRI 'Blur'
Lecturer Norikiyo Yoshida of Niigata University of Health and Welfare conducted research to verify an AI (deep learning) based correction method for motion artifacts in MRI examinations. The results showed the potential for more accurate evaluation of hippocampal volume, crucial for Alzheimer's disease diagnosis, by correcting blurred brain MRI images, leading to reduced rescans.
📋 Article Processing Timeline
- 📰 Published: April 1, 2026 at 20:00
- 🔍 Collected: April 1, 2026 at 16:47
- 🤖 AI Analyzed: April 21, 2026 at 13:59 (477h 11m after Collected)
Norikiyo Yoshida, a lecturer in the Department of Radiological Technology at Niigata University of Health and Welfare, part of the NSG Group, conducted research to verify a correction method using AI (deep learning) for image blur caused by patient movement during MRI examinations. The results indicated the potential for more accurate evaluation of hippocampal volume, which is crucial for Alzheimer's disease diagnosis, by correcting blurred brain MRI images.
This research outcome was published on March 22, 2026, in "Scientific Reports," an international journal covering all fields of natural and health sciences.
### About the Research
**[Research Overview]**
This study verified the effectiveness of an AI (deep learning) based correction method for image blur caused by patient movement during MRI examinations. Brain MRI images of 24 healthy adults were compared across original images, images with motion, and AI-corrected images. The evaluation focused not only on visual appearance but also on the accuracy of hippocampal volume measurement.
The results showed that even images degraded by motion could have their quality improved through AI correction, indicating the potential for more accurate evaluation of hippocampal volume, which is important for Alzheimer's disease diagnosis.
Furthermore, this method suggested the possibility of reducing the need for rescans, leading to shorter examination times and reduced patient burden.
This research applies AI-based image correction technology to the quantitative evaluation of brain MRI, and its future application in diagnostic support for the dementia field is highly anticipated.
**[Researcher's Comment]**
◆ Norikiyo Yoshida, Lecturer, Department of Radiological Technology
In this research, we investigated the potential of image correction using deep learning to address the challenge that slight patient movement during MRI examinations affects image quality and diagnostic accuracy. The results showed that even brain MRI images affected by motion could be corrected by AI, enabling more accurate evaluation of hippocampal volume, which is crucial for Alzheimer's disease diagnosis. Especially for elderly patients or those with difficulty controlling body movement, rescans are often necessary. If this method is put into practical use, we believe it can reduce the need for repeat examinations, alleviate patient burden, and contribute to the efficiency of medical sites.
**[Original Paper Information]**
Yoshida, N., Kageyama, H., Akai, H., Sasaki, K., Sakurai, N., Koori, N., Yamamoto, S., & Kodama, N.
Deep learning approach to super-resolution correction of brain MRI motion artifacts for accurate hippocampal volumetry.
Scientific Reports (2026).
DOI: 10.1038/s41598-026-44834-5
**[Researcher Information]**
Norikiyo Yoshida, Lecturer
Department of Radiological Technology, Faculty of Medical Technology, Niigata University of Health and Welfare
**[Contact]**
Public Relations Division, Admissions and Public Relations Department, Niigata University of Health and Welfare
Location: 1398 Shimami-cho, Kita-ku, Niigata City, Niigata Prefecture
TEL: 025-257-4459
This research outcome was published on March 22, 2026, in "Scientific Reports," an international journal covering all fields of natural and health sciences.
### About the Research
**[Research Overview]**
This study verified the effectiveness of an AI (deep learning) based correction method for image blur caused by patient movement during MRI examinations. Brain MRI images of 24 healthy adults were compared across original images, images with motion, and AI-corrected images. The evaluation focused not only on visual appearance but also on the accuracy of hippocampal volume measurement.
The results showed that even images degraded by motion could have their quality improved through AI correction, indicating the potential for more accurate evaluation of hippocampal volume, which is important for Alzheimer's disease diagnosis.
Furthermore, this method suggested the possibility of reducing the need for rescans, leading to shorter examination times and reduced patient burden.
This research applies AI-based image correction technology to the quantitative evaluation of brain MRI, and its future application in diagnostic support for the dementia field is highly anticipated.
**[Researcher's Comment]**
◆ Norikiyo Yoshida, Lecturer, Department of Radiological Technology
In this research, we investigated the potential of image correction using deep learning to address the challenge that slight patient movement during MRI examinations affects image quality and diagnostic accuracy. The results showed that even brain MRI images affected by motion could be corrected by AI, enabling more accurate evaluation of hippocampal volume, which is crucial for Alzheimer's disease diagnosis. Especially for elderly patients or those with difficulty controlling body movement, rescans are often necessary. If this method is put into practical use, we believe it can reduce the need for repeat examinations, alleviate patient burden, and contribute to the efficiency of medical sites.
**[Original Paper Information]**
Yoshida, N., Kageyama, H., Akai, H., Sasaki, K., Sakurai, N., Koori, N., Yamamoto, S., & Kodama, N.
Deep learning approach to super-resolution correction of brain MRI motion artifacts for accurate hippocampal volumetry.
Scientific Reports (2026).
DOI: 10.1038/s41598-026-44834-5
**[Researcher Information]**
Norikiyo Yoshida, Lecturer
Department of Radiological Technology, Faculty of Medical Technology, Niigata University of Health and Welfare
**[Contact]**
Public Relations Division, Admissions and Public Relations Department, Niigata University of Health and Welfare
Location: 1398 Shimami-cho, Kita-ku, Niigata City, Niigata Prefecture
TEL: 025-257-4459