Multi-contrast high-field quality image synthesis for portable low-field MRI using generative adversarial networks and paired data.
العنوان: | Multi-contrast high-field quality image synthesis for portable low-field MRI using generative adversarial networks and paired data. |
---|---|
المؤلفون: | Lucas A; Perelman School of Medicine, University of Pennsylvania.; Center for Neuroengineering and Therapeutics, Departments of Bioengineering and Neurology, University of Pennsylvania., Campbell Arnold T; Center for Neuroengineering and Therapeutics, Departments of Bioengineering and Neurology, University of Pennsylvania., Okar SV; National Institute of Neurological Disorders and Stroke, National Institutes of Health., Vadali C; Center for Neuroengineering and Therapeutics, Departments of Bioengineering and Neurology, University of Pennsylvania.; Department of Radiology, University of Pennsylvania., Kawatra KD; National Institute of Neurological Disorders and Stroke, National Institutes of Health., Ren Z; Penn Statistics in Imaging and Visualization Center, Department of Biostatistics, Epidemiology, and Informatics, University of Pennsylvania., Cao Q; Penn Statistics in Imaging and Visualization Center, Department of Biostatistics, Epidemiology, and Informatics, University of Pennsylvania., Shinohara RT; Penn Statistics in Imaging and Visualization Center, Department of Biostatistics, Epidemiology, and Informatics, University of Pennsylvania.; Center for Biomedical Image Computing and Analytics (CBICA), University of Pennsylvania., Schindler MK; Perelman School of Medicine, University of Pennsylvania.; Department of Neurology, University of Pennsylvania., Davis KA; Perelman School of Medicine, University of Pennsylvania.; Center for Neuroengineering and Therapeutics, Departments of Bioengineering and Neurology, University of Pennsylvania.; Department of Neurology, University of Pennsylvania., Litt B; Perelman School of Medicine, University of Pennsylvania.; Center for Neuroengineering and Therapeutics, Departments of Bioengineering and Neurology, University of Pennsylvania.; Department of Neurology, University of Pennsylvania., Reich DS; National Institute of Neurological Disorders and Stroke, National Institutes of Health., Stein JM; Perelman School of Medicine, University of Pennsylvania.; Center for Neuroengineering and Therapeutics, Departments of Bioengineering and Neurology, University of Pennsylvania.; Department of Radiology, University of Pennsylvania. |
المصدر: | MedRxiv : the preprint server for health sciences [medRxiv] 2023 Dec 29. Date of Electronic Publication: 2023 Dec 29. |
نوع المنشور: | Preprint |
اللغة: | English |
بيانات الدورية: | Country of Publication: United States NLM ID: 101767986 Publication Model: Electronic Cited Medium: Internet NLM ISO Abbreviation: medRxiv Subsets: PubMed not MEDLINE |
مستخلص: | Introduction: Portable low-field strength (64mT) MRI scanners promise to increase access to neuroimaging for clinical and research purposes, however these devices produce lower quality images compared to high-field scanners. In this study, we developed and evaluated a deep learning architecture to generate high-field quality brain images from low-field inputs using a paired dataset of multiple sclerosis (MS) patients scanned at 64mT and 3T. Methods: A total of 49 MS patients were scanned on portable 64mT and standard 3T scanners at Penn (n=25) or the National Institutes of Health (NIH, n=24) with T1-weighted, T2-weighted and FLAIR acquisitions. Using this paired data, we developed a generative adversarial network (GAN) architecture for low- to high-field image translation (LowGAN). We then evaluated synthesized images with respect to image quality, brain morphometry, and white matter lesions. Results: Synthetic high-field images demonstrated visually superior quality compared to low-field inputs and significantly higher normalized cross-correlation (NCC) to actual high-field images for T1 (p=0.001) and FLAIR (p<0.001) contrasts. LowGAN generally outperformed the current state-of-the-art for low-field volumetrics. For example, thalamic, lateral ventricle, and total cortical volumes in LowGAN outputs did not differ significantly from 3T measurements. Synthetic outputs preserved MS lesions and captured a known inverse relationship between total lesion volume and thalamic volume. Conclusions: LowGAN generates synthetic high-field images with comparable visual and quantitative quality to actual high-field scans. Enhancing portable MRI image quality could add value and boost clinician confidence, enabling wider adoption of this technology. |
معلومات مُعتمدة: | DP1 NS122038 United States NS NINDS NIH HHS; R01 NS125137 United States NS NINDS NIH HHS; T32 EB009384 United States EB NIBIB NIH HHS |
تواريخ الأحداث: | Date Created: 20240118 Latest Revision: 20240210 |
رمز التحديث: | 20240210 |
مُعرف محوري في PubMed: | PMC10793526 |
DOI: | 10.1101/2023.12.28.23300409 |
PMID: | 38234785 |
قاعدة البيانات: | MEDLINE |
DOI: | 10.1101/2023.12.28.23300409 |
---|