{"id":"CU23246","slug":"noise-reducing-magnetic--CU23246","source":{"id":"CU23246","dataset":"techtransfer","title":"Noise-reducing magnetic resonance image reconstruction software","description_":"<p>This technology is an AI-based software for denoising and reconstructing MRI images.</p>\r\r<h2>Unmet Need: Fast, anatomically accurate, and de-noised reconstruction of medical MR images</h2>\r\r<p>Currently, acquiring magnetic resonance imaging (MRI) data requires long scan times, followed by multiple steps of image reconstruction and post-processing. Extended scan times create logistical burdens for patients and medical professionals. Additionally, slow reconstruction and limited image accuracy can reduce the speed and reliability of diagnosis from MRI images. Existing deep learning approaches typically operate on already reconstructed images and do not fully utilize information contained in raw magnetic resonance data.  </p>\r\r<h2>The Technology: A deep learning-based MR image reconstruction and denoising software</h2>\r\r<p>This technology is an MRI denoising and reconstruction software that uses an Artificial Fourier Transform Network (AFT-Net) to perform deep learning-based reconstruction of MRI data. This network learns the transformation between raw MRI signal data and reconstructed images, enabling simultaneous image reconstruction and noise reduction while improving robustness to variations in acquisition quality.  </p>\r\r<p>This software was validated on publicly available human MRI datasets and preclinical mouse MRI scans, demonstrating improved image reconstruction, denoising performance, and robustness across different imaging conditions.  </p>\r\r<h2>Applications:</h2>\r\r<ul>\r<li>Accelerated MRI imaging with reduced sampling</li>\r<li>Clinical MRI reconstruction for diagnostic imaging</li>\r<li>Preclinical MR image reconstruction for enhanced MRI image quality </li>\r<li>Preclinical MR image denoising and artifact reduction</li>\r<li>Preclinical MRI image reconstruction for animal imaging studies</li>\r<li>Magnetic resonance spectroscopy reconstruction for metabolic analysis in clinical imaging </li>\r</ul>\r\r<h2>Advantages:</h2>\r\r<ul>\r<li>Reconstruction directly from raw MRI data (k-space / FID)</li>\r<li>Ability to reconstruct from under-sampled and noisy data  </li>\r<li>Robust to noise and artifacts </li>\r<li>Can be incorporated into existing deep learning networks</li>\r<li>Trained on various datasets to enable robust to variable acquisition and experimental parameters</li>\r</ul>\r\r<h2>Lead Inventor:</h2>\r\r<p><a href=\"https://mr.research.columbia.edu/content/jia-guo\">Jia Guo, Ph.D.</a></p>\r\r<h2>Patent Information:</h2>\r\r<p>Patent Pending (<a href=\"https://patents.google.com/patent/US20260051099A1/en?oq=US2026%2f0051099\">US20260051099</a>)</p>\r\r<h2>Related Publications:</h2>\r\r<ul>\r<li><a href=\"https://pubmed.ncbi.nlm.nih.gov/40328027/\">Yang Y, Zhang Y, Li Z, Tian JS, Dagommer M, Guo J. “Deep learning-based MRI reconstruction with Artificial Fourier Transform Network (AFTNet).” Computers in Biology and Medicine. 2025 Jun 1; 192 110224.</a> </li>\r</ul>\r\r<h2>Tech Ventures Reference:</h2>\r\r<ul>\r<li><p>IR CU23246</p></li>\r<li><p>Licensing Contact: <a href=\"mailto:jk2108@ctv.columbia.edu\">Jerry Kokosha</a> </p></li>\r</ul>\r","tags":["Deep learning","Fourier transform","Frequency domain","Magnetic resonance imaging","Medical imaging","Noise reduction","Nuclear magnetic resonance","Nuclear magnetic resonance spectroscopy","Plastic surgery"],"file_number":"CU23246","collections":[{"key":433,"name":"Digital Health (Software, AI, bioinformatics, etc)"}],"meta_description":"AI-driven MRI reconstruction and denoising using AFT-Net, enabling rapid, noise-robust imaging from raw data.","apriori_judge_output":"{\"scores\":{\"novelty\":4.0,\"potential_impact\":4.0,\"readiness\":3.0,\"scalability\":3.0,\"timeliness\":3.0},\"weighted_score\":3.65,\"risks\":[\"Patent-pending status may face patentability/defensibility questions\",\"Overlap with existing DI/ML MRI reconstruction methods could affect differentiation\",\"Regulatory and clinical adoption uncertainty for new AI-based reconstruction in diverse MRI platforms\",\"Need for extensive multi-site validation to confirm generalizability across scanners and protocols\"],\"one_sentence_take\":\"Strong novelty and potential impact, but readiness and scalability require more validation and regulatory/competitive differentiation.\"}","inventors":["Jia Guo","Yanting Yang"],"manager":"Jerry Kokoshka","depts":["Biomedical Engineering","Psychiatry"],"divs":["Columbia University Medical Center (CUMC)","Fu Foundation School of Engineering and Applied Science (SEAS)"],"date_released":"2023-07-01"},"highlight":{},"matched_queries":null,"score":0.0}