no code implementations • 19 Sep 2023 • Rucha Deshpande, Muzaffer Özbey, Hua Li, Mark A. Anastasio, Frank J. Brooks
However, there remains an important need to understand the extent to which DDPMs can reliably learn medical imaging domain-relevant information, which is referred to as `spatial context' in this work.
1 code implementation • 6 Jan 2023 • Salman UH Dar, Şaban Öztürk, Muzaffer Özbey, Tolga Çukur
To alleviate error propagation, PSFNet combines its SS and SG priors via a novel parallel-stream architecture with learnable fusion parameters.
1 code implementation • 17 Jul 2022 • Muzaffer Özbey, Onat Dalmaz, Salman UH Dar, Hasan A Bedel, Şaban Özturk, Alper Güngör, Tolga Çukur
Extensive assessments are reported on the utility of SynDiff against competing GAN and diffusion models in multi-contrast MRI and MRI-CT translation.
Ranked #7 on Image-to-Image Translation on IXI
1 code implementation • 13 Jul 2022 • Onat Dalmaz, Usama Mirza, Gökberk Elmas, Muzaffer Özbey, Salman UH Dar, Emir Ceyani, Salman Avestimehr, Tolga Çukur
As such, pFLSynth enables training of a unified synthesis model that can reliably generalize across multiple sites and translation tasks.
1 code implementation • 12 Jul 2022 • Alper Güngör, Salman UH Dar, Şaban Öztürk, Yilmaz Korkmaz, Gokberk Elmas, Muzaffer Özbey, Tolga Çukur
A two-phase reconstruction is executed following training: a rapid-diffusion phase that produces an initial reconstruction with the trained prior, and an adaptation phase that further refines the result by updating the prior to minimize data-consistency loss.
1 code implementation • 8 Feb 2022 • Gokberk Elmas, Salman UH Dar, Yilmaz Korkmaz, Emir Ceyani, Burak Susam, Muzaffer Özbey, Salman Avestimehr, Tolga Çukur
Specificity in the prior is preserved via a mapper subnetwork that produces site-specific latents.
1 code implementation • 15 May 2021 • Yilmaz Korkmaz, Salman UH Dar, Mahmut Yurt, Muzaffer Özbey, Tolga Çukur
Supervised reconstruction models are characteristically trained on matched pairs of undersampled and fully-sampled data to capture an MRI prior, along with supervision regarding the imaging operator to enforce data consistency.
no code implementations • 18 Dec 2020 • Muzaffer Özbey, Mahmut Yurt, Salman Ul Hassan Dar, Tolga Çukur
Mainstream deep models for three-dimensional MRI synthesis are either cross-sectional or volumetric depending on the input.
no code implementations • 29 Nov 2020 • Mahmut Yurt, Salman Ul Hassan Dar, Muzaffer Özbey, Berk Tınaz, Kader Karlı Oğuz, Tolga Çukur
Here, we propose a novel semi-supervised deep generative model that instead learns to recover high-quality target images directly from accelerated acquisitions of source and target contrasts.
no code implementations • 27 Nov 2020 • Mahmut Yurt, Muzaffer Özbey, Salman Ul Hassan Dar, Berk Tınaz, Kader Karlı Oğuz, Tolga Çukur
Comprehensive demonstrations on mainstream MRI reconstruction and synthesis tasks show that ProvoGAN yields superior performance to state-of-the-art volumetric and cross-sectional models.
no code implementations • 7 Oct 2017 • Salman Ul Hassan Dar, Muzaffer Özbey, Ahmet Burak Çatlı, Tolga Çukur
Methods: Neural networks were trained on thousands of samples from public datasets of either natural images or brain MR images.