Zobrazeno 1 - 10
of 128
pro vyhledávání: '"Dar, Salman"'
Contemporary developments in generative AI are rapidly transforming the field of medical AI. These developments have been predominantly driven by the availability of large datasets and high computing power, which have facilitated a significant increa
Externí odkaz:
http://arxiv.org/abs/2407.14892
Autor:
Dar, Salman Ul Hassan, Seyfarth, Marvin, Kahmann, Jannik, Ayx, Isabelle, Papavassiliu, Theano, Schoenberg, Stefan O., Frey, Norbert, Baeßler, Bettina, Foersch, Sebastian, Truhn, Daniel, Kather, Jakob Nikolas, Engelhardt, Sandy
AI models present a wide range of applications in the field of medicine. However, achieving optimal performance requires access to extensive healthcare data, which is often not readily available. Furthermore, the imperative to preserve patient privac
Externí odkaz:
http://arxiv.org/abs/2402.01054
Autor:
Mirza, Muhammad U., Dalmaz, Onat, Bedel, Hasan A., Elmas, Gokberk, Korkmaz, Yilmaz, Gungor, Alper, Dar, Salman UH, Çukur, Tolga
Deep generative models have gained recent traction in accelerated MRI reconstruction. Diffusion priors are particularly promising given their representational fidelity. Instead of the target transformation from undersampled to fully-sampled data requ
Externí odkaz:
http://arxiv.org/abs/2308.01096
Autor:
Dar, Salman Ul Hassan, Ghanaat, Arman, Kahmann, Jannik, Ayx, Isabelle, Papavassiliu, Theano, Schoenberg, Stefan O., Engelhardt, Sandy
Generative latent diffusion models have been established as state-of-the-art in data generation. One promising application is generation of realistic synthetic medical imaging data for open data sharing without compromising patient privacy. Despite t
Externí odkaz:
http://arxiv.org/abs/2307.01148
Autor:
Turk, Sevcan, Demirkaya, Ahmet, Turali, M Yigit, Hepdurgun, Cenk, Dar, Salman UH, Karabulut, Ahmet K, Azizova, Aynur, Orman, Mehmet, Tamsel, Ipek, Aydingoz, Ustun, Argin, Mehmet, Cukur, Tolga
Purpose: To develop a deep learning model that predicts active inflammation from sacroiliac joint radiographs and to compare the success with radiologists. Materials and Methods: A total of 1,537 (augmented 1752) grade 0 SIJs of 768 patients were ret
Externí odkaz:
http://arxiv.org/abs/2301.10769
Magnetic resonance imaging (MRI) is an essential diagnostic tool that suffers from prolonged scan times. Reconstruction methods can alleviate this limitation by recovering clinically usable images from accelerated acquisitions. In particular, learnin
Externí odkaz:
http://arxiv.org/abs/2301.02613
Autor:
Özbey, Muzaffer, Dalmaz, Onat, Dar, Salman UH, Bedel, Hasan A, Özturk, Şaban, Güngör, Alper, Çukur, Tolga
Imputation of missing images via source-to-target modality translation can improve diversity in medical imaging protocols. A pervasive approach for synthesizing target images involves one-shot mapping through generative adversarial networks (GAN). Ye
Externí odkaz:
http://arxiv.org/abs/2207.08208
Autor:
Dalmaz, Onat, Mirza, Usama, Elmas, Gökberk, Özbey, Muzaffer, Dar, Salman UH, Ceyani, Emir, Avestimehr, Salman, Çukur, Tolga
Multi-institutional collaborations are key for learning generalizable MRI synthesis models that translate source- onto target-contrast images. To facilitate collaboration, federated learning (FL) adopts decentralized training and mitigates privacy co
Externí odkaz:
http://arxiv.org/abs/2207.06509
Autor:
Güngör, Alper, Dar, Salman UH, Öztürk, Şaban, Korkmaz, Yilmaz, Elmas, Gokberk, Özbey, Muzaffer, Çukur, Tolga
Publikováno v:
A. G\"ung\"or, S. U. Dar, S. \"Ozt\"urk, Y. Korkmaz, G. Elmas, M. \"Ozbey, and T. \c{C}ukur, "Adaptive diffusion priors for accelerated MRI reconstruction," Medical Image Analysis, vol. 88, p. 102872, 2023
Deep MRI reconstruction is commonly performed with conditional models that de-alias undersampled acquisitions to recover images consistent with fully-sampled data. Since conditional models are trained with knowledge of the imaging operator, they can
Externí odkaz:
http://arxiv.org/abs/2207.05876
Deep-learning models have enabled performance leaps in analysis of high-dimensional functional MRI (fMRI) data. Yet, many previous methods are suboptimally sensitive for contextual representations across diverse time scales. Here, we present BolT, a
Externí odkaz:
http://arxiv.org/abs/2205.11578