Feliks Kogan

Bio: Feliks Kogan is an academic researcher from Stanford University. The author has contributed to research in topics: Osteoarthritis & Medicine. The author has an hindex of 21, co-authored 55 publications receiving 2025 citations. Previous affiliations of Feliks Kogan include University of Pennsylvania.

Magnetic resonance imaging of glutamate

Abstract: Glutamate (Glu) exhibits a pH and concentration dependent chemical exchange saturation transfer effect (CEST) between its -amine group and bulk water, here termed GluCEST. GluCEST asymmetry is observed at ~3 parts per million downfield from bulk water. Following middle cerebral artery occlusion in the rat brain, an approximately 100% elevation of GluCEST in the ipsilateral side compared to the contralateral side was observed, and is predominantly due to pH changes. In a rat brain tumor model with blood brain barrier disruption, intravenous Glu injection resulted in a clear elevation of GluCEST and a comparable increase in the proton magnetic resonance spectroscopy signal of Glu. GluCEST maps from healthy human brain at 7T were also obtained. These results demonstrate the feasibility and potential of GluCEST for mapping relative changes in Glu concentration as well as pH in vivo. Potential contributions from other brain metabolites to the GluCEST effect are also discussed.

Super-resolution musculoskeletal MRI using deep learning.

Abstract: PURPOSE To develop a super-resolution technique using convolutional neural networks for generating thin-slice knee MR images from thicker input slices, and compare this method with alternative through-plane interpolation methods. METHODS We implemented a 3D convolutional neural network entitled DeepResolve to learn residual-based transformations between high-resolution thin-slice images and lower-resolution thick-slice images at the same center locations. DeepResolve was trained using 124 double echo in steady-state (DESS) data sets with 0.7-mm slice thickness and tested on 17 patients. Ground-truth images were compared with DeepResolve, clinically used tricubic interpolation, and Fourier interpolation methods, along with state-of-the-art single-image sparse-coding super-resolution. Comparisons were performed using structural similarity, peak SNR, and RMS error image quality metrics for a multitude of thin-slice downsampling factors. Two musculoskeletal radiologists ranked the 3 data sets and reviewed the diagnostic quality of the DeepResolve, tricubic interpolation, and ground-truth images for sharpness, contrast, artifacts, SNR, and overall diagnostic quality. Mann-Whitney U tests evaluated differences among the quantitative image metrics, reader scores, and rankings. Cohen's Kappa (κ) evaluated interreader reliability. RESULTS DeepResolve had significantly better structural similarity, peak SNR, and RMS error than tricubic interpolation, Fourier interpolation, and sparse-coding super-resolution for all downsampling factors (p < .05, except 4 × and 8 × sparse-coding super-resolution downsampling factors). In the reader study, DeepResolve significantly outperformed (p < .01) tricubic interpolation in all image quality categories and overall image ranking. Both readers had substantial scoring agreement (κ = 0.73). CONCLUSION DeepResolve was capable of resolving high-resolution thin-slice knee MRI from lower-resolution thicker slices, achieving superior quantitative and qualitative diagnostic performance to both conventionally used and state-of-the-art methods.

A technique for in vivo mapping of myocardial creatine kinase metabolism

Abstract: ATP derived from the conversion of phosphocreatine to creatine by creatine kinase provides an essential chemical energy source that governs myocardial contraction. Here, we demonstrate that the exchange of amine protons from creatine with protons in bulk water can be exploited to image creatine through chemical exchange saturation transfer (CrEST) in myocardial tissue. We show that CrEST provides about two orders of magnitude higher sensitivity compared to (1)H magnetic resonance spectroscopy. Results of CrEST studies from ex vivo myocardial tissue strongly correlate with results from (1)H and (31)P magnetic resonance spectroscopy and biochemical analysis. We demonstrate the feasibility of CrEST measurement in healthy and infarcted myocardium in animal models in vivo on a 3-T clinical scanner. As proof of principle, we show the conversion of phosphocreatine to creatine by spatiotemporal mapping of creatine changes in the exercised human calf muscle. We also discuss the potential utility of CrEST in studying myocardial disorders.

Method for high‐resolution imaging of creatine in vivo using chemical exchange saturation transfer

Abstract: Purpose To develop a chemical exchange saturation transfer (CEST) based technique to measure free creatine (Cr) and to validate the technique by measuring the distribution of Cr in muscle with high spatial resolution before and after exercise.

Exchange Rates of Creatine Kinase Metabolites: Feasibility of Imaging Creatine by Chemical Exchange Saturation Transfer MRI

Abstract: Creatine (Cr), phosphocreatine (PCr) and adenosine-5-triphosphate (ATP) are major metabolites of the enzyme creatine kinase (CK). The exchange rate of amine protons of CK metabolites at physiological conditions has been limited. In the current study, the exchange rate and logarithmic dissociation constant (pKa) of amine protons of CK metabolites were calculated. Further, the chemical exchange saturation transfer effect (CEST) of amine protons of CK metabolites with bulk water was explored. At physiological temperature and pH, the exchange rate of amine protons in Cr was found to be 7-8 times higher than PCr and ATP. A higher exchange rate in Cr was associated with lower pKa value, suggesting faster dissociation of its amine protons compared to PCr and ATP. CEST MR imaging of these metabolites in vitro in phantoms displayed predominant CEST contrast from Cr and negligible contribution from PCr and ATP with the saturation pulse parameters used in the current study. These results provide a new method to perform high-resolution proton imaging of Cr without contamination from PCr. Potential applications of these finding are discussed.

An overview of deep learning in medical imaging focusing on MRI

Abstract: What has happened in machine learning lately, and what does it mean for the future of medical image analysis? Machine learning has witnessed a tremendous amount of attention over the last few years. The current boom started around 2009 when so-called deep artificial neural networks began outperforming other established models on a number of important benchmarks. Deep neural networks are now the state-of-the-art machine learning models across a variety of areas, from image analysis to natural language processing, and widely deployed in academia and industry. These developments have a huge potential for medical imaging technology, medical data analysis, medical diagnostics and healthcare in general, slowly being realized. We provide a short overview of recent advances and some associated challenges in machine learning applied to medical image processing and image analysis. As this has become a very broad and fast expanding field we will not survey the entire landscape of applications, but put particular focus on deep learning in MRI. Our aim is threefold: (i) give a brief introduction to deep learning with pointers to core references; (ii) indicate how deep learning has been applied to the entire MRI processing chain, from acquisition to image retrieval, from segmentation to disease prediction; (iii) provide a starting point for people interested in experimenting and perhaps contributing to the field of machine learning for medical imaging by pointing out good educational resources, state-of-the-art open-source code, and interesting sources of data and problems related medical imaging.

An overview of deep learning in medical imaging focusing on MRI

Abstract: What has happened in machine learning lately, and what does it mean for the future of medical image analysis? Machine learning has witnessed a tremendous amount of attention over the last few years. The current boom started around 2009 when so-called deep artificial neural networks began outperforming other established models on a number of important benchmarks. Deep neural networks are now the state-of-the-art machine learning models across a variety of areas, from image analysis to natural language processing, and widely deployed in academia and industry. These developments have a huge potential for medical imaging technology, medical data analysis, medical diagnostics and healthcare in general, slowly being realized. We provide a short overview of recent advances and some associated challenges in machine learning applied to medical image processing and image analysis. As this has become a very broad and fast expanding field we will not survey the entire landscape of applications, but put particular focus on deep learning in MRI. Our aim is threefold: (i) give a brief introduction to deep learning with pointers to core references; (ii) indicate how deep learning has been applied to the entire MRI processing chain, from acquisition to image retrieval, from segmentation to disease prediction; (iii) provide a starting point for people interested in experimenting and perhaps contributing to the field of deep learning for medical imaging by pointing out good educational resources, state-of-the-art open-source code, and interesting sources of data and problems related medical imaging.

DDcGAN: A Dual-Discriminator Conditional Generative Adversarial Network for Multi-Resolution Image Fusion

Abstract: In this paper, we proposed a new end-to-end model, termed as dual-discriminator conditional generative adversarial network (DDcGAN), for fusing infrared and visible images of different resolutions. Our method establishes an adversarial game between a generator and two discriminators. The generator aims to generate a real-like fused image based on a specifically designed content loss to fool the two discriminators, while the two discriminators aim to distinguish the structure differences between the fused image and two source images, respectively, in addition to the content loss. Consequently, the fused image is forced to simultaneously keep the thermal radiation in the infrared image and the texture details in the visible image. Moreover, to fuse source images of different resolutions, e.g. , a low-resolution infrared image and a high-resolution visible image, our DDcGAN constrains the downsampled fused image to have similar property with the infrared image. This can avoid causing thermal radiation information blurring or visible texture detail loss, which typically happens in traditional methods. In addition, we also apply our DDcGAN to fusing multi-modality medical images of different resolutions, e.g. , a low-resolution positron emission tomography image and a high-resolution magnetic resonance image. The qualitative and quantitative experiments on publicly available datasets demonstrate the superiority of our DDcGAN over the state-of-the-art, in terms of both visual effect and quantitative metrics. Our code is publicly available at https://github.com/jiayi-ma/DDcGAN .

Next-generation probes, particles, and proteins for neural interfacing

Abstract: Bidirectional interfacing with the nervous system enables neuroscience research, diagnosis, and therapy. This two-way communication allows us to monitor the state of the brain and its composite networks and cells as well as to influence them to treat disease or repair/restore sensory or motor function. To provide the most stable and effective interface, the tools of the trade must bridge the soft, ion-rich, and evolving nature of neural tissue with the largely rigid, static realm of microelectronics and medical instruments that allow for readout, analysis, and/or control. In this Review, we describe how the understanding of neural signaling and material-tissue interactions has fueled the expansion of the available tool set. New probe architectures and materials, nanoparticles, dyes, and designer genetically encoded proteins push the limits of recording and stimulation lifetime, localization, and specificity, blurring the boundary between living tissue and engineered tools. Understanding these approaches, their modality, and the role of cross-disciplinary development will support new neurotherapies and prostheses and provide neuroscientists and neurologists with unprecedented access to the brain.