• Investigative Magnetic Resonance Imaging
• /
• v.25 no.4
• /
• pp.293-299
• /
• 2021

Purpose: To accelerate magnetic resonance fingerprinting (MRF) by developing a flexible deep learning reconstruction method. Materials and Methods: Synthetic data were used to train a deep learning model. The trained model was then applied to MRF for different organs and diseases. Iterative reconstruction was performed outside the deep learning model, allowing a changeable encoding matrix, i.e., with flexibility of choice for image resolution, radiofrequency coil, k-space trajectory, and undersampling mask. In vivo experiments were performed on normal brain and prostate cancer volunteers to demonstrate the model performance and generalizability. Results: In 400-dynamics brain MRF, direct nonuniform Fourier transform caused a slight increase of random fluctuations on the T2 map. These fluctuations were reduced with the proposed method. In prostate MRF, the proposed method suppressed fluctuations on both T1 and T2 maps. Conclusion: The deep learning and iterative MRF reconstruction method described in this study was flexible with different acquisition settings such as radiofrequency coils. It is generalizable for different in vivo applications.

• Investigative Magnetic Resonance Imaging
• /
• v.25 no.3
• /
• pp.141-155
• /
• 2021

Purpose: To develop a 3D magnetic resonance fingerprinting (MRF) method for application in high resolution knee cartilage PD, T₁, T₂ mapping. Materials and Methods: A novel 3D acquisition trajectory with golden-angle rotating radial in k_xy direction and interleaved echo planar imaging (EPI) acquisition in the k_z direction was implemented in the MRF framework. A centric order was applied to the interleaved EPI acquisition to reduce Nyquist ghosting artifact due to field inhomogeneity. For the reconstruction, singular value decomposition (SVD) compression method was used to accelerate reconstruction time and conjugate gradient sensitivity-encoding (CG-SENSE) was performed to overcome low SNR of the high resolution data. Phantom experiments were performed to verify the proposed method. In vivo experiments were performed on 6 healthy volunteers and 2 early osteoarthritis (OA) patients. Results: In the phantom experiments, the T₁ and T₂ values of the proposed method were in good agreement with the spin-echo references. The results from the in vivo scans showed high quality proton density (PD), T₁, T₂ map with EPI echo train length (N_ETL = 4), acceleration factor in through plane (R_z = 5), and number of radial spokes (N_spk = 4). In patients, high T₂ values (50-60 ms) were seen in all transverse, sagittal, and coronal views and the damaged cartilage regions were in agreement with the hyper-intensity regions shown on conventional turbo spin-echo (TSE) images. Conclusion: The proposed 3D MRF method can acquire high resolution (0.5 mm³) quantitative maps in practical scan time (~ 7 min and 10 sec) with full coverage of the knee (FOV: 160 × 160 × 120 mm³).

• Proceedings of the KSRS Conference
• /
• 2007.10a
• /
• pp.270-273
• /
• 2007

In this paper, a SAR-despeck1ing approach of adaptive iteration based a Bayesian model using the lognormal distribution for image intensity and a Gibbs random field (GRF) for image texture is proposed for noise removal of the images that are corrupted by multiplicative speckle noise. When the image intensity is logarithmically transformed, the speckle noise is approximately Gaussian additive noise, and it tends to a normal probability much faster than the intensity distribution. The MRF is incorporated into digital image analysis by viewing pixel types as states of molecules in a lattice-like physical system. The iterative approach based on MRF is very effective for the inner areas of regions in the observed scene, but may result in yielding false reconstruction around the boundaries due to using wrong information of adjacent regions with different characteristics. The proposed method suggests an adaptive approach using variable parameters depending on the location of reconstructed area, that is, how near to the boundary. The proximity of boundary is estimated by the statistics based on edge value, standard deviation, entropy, and the 4th moment of intensity distribution.

• Journal of Broadcast Engineering
• /
• v.20 no.3
• /
• pp.449-463
• /
• 2015

Video object tracking is important for variety of applications, such as security, video indexing and retrieval, video surveillance, communication, and compression. This paper proposes an object tracking method in HEVC bitstreams. Without pixel reconstruction, motion vector (MV) and size of prediction unit in the bitstream are employed in an Spatio-Temporal Markov Random Fields (ST-MRF) model which represents the spatial and temporal aspects of the object's motion. Coefficient-based object shape adjustment is proposed to solve the over-segmentation and the error propagation problems caused in other methods. In the experimental results, the proposed method provides on average precision of 86.4%, recall of 79.8% and F-measure of 81.1%. The proposed method achieves an F-measure improvement of up to 9% for over-segmented results in the other method even though it provides only average F-measure improvement of 0.2% with respect to the other method. The total processing time is 5.4ms per frame, allowing the algorithm to be applied in real-time applications.

• ETRI Journal
• /
• v.40 no.2
• /
• pp.237-245
• /
• 2018

The Distributed Modulated Wideband Converter (DMWC) is a networking system developed from the Modulated Wideband Converter, which converts all sampling channels into sensing nodes with number variables to implement signal undersampling. When the number of sparse subbands changes, the number of nodes can be adjusted flexibly to improve the reconstruction rate. Owing to the different attenuations of distributed nodes in different locations, it is worthwhile to find out how to select the optimal sensing node as the sampling channel. This paper proposes the spectrum sensing of DMWC based on a Markov random field (MRF) to select the ideal node, which is compared to the image edge segmentation. The attenuation of the candidate nodes is estimated based on the attenuation of the neighboring nodes that have participated in the DMWC system. Theoretical analysis and numerical simulations show that neighboring attenuation plays an important role in determining the node selection, and selecting the node using MRF can avoid serious transmission attenuation. Furthermore, DMWC can greatly improve recovery performance by using a Markov random field compared with random selection.