1
|
Radial Undersampling-Based Interpolation Scheme for Multislice CSMRI Reconstruction Techniques. BIOMED RESEARCH INTERNATIONAL 2021; 2021:6638588. [PMID: 33954189 PMCID: PMC8057880 DOI: 10.1155/2021/6638588] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 12/17/2020] [Accepted: 04/05/2021] [Indexed: 11/18/2022]
Abstract
Magnetic Resonance Imaging (MRI) is an important yet slow medical imaging modality. Compressed sensing (CS) theory has enabled to accelerate the MRI acquisition process using some nonlinear reconstruction techniques from even 10% of the Nyquist samples. In recent years, interpolated compressed sensing (iCS) has further reduced the scan time, as compared to CS, by exploiting the strong interslice correlation of multislice MRI. In this paper, an improved efficient interpolated compressed sensing (EiCS) technique is proposed using radial undersampling schemes. The proposed efficient interpolation technique uses three consecutive slices to estimate the missing samples of the central target slice from its two neighboring slices. Seven different evaluation metrics are used to analyze the performance of the proposed technique such as structural similarity index measurement (SSIM), feature similarity index measurement (FSIM), mean square error (MSE), peak signal to noise ratio (PSNR), correlation (CORR), sharpness index (SI), and perceptual image quality evaluator (PIQE) and compared with the latest interpolation techniques. The simulation results show that the proposed EiCS technique has improved image quality and performance using both golden angle and uniform angle radial sampling patterns, with an even lower sampling ratio and maximum information content and using a more practical sampling scheme.
Collapse
|
2
|
Yang Q, Zhang H, Xia J, Zhang X. Evaluation of magnetic resonance image segmentation in brain low-grade gliomas using support vector machine and convolutional neural network. Quant Imaging Med Surg 2021; 11:300-316. [PMID: 33392030 PMCID: PMC7719950 DOI: 10.21037/qims-20-783] [Citation(s) in RCA: 8] [Impact Index Per Article: 2.7] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/21/2020] [Accepted: 08/18/2020] [Indexed: 11/06/2022]
Abstract
BACKGROUND Image segmentation of brain low-grade glioma (LGG) magnetic resonance imaging (MRI) contributes tremendously to diagnosis, classification and treatment of the disease. A tangible, accurate, reliable and fast image segmentation technique is demanded in clinical diagnosis and research. METHODS The emerging machine learning technique has been demonstrated its unique capability in the field of medical image processing, including medical image segmentation. Support vector machine (SVM) and convolutional neural network (CNN) are two widely used machine learning methods. In this work, image segmentation tools based on SVM and CNN are developed and evaluated for brain LGG MR image segmentation studies. The segmentation performance in terms of accuracy and cost is quantitatively analyzed and compared between the SVM and CNN techniques developed. RESULTS Computed on the Google CoLab, each of the 109 SVM models represents an individual patient, is trained using a single image of that patient and takes a few seconds to complete. The CNN model is trained on a drastically larger dataset of 19,760 data augmented images and takes approximately 2 hours to obtain the most optimal result. The SVM models achieved an average and median accuracy of 0.937 and 0.976 respectively, precision of 0.456 and 0.535 respectively, recall of 0.878 and 0.906 respectively, and F1 score of 0.546 and 0.662 respectively. Although the CNN model required a significantly longer calculation time, it surpassed the SVM models in performance in LGG MR image segmentation, achieving an accuracy of 0.998, a precision of 0.999, a recall of 0.999 and an F1 score of 0.999. CONCLUSIONS This study shows that SVM with appropriate filtering techniques is capable of obtaining reliable and fast segmentation of brain LGG MR images with sufficient accuracy and limited image data. CNN technique outperforms SVM in the accuracy of segmentation with requirements of significantly enlarged data set, long computation time and high-performance computer.
Collapse
Affiliation(s)
- Qifan Yang
- Department of Biomedical Engineering, Jacobs School of Medicine and Biomedical Sciences, and School of Engineering and Applied Sciences, State University of New York at Buffalo, Buffalo, NY, USA
| | - Huijuan Zhang
- Department of Biomedical Engineering, Jacobs School of Medicine and Biomedical Sciences, and School of Engineering and Applied Sciences, State University of New York at Buffalo, Buffalo, NY, USA
| | - Jun Xia
- Department of Biomedical Engineering, Jacobs School of Medicine and Biomedical Sciences, and School of Engineering and Applied Sciences, State University of New York at Buffalo, Buffalo, NY, USA
| | - Xiaoliang Zhang
- Department of Biomedical Engineering, Jacobs School of Medicine and Biomedical Sciences, and School of Engineering and Applied Sciences, State University of New York at Buffalo, Buffalo, NY, USA
| |
Collapse
|
3
|
Deng M, Chen SZ, Yuan J, Chan Q, Zhou J, Wáng YXJ. Chemical Exchange Saturation Transfer (CEST) MR Technique for Liver Imaging at 3.0 Tesla: an Evaluation of Different Offset Number and an After-Meal and Over-Night-Fast Comparison. Mol Imaging Biol 2016; 18:274-82. [PMID: 26391991 DOI: 10.1007/s11307-015-0887-8] [Citation(s) in RCA: 20] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/14/2023]
Abstract
PURPOSE This study seeks to explore whether chemical exchange saturation transfer (CEST) magnetic resonance imaging (MRI) can detect liver composition changes between after-meal and over-night-fast statuses. PROCEDURES Fifteen healthy volunteers were scanned on a 3.0-T human MRI scanner in the evening 1.5-2 h after dinner and in the morning after over-night (12-h) fasting. Among them, seven volunteers were scanned twice to assess the scan-rescan reproducibility. Images were acquired at offsets (n = 41, increment = 0.25 ppm) from -5 to 5 ppm using a turbo spin echo (TSE) sequence with a continuous rectangular saturation pulse. Amide proton transfer-weighted (APTw) and GlycoCEST signals were quantified with the asymmetric magnetization transfer ratio (MTRasym) at 3.5 ppm and the total MTRasym integrated from 0.5 to 1.5 ppm from the corrected Z-spectrum, respectively. To explore scan time reduction, CEST images were reconstructed using 31 offsets (with 20% time reduction) and 21 offsets (with 40% time reduction), respectively. RESULTS For reproducibility, GlycoCEST measurements in 41 offsets showed the smallest scan-rescan mean measurements variability, indicated by the lowest mean difference of -0.049% (95% limits of agreement, -0.209 to 0.111%); for APTw, the smallest mean difference was found to be 0.112% (95% limits of agreement, -0.698 to 0.921%) in 41 offsets. Compared with after-meal, both GlycoCEST measurement and APTw measurement under different offset number decreased after 12-h fasting. However, as the offsets number decreased (41 offsets vs. 31 offsets vs. 21 offsets), GlycoCEST map and APTw map became more heterogeneous and noisier. CONCLUSION Our results show that CEST liver imaging at 3.0 T has high sensitivity for fasting.
Collapse
Affiliation(s)
- Min Deng
- Department of Imaging and Interventional Radiology, Faculty of Medicine, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, New Territories, Hong Kong SAR
| | - Shu-Zhong Chen
- Department of Imaging and Interventional Radiology, Faculty of Medicine, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, New Territories, Hong Kong SAR
| | - Jing Yuan
- Medical Physics and Research Department, Hong Kong Sanatorium & Hospital, Happy Valley, Hong Kong SAR
| | - Queenie Chan
- MR Clinical Science, Philips Healthcare Greater China, Shanghai, China
| | - Jinyuan Zhou
- Department of Radiology, Johns Hopkins University, Baltimore, MD, USA.,F.M. Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute, Baltimore, MD, USA
| | - Yì-Xiáng J Wáng
- Department of Imaging and Interventional Radiology, Faculty of Medicine, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, New Territories, Hong Kong SAR.
| |
Collapse
|
4
|
P Krishnan A, Joy A, Paul JS. Improved image reconstruction of low-resolution multichannel phase contrast angiography. J Med Imaging (Bellingham) 2016; 3:014001. [PMID: 26835501 DOI: 10.1117/1.jmi.3.1.014001] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.1] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/12/2015] [Accepted: 12/18/2015] [Indexed: 11/14/2022] Open
Abstract
In low-resolution phase contrast magnetic resonance angiography, the maximum intensity projected channel images will be blurred with consequent loss of vascular details. The channel images are enhanced using a stabilized deblurring filter, applied to each channel prior to combining the individual channel images. The stabilized deblurring is obtained by the addition of a nonlocal regularization term to the reverse heat equation, referred to as nonlocally stabilized reverse diffusion filter. Unlike reverse diffusion filter, which is highly unstable and blows up noise, nonlocal stabilization enhances intensity projected parallel images uniformly. Application to multichannel vessel enhancement is illustrated using both volunteer data and simulated multichannel angiograms. Robustness of the filter applied to volunteer datasets is shown using statistically validated improvement in flow quantification. Improved performance in terms of preserving vascular structures and phased array reconstruction in both simulated and real data is demonstrated using structureness measure and contrast ratio.
Collapse
Affiliation(s)
- Akshara P Krishnan
- Indian Institute of Information Technology and Management-Kerala , Medical Image Computing and Signal Processing Laboratory, Trivandrum 695581, Kerala, India
| | - Ajin Joy
- Indian Institute of Information Technology and Management-Kerala , Medical Image Computing and Signal Processing Laboratory, Trivandrum 695581, Kerala, India
| | - Joseph Suresh Paul
- Indian Institute of Information Technology and Management-Kerala , Medical Image Computing and Signal Processing Laboratory, Trivandrum 695581, Kerala, India
| |
Collapse
|
5
|
Wang YXJ, Lo GG, Yuan J, Larson PEZ, Zhang X. Magnetic resonance imaging for lung cancer screen. J Thorac Dis 2014; 6:1340-8. [PMID: 25276380 DOI: 10.3978/j.issn.2072-1439.2014.08.43] [Citation(s) in RCA: 11] [Impact Index Per Article: 1.1] [Reference Citation Analysis] [Abstract] [Key Words] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/19/2014] [Accepted: 08/20/2014] [Indexed: 12/11/2022]
Abstract
Lung cancer is the leading cause of cancer related death throughout the world. Lung cancer is an example of a disease for which a large percentage of the high-risk population can be easily identified via a smoking history. This has led to the investigation of lung cancer screening with low-dose helical/multi-detector CT. Evidences suggest that early detection of lung cancer allow more timely therapeutic intervention and thus a more favorable prognosis for the patient. The positive relationship of lesion size to likelihood of malignancy has been demonstrated previously, at least 99% of all nodules 4 mm or smaller are benign, while noncalcified nodules larger than 8 mm diameter bear a substantial risk of malignancy. In the recent years, the availability of high-performance gradient systems, in conjunction with phased-array receiver coils and optimized imaging sequences, has made MR imaging of the lung feasible. It can now be assumed a threshold size of 3-4 mm for detection of lung nodules with MRI under the optimal conditions of successful breath-holds with reliable gating or triggering. In these conditions, 90% of all 3-mm nodules can be correctly diagnosed and that nodules 5 mm and larger are detected with 100% sensitivity. Parallel imaging can significantly shorten the imaging acquisition time by utilizing the diversity of sensitivity profile of individual coil elements in multi-channel radiofrequency receive coil arrays or transmit/receive coil arrays to reduce the number of phase encoding steps required in imaging procedure. Compressed sensing technique accelerates imaging acquisition from dramatically undersampled data set by exploiting the sparsity of the images in an appropriate transform domain. With the combined imaging algorithm of parallel imaging and compressed sensing and advanced 32-channel or 64-channel RF hardware, overall imaging acceleration of 20 folds or higher can then be expected, ultimately achieve free-breathing and no ECG gating acquisitions in lung cancer MRI screening. Further development of protocols, more clinical trials and the use of advanced analysis tools will further evaluate the real significance of lung MRI.
Collapse
Affiliation(s)
- Yi-Xiang J Wang
- 1 Department of Imaging and Interventional Radiology, Faculty of Medicine, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, Hong Kong, China ; 2 Department of Diagnostic Radiology, Hong Kong Sanatorium & Hospital, Happy Valley, Hong Kong, China ; 3 Medical Physics and Research Department, Hong Kong Sanatorium & Hospital, Happy Valley, Hong Kong, China ; 4 Department of Radiology and Biomedical Imaging, University of California San Francisco, San Francisco, CA, USA ; 5 UCSF/UC Berkeley Joint Bioengineering Program, San Francisco and Berkeley, CA, USA
| | - Gladys G Lo
- 1 Department of Imaging and Interventional Radiology, Faculty of Medicine, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, Hong Kong, China ; 2 Department of Diagnostic Radiology, Hong Kong Sanatorium & Hospital, Happy Valley, Hong Kong, China ; 3 Medical Physics and Research Department, Hong Kong Sanatorium & Hospital, Happy Valley, Hong Kong, China ; 4 Department of Radiology and Biomedical Imaging, University of California San Francisco, San Francisco, CA, USA ; 5 UCSF/UC Berkeley Joint Bioengineering Program, San Francisco and Berkeley, CA, USA
| | - Jing Yuan
- 1 Department of Imaging and Interventional Radiology, Faculty of Medicine, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, Hong Kong, China ; 2 Department of Diagnostic Radiology, Hong Kong Sanatorium & Hospital, Happy Valley, Hong Kong, China ; 3 Medical Physics and Research Department, Hong Kong Sanatorium & Hospital, Happy Valley, Hong Kong, China ; 4 Department of Radiology and Biomedical Imaging, University of California San Francisco, San Francisco, CA, USA ; 5 UCSF/UC Berkeley Joint Bioengineering Program, San Francisco and Berkeley, CA, USA
| | - Peder E Z Larson
- 1 Department of Imaging and Interventional Radiology, Faculty of Medicine, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, Hong Kong, China ; 2 Department of Diagnostic Radiology, Hong Kong Sanatorium & Hospital, Happy Valley, Hong Kong, China ; 3 Medical Physics and Research Department, Hong Kong Sanatorium & Hospital, Happy Valley, Hong Kong, China ; 4 Department of Radiology and Biomedical Imaging, University of California San Francisco, San Francisco, CA, USA ; 5 UCSF/UC Berkeley Joint Bioengineering Program, San Francisco and Berkeley, CA, USA
| | - Xiaoliang Zhang
- 1 Department of Imaging and Interventional Radiology, Faculty of Medicine, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, Hong Kong, China ; 2 Department of Diagnostic Radiology, Hong Kong Sanatorium & Hospital, Happy Valley, Hong Kong, China ; 3 Medical Physics and Research Department, Hong Kong Sanatorium & Hospital, Happy Valley, Hong Kong, China ; 4 Department of Radiology and Biomedical Imaging, University of California San Francisco, San Francisco, CA, USA ; 5 UCSF/UC Berkeley Joint Bioengineering Program, San Francisco and Berkeley, CA, USA
| |
Collapse
|
6
|
Ji JX, Zhang X. Parallel and sparse MR imaging: methods and instruments-Part 2. Quant Imaging Med Surg 2014; 4:68-70. [PMID: 24834417 DOI: 10.3978/j.issn.2223-4292.2014.04.16] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.2] [Reference Citation Analysis] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/28/2014] [Accepted: 04/29/2014] [Indexed: 11/14/2022]
Affiliation(s)
- Jim X Ji
- 1 Department of Electrical and Computer Engineering, Texas A&M University, 309E WERC, College Station, TX, USA ; 2 Department of Radiology and Biomedical Imaging, University of California San Francisco, San Francisco, CA, USA ; 3 UC Berkeley/UCSF Joint Bioengineering Program, Berkeley, CA, USA ; 4 California Institute for Quantitative Biosciences (QB3), Byers Hall 102D, San Francisco, CA, USA
| | - Xiaoliang Zhang
- 1 Department of Electrical and Computer Engineering, Texas A&M University, 309E WERC, College Station, TX, USA ; 2 Department of Radiology and Biomedical Imaging, University of California San Francisco, San Francisco, CA, USA ; 3 UC Berkeley/UCSF Joint Bioengineering Program, Berkeley, CA, USA ; 4 California Institute for Quantitative Biosciences (QB3), Byers Hall 102D, San Francisco, CA, USA
| |
Collapse
|