M. R. Avendi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
M. R. Avendi has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Hamid Jafarkhani
Center for Pervasive Communications and Computing, University of California, Irvine, Irvine, CA, USA
5 shared publications - Arash Kheradvar
Department of Biomedical Engineering, University of California, Irvine, USA
2 shared publications - Dac-Binh Ha
Duy Tan University
1 shared publication - Ha H. Nguyen
Department of Electrical and Computer Engineering, University of Saskatchewan, Saskatoon, SK, Canada
1 shared publication Sina Poorkasmaei
Center for Pervasive Communications and Computing University of California Irvine USA
1 shared publication
Latest Publications
- Top Cited Papers
- 1 May 2016
- Published by Elsevier in Medical Image Analysis
- Vol. 30, 108-119
A combined deep-learning and deformable-model approach to fully automatic segmentation of the left ventricle in cardiac MRI
- 1 May 2016
- Published by Elsevier in Medical Image Analysis
- Vol. 30, 108-119
Top Cited PapersA combined deep-learning and deformable-model approach to fully automatic segmentation of the left ventricle in cardiac MRI
- Open Access
A Combined Deep-Learning and Deformable-Model Approach to Fully Automatic Segmentation of the Left Ventricle in Cardiac MRI
Segmentation of the left ventricle (LV) from cardiac magnetic resonance imaging (MRI) datasets is an essential step for calculation of clinical indices such as ventricular volume and ejection fraction. In this work, we employ deep learning algorithms combined with deformable models to develop and evaluate a fully automatic segmentation tool for the LV from short-axis cardiac MRI datasets. The method employs deep learning algorithms to learn the segmentation task from the ground true data. Convolutional networks are employed to automatically detect the LV chamber in MRI dataset. Stacked autoencoders are utilized to infer the shape of the LV. The inferred shape is incorporated into deformable models to improve the accuracy and robustness of the segmentation. We validated our method using 45 cardiac MR datasets taken from the MICCAI 2009 LV segmentation challenge and showed that it outperforms the state-of-the art methods. Excellent agreement with the ground truth was achieved. Validation metrics, percentage of good contours, Dice metric, average perpendicular distance and conformity, were computed as 96.69%, 0.94, 1.81mm and 0.86, versus those of 79.2%-95.62%, 0.87-0.9, 1.76-2.97mm and 0.67-0.78, obtained by other methods, respectively.
Open AccessA Combined Deep-Learning and Deformable-Model Approach to Fully Automatic Segmentation of the Left Ventricle in Cardiac MRI
Segmentation of the left ventricle (LV) from cardiac magnetic resonance imaging (MRI) datasets is an essential step for calculation of clinical indices such as ventricular volume and ejection fraction. In this work, we employ deep learning algorithms combined with deformable models to develop and evaluate a fully automatic segmentation tool for the LV from short-axis cardiac MRI datasets. The method employs deep learning algorithms to learn the segmentation task from the ground true data. Convolutional networks are employed to automatically detect the LV chamber in MRI dataset. Stacked autoencoders are utilized to infer the shape of the LV. The inferred shape is incorporated into deformable models to improve the accuracy and robustness of the segmentation. We validated our method using 45 cardiac MR datasets taken from the MICCAI 2009 LV segmentation challenge and showed that it outperforms the state-of-the art methods. Excellent agreement with the ground truth was achieved. Validation metrics, percentage of good contours, Dice metric, average perpendicular distance and conformity, were computed as 96.69%, 0.94, 1.81mm and 0.86, versus those of 79.2%-95.62%, 0.87-0.9, 1.76-2.97mm and 0.67-0.78, obtained by other methods, respectively.
- Open Access
Differential Amplify-and-Forward Relaying Using Linear Combining in Time-Varying Channels
Differential encoding and decoding can be employed to circumvent channel estimation in wireless relay networks. This article studies differential amplify-and-forward relaying using linear combining with arbitrary fixed combining weights in time-varying channels. An exact bit error rate (BER) analysis is obtained for this system using DBPSK modulation and over time-varying Rayleigh fading channels. The analysis is verified with simulation results for several sets of combining weights and in various fading scenarios.
Open AccessDifferential Amplify-and-Forward Relaying Using Linear Combining in Time-Varying Channels
Differential encoding and decoding can be employed to circumvent channel estimation in wireless relay networks. This article studies differential amplify-and-forward relaying using linear combining with arbitrary fixed combining weights in time-varying channels. An exact bit error rate (BER) analysis is obtained for this system using DBPSK modulation and over time-varying Rayleigh fading channels. The analysis is verified with simulation results for several sets of combining weights and in various fading scenarios.
- 24 November 2014
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 14 (4) , 1811-1822
Differential Distributed Space-Time Coding With Imperfect Synchronization in Frequency-Selective Channels
Differential distributed space-time coding (D-DSTC) is a cooperative transmission technique that can improve diversity in wireless relay networks in the absence of channel information. Conventionally, it is assumed that channels are flat-fading and relays are perfectly synchronized at the symbol level. However, due to the delay spread in broadband systems and the distributed nature of relay networks, these assumptions may be violated. Hence, inter-symbol interference (ISI) may appear. This paper proposes a new differential encoding and decoding process for D-DSTC systems with multiple relays over slow frequency-selective fading channels with imperfect synchronization. The proposed method overcomes the ISI caused by frequency-selectivity and is robust against synchronization errors while not requiring any channel information at the relays and destination. Moreover, the maximum possible diversity with a decoding complexity similar to that of the conventional D-DSTC is attained. Simulation results are provided to show the performance of the proposed method in various scenarios.
- 24 November 2014
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 14 (4) , 1811-1822
Differential Distributed Space-Time Coding With Imperfect Synchronization in Frequency-Selective Channels
Differential distributed space-time coding (D-DSTC) is a cooperative transmission technique that can improve diversity in wireless relay networks in the absence of channel information. Conventionally, it is assumed that channels are flat-fading and relays are perfectly synchronized at the symbol level. However, due to the delay spread in broadband systems and the distributed nature of relay networks, these assumptions may be violated. Hence, inter-symbol interference (ISI) may appear. This paper proposes a new differential encoding and decoding process for D-DSTC systems with multiple relays over slow frequency-selective fading channels with imperfect synchronization. The proposed method overcomes the ISI caused by frequency-selectivity and is robust against synchronization errors while not requiring any channel information at the relays and destination. Moreover, the maximum possible diversity with a decoding complexity similar to that of the conventional D-DSTC is attained. Simulation results are provided to show the performance of the proposed method in various scenarios.
- Open Access
Differential Distributed Space-Time Coding with Imperfect Synchronization in Frequency-Selective Channels
Differential distributed space-time coding (D-DSTC) is a cooperative transmission technique that can improve diversity in wireless relay networks in the absence of channel information. Conventionally, it is assumed that channels are flat-fading and relays are perfectly synchronized at the symbol level. However, due to the delay spread in broadband systems and the distributed nature of relay networks, these assumptions may be violated. Hence, inter-symbol interference (ISI) may appear. This paper proposes a new differential encoding and decoding process for D-DSTC systems with multiple relays over slow frequency-selective fading channels with imperfect synchronization. The proposed method overcomes the ISI caused by frequency-selectivity and is robust against synchronization errors while not requiring any channel information at the relays and destination. Moreover, the maximum possible diversity with a decoding complexity similar to that of the conventional D-DSTC is attained. Simulation results are provided to show the performance of the proposed method in various scenarios.
Open AccessDifferential Distributed Space-Time Coding with Imperfect Synchronization in Frequency-Selective Channels
Differential distributed space-time coding (D-DSTC) is a cooperative transmission technique that can improve diversity in wireless relay networks in the absence of channel information. Conventionally, it is assumed that channels are flat-fading and relays are perfectly synchronized at the symbol level. However, due to the delay spread in broadband systems and the distributed nature of relay networks, these assumptions may be violated. Hence, inter-symbol interference (ISI) may appear. This paper proposes a new differential encoding and decoding process for D-DSTC systems with multiple relays over slow frequency-selective fading channels with imperfect synchronization. The proposed method overcomes the ISI caused by frequency-selectivity and is robust against synchronization errors while not requiring any channel information at the relays and destination. Moreover, the maximum possible diversity with a decoding complexity similar to that of the conventional D-DSTC is attained. Simulation results are provided to show the performance of the proposed method in various scenarios.
- Open Access
Differential Distributed Space-Time Coding with Imperfect Synchronization
Differential distributed space-time coding (D-DSTC) has been considered to improve both diversity and data-rate in cooperative communications in the absence of channel information. However, conventionally, it is assumed that relays are perfectly synchronized in the symbol level. In practice, this assumption is easily violated due to the distributed nature of the relay networks. This paper proposes a new differential encoding and decoding process for D-DSTC systems with two relays. The proposed method is robust against synchronization errors and does not require any channel information at the destination. Moreover, the maximum possible diversity and symbol-by-symbol decoding are attained. Simulation results are provided to show the performance of the proposed method for various synchronization errors and the fact that our algorithm is not sensitive to synchronization error.
Open AccessDifferential Distributed Space-Time Coding with Imperfect Synchronization
Differential distributed space-time coding (D-DSTC) has been considered to improve both diversity and data-rate in cooperative communications in the absence of channel information. However, conventionally, it is assumed that relays are perfectly synchronized in the symbol level. In practice, this assumption is easily violated due to the distributed nature of the relay networks. This paper proposes a new differential encoding and decoding process for D-DSTC systems with two relays. The proposed method is robust against synchronization errors and does not require any channel information at the destination. Moreover, the maximum possible diversity and symbol-by-symbol decoding are attained. Simulation results are provided to show the performance of the proposed method for various synchronization errors and the fact that our algorithm is not sensitive to synchronization error.