Temporal phase unwrapping using deep learning

Yin, Wei; Chen, Qian; Feng, Shijie; Tao, Tianyang; Huang, Lei; Trusiak, Maciej; Asundi, Anand Krishna; Zuo, Chao

doi:10.1038/s41598-019-56222-3

Cited by 106 publications

(43 citation statements)

References 48 publications

(47 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One approach would be to train a network to directly estimate the unwrapped phase from the potentially-wrapped input phase, i.e., treating the problem as a regression problem [ 42 , 43 ]. Another approach, the one we took, is to estimate the integer number of wrap cycles at each pixel of the phase map by training a semantic-segmentation network to label each pixel according to its wrap class as defined in Table 1 [ 35 , 44 – 46 ]. While both approaches are reasonable, we selected the semantic-segmentation approach because, by recognizing DENSE phase wrap patterns, we reasoned that the method may be effective even for low-SNR images; furthermore, the regression output does not apply any constraints to the output phase so that the network may yield unrealistic values.…”

Section: Discussionmentioning

confidence: 99%

Fully-automated global and segmental strain analysis of DENSE cardiovascular magnetic resonance using deep learning for segmentation and phase unwrapping

Ghadimi

Auger

Feng

et al. 2021

Journal of Cardiovascular Magnetic Resonance

View full text Add to dashboard Cite

Background Cardiovascular magnetic resonance (CMR) cine displacement encoding with stimulated echoes (DENSE) measures heart motion by encoding myocardial displacement into the signal phase, facilitating high accuracy and reproducibility of global and segmental myocardial strain and providing benefits in clinical performance. While conventional methods for strain analysis of DENSE images are faster than those for myocardial tagging, they still require manual user assistance. The present study developed and evaluated deep learning methods for fully-automatic DENSE strain analysis. Methods Convolutional neural networks (CNNs) were developed and trained to (a) identify the left-ventricular (LV) epicardial and endocardial borders, (b) identify the anterior right-ventricular (RV)-LV insertion point, and (c) perform phase unwrapping. Subsequent conventional automatic steps were employed to compute strain. The networks were trained using 12,415 short-axis DENSE images from 45 healthy subjects and 19 heart disease patients and were tested using 10,510 images from 25 healthy subjects and 19 patients. Each individual CNN was evaluated, and the end-to-end fully-automatic deep learning pipeline was compared to conventional user-assisted DENSE analysis using linear correlation and Bland Altman analysis of circumferential strain. Results LV myocardial segmentation U-Nets achieved a DICE similarity coefficient of 0.87 ± 0.04, a Hausdorff distance of 2.7 ± 1.0 pixels, and a mean surface distance of 0.41 ± 0.29 pixels in comparison with manual LV myocardial segmentation by an expert. The anterior RV-LV insertion point was detected within 1.38 ± 0.9 pixels compared to manually annotated data. The phase-unwrapping U-Net had similar or lower mean squared error vs. ground-truth data compared to the conventional path-following method for images with typical signal-to-noise ratio (SNR) or low SNR (p < 0.05), respectively. Bland–Altman analyses showed biases of 0.00 ± 0.03 and limits of agreement of − 0.04 to 0.05 or better for deep learning-based fully-automatic global and segmental end-systolic circumferential strain vs. conventional user-assisted methods. Conclusions Deep learning enables fully-automatic global and segmental circumferential strain analysis of DENSE CMR providing excellent agreement with conventional user-assisted methods. Deep learning-based automatic strain analysis may facilitate greater clinical use of DENSE for the quantification of global and segmental strain in patients with cardiac disease.

show abstract

Section: Discussionmentioning

confidence: 99%

Fully-automated global and segmental strain analysis of DENSE cardiovascular magnetic resonance using deep learning for segmentation and phase unwrapping

Ghadimi

Auger

Feng

et al. 2021

Journal of Cardiovascular Magnetic Resonance

View full text Add to dashboard Cite

show abstract

“…Feng et al [167] employed a deep neural network to improve the accuracy of phase demodulation from an input fringe pattern. Compared with FT and WFT methods, the deep learning-based method is more accurate and can effectively perform temporal phase unwrapping even under harsh conditions [245]. A properly trained deep neural network is capable of high-quality 3D shape reconstructions for transient scenes [246].…”

Section: Deep Learningmentioning

confidence: 99%

Development of surface reconstruction algorithms for optical interferometric measurement

2020

Front. Mech. Eng.

View full text Add to dashboard Cite

Optical interferometry is a powerful tool for measuring and characterizing areal surface topography in precision manufacturing. A variety of instruments based on optical interferometry have been developed to meet the measurement needs in various applications, but the existing techniques are simply not enough to meet the ever-increasing requirements in terms of accuracy, speed, robustness, and dynamic range, especially in on-line or on-machine conditions. This paper provides an in-depth perspective of surface topography reconstruction for optical interferometric measurements. Principles, configurations, and applications of typical optical interferometers with different capabilities and limitations are presented. Theoretical background and recent advances of fringe analysis algorithms, including coherence peak sensing and phase-shifting algorithm, are summarized. The new developments in measurement accuracy and repeatability, noise resistance, self-calibration ability, and computational efficiency are discussed. This paper also presents the new challenges that optical interferometry techniques are facing in surface topography measurement. To address these challenges, advanced techniques in image stitching, on-machine measurement, intelligent sampling, parallel computing, and deep learning are explored to improve the functional performance of optical interferometry in future manufacturing metrology.

show abstract

“…For instance, Feng et al [ 27 , 28 ] integrated the deep CNNs with the phase-shifting scheme for unwrapped phase detections. Yin et al [ 29 ] removed the phase ambiguities in the phase unwrapping process with two groups of phase-shifted fringe patterns and deep learning. Jeught [ 30 ] proposed a neural network with a large simulated training dataset to acquire the height information from a single fringe pattern.…”

Section: Introductionmentioning

confidence: 99%

Single-Shot 3D Shape Reconstruction Using Structured Light and Deep Convolutional Neural Networks

Nguyen

Wang

2020

Sensors

View full text Add to dashboard Cite

Single-shot 3D imaging and shape reconstruction has seen a surge of interest due to the ever-increasing evolution in sensing technologies. In this paper, a robust single-shot 3D shape reconstruction technique integrating the structured light technique with the deep convolutional neural networks (CNNs) is proposed. The input of the technique is a single fringe-pattern image, and the output is the corresponding depth map for 3D shape reconstruction. The essential training and validation datasets with high-quality 3D ground-truth labels are prepared by using a multi-frequency fringe projection profilometry technique. Unlike the conventional 3D shape reconstruction methods which involve complex algorithms and intensive computation to determine phase distributions or pixel disparities as well as depth map, the proposed approach uses an end-to-end network architecture to directly carry out the transformation of a 2D image to its corresponding 3D depth map without extra processing. In the approach, three CNN-based models are adopted for comparison. Furthermore, an accurate structured-light-based 3D imaging dataset used in this paper is made publicly available. Experiments have been conducted to demonstrate the validity and robustness of the proposed technique. It is capable of satisfying various 3D shape reconstruction demands in scientific research and engineering applications.

show abstract

Temporal phase unwrapping using deep learning

Cited by 106 publications

References 48 publications

Fully-automated global and segmental strain analysis of DENSE cardiovascular magnetic resonance using deep learning for segmentation and phase unwrapping

Fully-automated global and segmental strain analysis of DENSE cardiovascular magnetic resonance using deep learning for segmentation and phase unwrapping

Development of surface reconstruction algorithms for optical interferometric measurement

Single-Shot 3D Shape Reconstruction Using Structured Light and Deep Convolutional Neural Networks

Contact Info

Product

Resources

About