Evaluation of motion artifact metrics for coronary CT angiography

TT U-Net: Temporal Transformer U-Net for Motion Artifact Reduction Using PAD (Pseudo All-Phase Clinical-Dataset) in Cardiac CT

Involuntary motion of the heart remains a challenge for cardiac computed tomography (CT) imaging. Although the electrocardiogram (ECG) gating strategy is widely adopted to perform CT scans at the quasi-quiescent cardiac phase, motion-induced artifacts are still unavoidable for patients with high heart rates or irregular rhythms. Dynamic cardiac CT, which provides functional information of the heart, suffers even more severe motion artifacts. In this paper, we develop a deep learning based framework for motion artifact reduction in dynamic cardiac CT. First, we build a PAD (Pseudo All-phase clinical-Dataset) based on a whole-heart motion model and single-phase cardiac CT images. This dataset provides dynamic CT images with realistic-looking motion artifacts that help to develop data-driven approaches. Second, we formulate the problem of motion artifact reduction as a video deblurring task according to its dynamic nature. A novel TT U-Net (Temporal Transformer U-Net) is proposed to excavate the spatiotemporal features for better motion artifact reduction. The self-attention mechanism along the temporal dimension effectively encodes motion information and thus aids image recovery. Experiments show that the TT U-Net trained on the proposed PAD performs well on clinical CT scans, which substantiates the effectiveness and fine generalization ability of our method. The source code, trained models, and dynamic demo will be available at https://github.com/ivy9092111111/TT-U-Net.

Deep learning-based motion correction algorithm for coronary CT angiography: Lowering the phase requirement for morphological and functional evaluation

PURPOSE

To investigate the performance of a deep learning-based motion correction algorithm (MCA) at various cardiac phases of coronary computed tomography angiography (CCTA), and determine the extent to which it may allow for reliable morphological and functional evaluation.

MATERIALS AND METHODS

The acquired image data of 53 CCTA cases, where the patient heart rate (HR) was ≥75 bpm, were reconstructed at 0, ±2, ±4, ±6, and ±8% deviations from each optimal systolic phase, with and without the MCA, yielding a total of 954 images (53 cases × 9 phases × 2 reconstructions). The overall image quality and diagnostic confidence were graded by two radiologists using a 5-point scale, with scores ≥3 being deemed clinically interpretable. Signal-to-noise ratio, contrast-to-noise ratio, vessel sharpness, and circularity were measured. The CCTA-derived fractional flow reserve (CT-FFR) was calculated in 38 vessels on 24 patients to identify functionally significant stenosis, using the invasive fractional flow reserve (FFR) as reference. All metrics were compared between two reconstructions at various phases.

RESULTS

Inferior image quality was observed as the phase deviation was enlarged. However, MCA significantly improved the image quality at nonoptimal phases and the optimal phase. Coronary artery evaluation was feasible within 4% phase deviation using MCA, with interpretable overall image quality and high diagnostic confidence. With MCA, the performance of identifying functionally significant stenosis via CT-FFR was increased for images at various phase deviations. However, obvious decrease in accuracy, as compared to the image at the optimal phase, was found on those with deviations >4%.

CONCLUSION

The deep learning-based MCA allows up to 4% phase deviation in acquiring CCTA for reliable morphological and functional evaluation on patients with high HRs.

High-pitch, high temporal resolution, multi-energy cardiac imaging on a dual-source photon-counting-detector CT

OBJECTIVE

To measure the accuracy of material decomposition using a dual-source photon-counting-detector (DS-PCD) CT operated in the high-pitch helical scanning mode and compare the results against dual-source energy-integrating-detector (DS-EID) CT, which requires use of a low-pitch value in dual-energy mode.

METHODS

A DS-PCD CT and a DS-EID CT were used to scan a cardiac motion phantom consisting of a 3-mm diameter iodine cylinder. Iodine maps were reconstructed using DS-PCD in high-pitch mode and DS-EID in low-pitch mode. Image-based circularity, diameter, and iodine concentration of the iodine cylinder were calculated and compared between the two scanners. With institutional review board approval, in vivo exams were performed with the DS-PCD CT in high-pitch mode. Images were qualitatively compared against patients with similar heart rates that were scanned with DS-EID CT in low-pitch dual-energy mode.

RESULTS

On iodine maps, the mean circularity was 0.97 ± 0.02 with DS-PCD in high-pitch mode and 0.95 ± 0.06 with DS-EID in low-pitch mode. The mean diameter was 2.9 ± 0.2 mm with DS-PCD and 3.1 ± 0.2 mm with DS-EID, both of which are close to the 3 mm ground truth. For DS-PCD, the mean iodine concentration was 9.6 ± 0.8 mg/ml and this was consistent with the 9.4 ± 0.6 mg/ml value obtained with the cardiac motion disabled. For DS-EID, the concentration was 12.7 ± 1.2 mg/ml with motion enabled and 11.7 ± 0.5 mg/ml disabled. The background noise in the iodine maps was 15.1 HU with DS-PCD and 14.4 HU with DS-EID, whereas the volume CT dose index (CTDIvol ) was 3 mGy with DS-PCD and 11 mGy with DS-EID. On comparison of six patients (three on PCD, three on EID) with similar heart rates, DS-PCD provided iodine maps with well-defined coronaries even at a high heart rate of 86 beats per minute. Meanwhile, there were substantial motion artifacts in iodine maps obtained with DS-EID for patients with similar heart rates.

CONCLUSION

In a cardiac motion phantom, DS-PCD CT can perform accurate material decomposition in high-pitch mode, providing iodine maps with excellent geometric accuracy and robustness to motion at approximately 38% of the dose for similar noise as DS-EID CT.

Quantitative assessment of motion effects in dual-source dual-energy CT and dual-source photon-counting detector CT

Conventional dual-source CT scanners can be used to either provide better temporal resolution or dual-energy imaging, but not both at the same time. This presents a dilemma in cardiac CT as both high temporal resolution and multi-energy imaging are desirable. The current study evaluated a dual-source photon-counting-detector (DS-PCD) CT which can acquire multi-energy images at high temporal resolution. A cardiac motion phantom with a 3-mm diameter iodinated rod, mimicking the right coronary artery, was scanned 25 times using a DS-PCD CT (66 ms resolution) and a dual-source dual-energy (DS-DE, 125 ms resolution) CT. Low/high energy images and iodine maps were reconstructed at 40% and 75% cardiac phases. To quantify the impact of motion on image quality, dice similarity coefficient was computed between the low/high energy images while the circularity and effective diameter of the iodinated rod were computed on the iodine maps. The dice coefficients were higher for DS-PCD with a mean of 0.89 and 0.91 at the 40% and 70% phases, while DS-DE had a lower mean of 0.20 and 0.78, respectively. The circularity was excellent for DS-PCD with a mean of 0.97 and 0.98 at the 40% and 75% phases, while DS-DE had a mean of 0.71 and 0.98, respectively. The effective diameter was accurate for DS-PCD with a mean of 2.9 mm (true size of 3 mm) at both phases, while DS-DE had a mean of 4.0 mm and 3.2 mm at the 40% and 75% phases, respectively. These results indicate that DS-PCD CT enables simultaneous high temporal resolution and multi-energy cardiac imaging with minimal motion artifacts.

Motion correction for routine X-ray lung CT imaging

A novel motion correction algorithm for X-ray lung CT imaging has been developed recently. It was designed to perform for routine chest or thorax CT scans without gating, namely axial or helical scans with pitch around 1.0. The algorithm makes use of two conjugate partial angle reconstruction images for motion estimation via non-rigid registration which is followed by a motion compensated reconstruction. Differently from other conventional approaches, no segmentation is adopted in motion estimation. This makes motion estimation of various fine lung structures possible. The aim of this study is to explore the performance of the proposed method in correcting the lung motion artifacts which arise even under routine CT scans with breath-hold. The artifacts are known to mimic various lung diseases, so it is of great interest to address the problem. For that purpose, a moving phantom experiment and clinical study (seven cases) were conducted. We selected the entropy and positivity as figure of merits to compare the reconstructed images before and after the motion correction. Results of both phantom and clinical studies showed a statistically significant improvement by the proposed method, namely up to 53.6% (p < 0.05) and up to 35.5% (p < 0.05) improvement by means of the positivity measure, respectively. Images of the proposed method show significantly reduced motion artifacts of various lung structures such as lung parenchyma, pulmonary vessels, and airways which are prominent in FBP images. Results of two exemplary cases also showed great potential of the proposed method in correcting motion artifacts of the aorta which is known to mimic aortic dissection. Compared to other approaches, the proposed method provides an excellent performance and a fully automatic workflow. In addition, it has a great potential to handle motions in wide range of organs such as lung structures and the aorta. We expect that this would pave a way toward innovations in chest and thorax CT imaging.

Task-dependent estimability index to assess the quality of cardiac computed tomography angiography for quantifying coronary stenosis

Purpose: Quantifying stenosis in cardiac computed tomography angiography (CTA) images remains a difficult task, as image noise and cardiac motion can degrade image quality and distort underlying anatomic information. The purpose of this study was to develop a computational framework to objectively assess the precision of quantifying coronary stenosis in cardiac CTA.

Approach: The framework used models of coronary vessels and plaques, asymmetric motion point spread functions, CT image blur (task-based modulation transfer functions) and noise (noise-power spectrums), and an automated maximum-likelihood estimator implemented as a matched template squared-difference operator. These factors were integrated into an estimability index (e′) as a task-based measure of image quality in cardiac CTA. The e′ index was applied to assess how well it can to predict the quality of 132 clinical cases selected from the Prospective Multicenter Imaging Study for Evaluation of Chest Pain trial. The cases were divided into two cohorts, high quality and low quality, based on clinical scores and the concordance of clinical evaluations of cases by experienced cardiac imagers. The framework was also used to ascertain protocol factors for CTA Biomarker initiative of the Quantitative Imaging Biomarker Alliance (QIBA).

Results: The e′ index categorized the patient datasets with an area under the curve of 0.985, an accuracy of 0.977, and an optimal e′ threshold of 25.58 corresponding to a stenosis estimation precision (standard deviation) of 3.91%. Data resampling and training–test validation methods demonstrated stable classifier thresholds and receiver operating curve performance. The framework was successfully applicable to the QIBA objective.

Conclusions: A computational framework to objectively quantify stenosis estimation task performance was successfully implemented and was reflective of clinical results in the context of a prominent clinical trial with diverse sites, readers, scanners, acquisition protocols, and patients. It also demonstrated the potential for prospective optimization of imaging protocols toward targeted precision and measurement consistency in cardiac CT images.

High-quality initial image-guided 4D CBCT reconstruction

PURPOSE

Four-dimensional cone-beam computed tomography (4D CBCT) has been developed to provide a sequence of phase-resolved reconstructions in image-guided radiation therapy. However, 4D CBCT images are degraded by severe streaking artifacts because the 4D CBCT reconstruction process is an extreme sparse-view CT procedure wherein only under-sampled projections are used for the reconstruction of each phase. To obtain a set of 4D CBCT images achieving both high spatial and temporal resolution, we propose an algorithm by providing a high-quality initial image at the beginning of the iterative reconstruction process for each phase to guide the final reconstructed result toward its optimal solution.

METHODS

The proposed method consists of three steps to generate the initial image. First, a prior image is obtained by an iterative reconstruction method using the measured projections of the entire set of 4D CBCT images. The prior image clearly shows the appearance of structures in static regions, although it contains blurring artifacts in motion regions. Second, the robust principal component analysis (RPCA) model is adopted to extract the motion components corresponding to each phase-resolved image. Third, a set of initial images are produced by the proposed linear estimation model that combines the prior image and the RPCA-decomposed motion components. The final 4D CBCT images are derived from the simultaneous algebraic reconstruction technique (SART) equipped with the initial images. Qualitative and quantitative evaluations were performed by using two extended cardiac-torso (XCAT) phantoms and two sets of patient data. Several state-of-the-art 4D CBCT algorithms were performed for comparison to validate the performance of the proposed method.

RESULTS

The image quality of phase-resolved images is greatly improved by the proposed method in both phantom and patient studies. The results show an outstanding spatial resolution, in which streaking artifacts are suppressed to a large extent, while detailed structures such as tumors and blood vessels are well restored. Meanwhile, the proposed method depicts a high temporal resolution with a distinct respiratory motion change at different phases. For simulation phantom, quantitative evaluations of the simulation data indicate that an average of 36.72% decrease at EI phase and 42% decrease at EE phase in terms of root-mean-square error (RMSE) are achieved by our method when comparing with PICCS algorithm in Phantom 1 and Phantom 2. In addition, the proposed method has the lowest entropy and the highest normalized mutual information compared with the existing methods in simulation experiments, such as PRI, RPCA-4DCT, SMART, and PICCS. And for real patient cases, the proposed method also achieves the lowest entropy value compared with the competitive method.

CONCLUSIONS

The proposed algorithm can generate an optimal initial image to improve iterative reconstruction performance. The final sequence of phase-resolved volumes guided by the initial image achieves high spatiotemporal resolution by eliminating motion-induced artifacts. This study presents a practical 4D CBCT reconstruction method with leading image quality.

Motion-corrected coronary calcium scores by a convolutional neural network: a robotic simulating study

OBJECTIVE

To classify motion-induced blurred images of calcified coronary plaques so as to correct coronary calcium scores on nontriggered chest CT, using a deep convolutional neural network (CNN) trained by images of motion artifacts.

METHODS

Three artificial coronary arteries containing nine calcified plaques of different densities (high, medium, and low) and sizes (large, medium, and small) were attached to a moving robotic arm. The artificial arteries moving at 0-90 mm/s were scanned to generate nine categories (each from one calcified plaque) of images with motion artifacts. An inception v3 CNN was fine-tuned and validated. Agatston scores of the predicted classification by CNN were considered as corrected scores. Variation of Agatston scores on moving plaque and by CNN correction was calculated using the scores at rest as reference.

RESULTS

The overall accuracy of CNN classification was 79.2 ± 6.1% for nine categories. The accuracy was 88.3 ± 4.9%, 75.9 ± 6.4%, and 73.5 ± 5.0% for the high-, medium-, and low-density plaques, respectively. Compared with the Agatston score at rest, the overall median score variation was 37.8% (1st and 3rd quartile, 10.5% and 68.8%) in moving plaques. CNN correction largely decreased the variation to 3.7% (1.9%, 9.1%) (p < 0.001, Mann-Whitney U test) and improved the sensitivity (percentage of non-zero scores among all the scores) from 65 to 85% for detection of coronary calcifications.

CONCLUSIONS

In this experimental study, CNN showed the ability to classify motion-induced blurred images and correct calcium scores derived from nontriggered chest CT. CNN correction largely reduces the overall Agatston score variation and increases the sensitivity to detect calcifications.

KEY POINTS

• A deep CNN architecture trained by CT images of motion artifacts showed the ability to correct coronary calcium scores from blurred images. • A correction algorithm based on deep CNN can be used for a tenfold reduction in Agatston score variations from 38 to 3.7% of moving coronary calcified plaques and to improve the sensitivity from 65 to 85% for the detection of calcifications. • This experimental study provides a method to improve its accuracy for coronary calcium scores that is a fundamental step towards a real clinical scenario.

Cardiovascular optoacoustics: From mice to men - A review

Imaging has become an indispensable tool in the research and clinical management of cardiovascular disease (CVD). An array of imaging technologies is considered for CVD diagnostics and therapeutic assessment, ranging from ultrasonography, X-ray computed tomography and magnetic resonance imaging to nuclear and optical imaging methods. Each method has different operational characteristics and assesses different aspects of CVD pathophysiology; nevertheless, more information is desirable for achieving a comprehensive view of the disease. Optoacoustic (photoacoustic) imaging is an emerging modality promising to offer novel information on CVD parameters by allowing high-resolution imaging of optical contrast several centimeters deep inside tissue. Implemented with illumination at several wavelengths, multi-spectral optoacoustic tomography (MSOT) in particular, is sensitive to oxygenated and deoxygenated hemoglobin, water and lipids allowing imaging of the vasculature, tissue oxygen saturation and metabolic or inflammatory parameters. Progress with fast-tuning lasers, parallel detection and advanced image reconstruction and data-processing algorithms have recently transformed optoacoustics from a laboratory tool to a promising modality for small animal and clinical imaging. We review progress with optoacoustic CVD imaging, highlight the research and diagnostic potential and current applications and discuss the advantages, limitations and possibilities for integration into clinical routine.

Motion artifact recognition and quantification in coronary CT angiography using convolutional neural networks

Excellent image quality is a primary prerequisite for diagnostic non-invasive coronary CT angiography. Artifacts due to cardiac motion may interfere with detection and diagnosis of coronary artery disease and render subsequent treatment decisions more difficult. We propose deep-learning-based measures for coronary motion artifact recognition and quantification in order to assess the diagnostic reliability and image quality of coronary CT angiography images. More specifically, the application, steering and evaluation of motion compensation algorithms can be triggered by these measures. A Coronary Motion Forward Artifact model for CT data (CoMoFACT) is developed and applied to clinical cases with excellent image quality to introduce motion artifacts using simulated motion vector fields. The data required for supervised learning is generated by the CoMoFACT from 17 prospectively ECG-triggered clinical cases with controlled motion levels on a scale of 0-10. Convolutional neural networks achieve an accuracy of 93.3% ± 1.8% for the classification task of separating motion-free from motion-perturbed coronary cross-sectional image patches. The target motion level is predicted by a corresponding regression network with a mean absolute error of 1.12 ± 0.07. Transferability and generalization capabilities are demonstrated by motion artifact measurements on eight additional CCTA cases with real motion artifacts.

Automated quantification and evaluation of motion artifact on coronary CT angiography images

PURPOSE

This study developed and validated a Motion Artifact Quantification algorithm to automatically quantify the severity of motion artifacts on coronary computed tomography angiography (CCTA) images. The algorithm was then used to develop a Motion IQ Decision method to automatically identify whether a CCTA dataset is of sufficient diagnostic image quality or requires further correction.

METHOD

The developed Motion Artifact Quantification algorithm includes steps to identify the right coronary artery (RCA) regions of interest (ROIs), segment vessel and shading artifacts, and to calculate the motion artifact score (MAS) metric. The segmentation algorithms were verified against ground-truth manual segmentations. The segmentation algorithms were also verified by comparing and analyzing the MAS calculated from ground-truth segmentations and the algorithm-generated segmentations. The Motion IQ Decision algorithm first identifies slices with unsatisfactory image quality using a MAS threshold. The algorithm then uses an artifact-length threshold to determine whether the degraded vessel segment is large enough to cause the dataset to be nondiagnostic. An observer study on 30 clinical CCTA datasets was performed to obtain the ground-truth decisions of whether the datasets were of sufficient image quality. A five-fold cross-validation was used to identify the thresholds and to evaluate the Motion IQ Decision algorithm.

RESULTS

The automated segmentation algorithms in the Motion Artifact Quantification algorithm resulted in Dice coefficients of 0.84 for the segmented vessel regions and 0.75 for the segmented shading artifact regions. The MAS calculated using the automated algorithm was within 10% of the values obtained using ground-truth segmentations. The MAS threshold and artifact-length thresholds were determined by the ROC analysis to be 0.6 and 6.25 mm by all folds. The Motion IQ Decision algorithm demonstrated 100% sensitivity, 66.7% ± 27.9% specificity, and a total accuracy of 86.7% ± 12.5% for identifying datasets in which the RCA required correction. The Motion IQ Decision algorithm demonstrated 91.3% sensitivity, 71.4% specificity, and a total accuracy of 86.7% for identifying CCTA datasets that need correction for any of the three main vessels.

CONCLUSION

The Motion Artifact Quantification algorithm calculated accurate (<10% error) motion artifact scores using the automated segmentation methods. The developed algorithms demonstrated high sensitivity (91.3%) and specificity (71.4%) in identifying datasets of insufficient image quality. The developed algorithms for automatically quantifying motion artifact severity may be useful for comparing acquisition techniques, improving best-phase selection algorithms, and evaluating motion compensation techniques.