Self-supervised learning enables 3D digital subtraction angiography reconstruction from ultra-sparse 2D projection views: A multicenter study.
Ontology highlight
ABSTRACT:
SUBMITTER: Zhao H
PROVIDER: S-EPMC9589028 | biostudies-literature | 2022 Oct
REPOSITORIES: biostudies-literature
Publications
Self-supervised learning enables 3D digital subtraction angiography reconstruction from ultra-sparse 2D projection views: A multicenter study.
Zhao Huangxuan H Zhou Zhenghong Z Wu Feihong F Xiang Dongqiao D Zhao Hui H Zhang Wei W Li Lin L Li Zhong Z Huang Jia J Hu Hongyao H Liu Chengbo C Wang Tao T Liu Wenyu W Ma Jinqiang J Yang Fan F Wang Xinggang X Zheng Chuansheng C
Cell reports. Medicine 20221007 10
3D digital subtraction angiography (DSA) reconstruction from rotational 2D projection X-ray angiography is an important basis for diagnosis and treatment of intracranial aneurysms (IAs). The gold standard requires approximately 133 different projection views for 3D reconstruction. A method to significantly reduce the radiation dosage while ensuring the reconstruction quality is yet to be developed. We propose a self-supervised learning method to realize 3D-DSA reconstruction using ultra-sparse 2 ...[more]
Similar Datasets
Project description:BackgroundComputed tomography (CT) is an advanced medical imaging technology. The images obtained by CT are helpful for improving diagnostic accuracy. Currently, CT is widely used in clinical settings for diagnosis and health examinations. However, full angle CT scanning has the disadvantage of causing radiation damage to the human body. Sparse angle projection CT scanning is the most effective way to minimize this damage, but the quality of the reconstructed image is reduced. Therefore, it is important to improve the reconstructed image quality produced by sparse angle projection.MethodsIn this paper, we focused on the algebraic reconstruction algorithm. To reduce the accumulation of random noise, we formulated a modified algebraic reconstruction algorithm. Firstly, the algebraic reconstruction algorithm was used to compute two consecutive results, and then the weighted sum of these two results was used to correct the reconstructed image, and an iterative result was obtained. Using this method, we aimed to reduce the noise accumulation caused by iteration.ResultsIn this study, 20 angle projections were used for the reconstruction. The experimental object was the Shepp-Logan phantom test image. The experiments were implemented under two conditions: without noise and with noise. The peak signal to noise ratio (PSNR) and the mean squared error (MSE) of the reconstructed image from projections without noise were 76.0896 and 0.0016, respectively. The PSNR and MSE of the reconstructed image from projections with noise were 75.8263 and 0.0017, respectively. The reconstructed performance was superior to the previous algebraic reconstruction algorithm.ConclusionsThe performance of the proposed method was superior to other algorithms, which confirms that noise accumulation caused by iteration can be effectively reduced by the weighted summation of two consecutive reconstruction results. Moreover, the reconstruction performance under noisy projection is superior to other algorithms, which demonstrates that the proposed method improves anti-noise performance.
| S-EPMC8506772 | biostudies-literature
Project description:ObjectiveStimulated Raman projection tomography (SRPT), a recently developed label-free volumetric chemical imaging technology, has been reported to quantitatively reconstruct the distribution of chemicals in a three-dimensional (3D) complex system. The current image reconstruction scheme used in SRPT is based on a filtered back projection (FBP) algorithm that requires at least 180 angular-dependent projections to rebuild a reasonable SRPT image, resulting in a long total acquisition time. This is a big limitation for longitudinal studies on live systems.MethodsWe present a sparse-view data-based sparse reconstruction scheme, in which sparsely sampled projections at 180 degrees were used to reconstruct the volumetric information. In the scheme, the simultaneous algebra reconstruction technique (SART), combined with total variation regularization, was used for iterative reconstruction. To better describe the projection process, a pixel vertex driven model (PVDM) was developed to act as projectors, whose performance was compared with those of the distance driven model (DDM).ResultsWe evaluated our scheme with numerical simulations and validated it for SRPT by mapping lipid contents in adipose cells. Simulation results showed that the PVDM performed better than the DDM in the case of using sparse-view data. Our scheme could maintain the quality of the reconstructed images even when the projection number was reduced to 15. The cell-based experimental results demonstrated that the proposed scheme can improve the imaging speed of the current FBP-based SRPT scheme by a factor of 9-12 without sacrificing discernible imaging details.ConclusionOur proposed scheme significantly reduces the total acquisition time required for SRPT at a speed of one order of magnitude faster than the currently used scheme. This significant improvement in imaging speed would potentially promote the applicability of SRPT for imaging living organisms.
| S-EPMC7329365 | biostudies-literature
Project description:BackgroundAccurate detection of Hepatocellular carcinoma (HCC) feeding vessels during transcatheter arterial chemoembolization (TACE) is important for an effective treatment, while limiting non-target embolization. This study aimed to investigate the feasibility and accuracy of pre-TACE three dimensional (3D) CT angiography for tumor-feeding vessels detection compared to DSA.MethodsSixty-nine consecutive patients referred for TACE from May 2022 to May 2023 were included. (3D) CT images were reconstructed from the pre-TACE diagnostic multiphasic contrast enhanced CT images and compared with non-selective digital subtraction angiography (DSA) images obtained during TACE for detection of HCC feeding vessels. A "Ground truth" made by consensus between observers after reviewing all available pre-TACE CT images, and DSA and CBCT images during TACE to detect the true feeding vessels was the gold standard. Sensitivity, specificity, negative predictive value (NPV), positive predictive value (PPV), accuracy and ROC curve with AUC were calculated for each modality and compared.ResultsA total of 136 active HCCs were detected in the 69 consecutive patients included in the study. 185 feeding arteries were detected by 3D CT and DSA and included in the analysis. 3D CT detection of feeding arteries revealed mean sensitivity, specificity, PPV, NPV and accuracy of 91%, 71%, 98%, 36%, and 90%, respectively, with mean AUC = 0.81. DSA detection of feeding arteries revealed mean sensitivity, specificity, PPV, NPV, and accuracy of 80%, 58%, 96.5%, 16.5% and 78%, respectively, with mean AUC = 0.69.ConclusionsPre-TACE 3D CT angiography has shown promise in improving the detection of HCC feeding vessels compared to DSA. However, further studies are required to confirm these findings across different clinical settings and patient populations.Trial registrationThis study was prospectively registered at Clinicaltrials.gov with ID NCT05304572; Date of registration: 2-4-2022.
| S-EPMC11412056 | biostudies-literature
Project description:Circulating whole blood transcriptomes can detect the presence of unruptured cerebral aneurysms. Pending additional testing in larger cohorts, this could serve as a foundation to develop a simple blood-based test to facilitate screening and early detection of cerebral aneurysms.
2020-10-22 | GSE159610 | GEO
Project description:Background2D digital subtraction angiography (DSA) is utilized qualitatively to assess blood velocity changes that occur during arterial interventions. Quantitative angiographic metrics, such as blood velocity, could be used to standardize endpoints during angiographic interventions.PurposeTo assess the accuracy and precision of a quantitative 2D DSA (qDSA) technique and to determine its feasibility for in vivo measurements of blood velocity.Materials and methodsA quantitative DSA technique was developed to calculate intra-procedural blood velocity. In vitro validation was performed by comparing velocities from the qDSA method and an ultrasonic flow probe in a bifurcation phantom. Parameters of interest included baseline flow rate, contrast injection rate, projection angle, and magnification. In vivo qDSA analysis was completed in five different branches of the abdominal aorta in two 50 kg swine and compared to 4D Flow MRI. Linear regression, Bland-Altman, Pearson's correlation coefficient and chi squared tests were used to assess the accuracy and precision of the technique.ResultsIn vitro validation showed strong correlation between qDSA and flow probe velocities over a range of contrast injection and baseline flow rates (slope = 1.012, 95% CI [0.989,1.035], Pearson's r = 0.996, p < .0001). The application of projection angle and magnification corrections decreased variance to less than 5% the average baseline velocity (p = 0.999 and p = 0.956, respectively). In vivo validation showed strong correlation with a small bias between qDSA and 4D Flow MRI velocities for all five abdominopelvic arterial vessels of interest (slope = 1.01, Pearson's r = 0.880, p = <.01, Bias = 0.117 cm/s).ConclusionThe proposed method allows for accurate and precise calculation of blood velocities, in near real-time, from time resolved 2D DSAs.
| S-EPMC7790988 | biostudies-literature
Project description:ObjectiveEvaluation of a new iterative reconstruction algorithm (IMR) for detection/rule-out of pulmonary embolism (PE) in ultra-low dose computed tomography pulmonary angiography (CTPA).MethodsLower dose CT data sets were simulated based on CTPA examinations of 16 patients with pulmonary embolism (PE) with dose levels (DL) of 50%, 25%, 12.5%, 6.3% or 3.1% of the original tube current setting. Original CT data sets and simulated low-dose data sets were reconstructed with three reconstruction algorithms: the standard reconstruction algorithm "filtered back projection" (FBP), the first generation iterative reconstruction algorithm iDose and the next generation iterative reconstruction algorithm "Iterative Model Reconstruction" (IMR). In total, 288 CTPA data sets (16 patients, 6 tube current levels, 3 different algorithms) were evaluated by two blinded radiologists regarding image quality, diagnostic confidence, detectability of PE and contrast-to-noise ratio (CNR).ResultsiDose and IMR showed better detectability of PE than FBP. With IMR, sensitivity for detection of PE was 100% down to a dose level of 12.5%. iDose and IMR showed superiority to FBP regarding all characteristics of subjective (diagnostic confidence in detection of PE, image quality, image noise, artefacts) and objective image quality. The minimum DL providing acceptable diagnostic performance was 12.5% (= 0.45 mSv) for IMR, 25% (= 0.89 mSv) for iDose and 100% (= 3.57 mSv) for FBP. CNR was significantly (p < 0.001) improved by IMR compared to FBP and iDose at all dose levels.ConclusionBy using IMR for detection of PE, dose reduction for CTPA of up to 75% is possible while maintaining full diagnostic confidence. This would result in a mean effective dose of approximately 0.9 mSv for CTPA.
| S-EPMC5017721 | biostudies-literature
Project description:Background/objectivesDetailed morphometric analysis of an aneurysm and the related vascular bifurcation are critical factors when determining rupture risk and planning treatment for unruptured intracranial aneurysms (UIAs). The standard visualization of digital subtraction angiography (DSA) and its 3D reconstruction on a 2D monitor provide precise measurements but are subject to variability based on the rater. Visualization using virtual (VR) and augmented reality platforms can overcome those limitations. It is, however, unclear whether accurate measurements of the aneurysm and adjacent arterial branches can be obtained on VR models. This study aimed to assess interrater reliability and compare measurements between 3D VR, standard 2D DSA, and 3D DSA reconstructions, evaluating the reliability and accuracy of 3D VR as a measurement tool.MethodsA pool of five neurosurgeons performed three individual analyses on each of the ten UIA cases, measuring them in completely immersed 3D VR and the standard on-screen format (2D DSA and 3D reconstruction). This resulted in three independent measurements per modality for each case. Interrater reliability of measurements and morphology characterization, comparative differences, measurement duration, and VR user experience were assessed.ResultsInterrater reliability for 3D VR measurements was significantly higher than for 3D DSA measurements (3D VR mean intraclass correlation coefficient [ICC]: 0.69 ± 0.22 vs. 3D DSA mean ICC: 0.36 ± 0.37, p = 0.042). No significant difference was observed between 3D VR and 2D DSA (3D VR mean ICC: 0.69 ± 0.22 vs. 2D DSA mean ICC: 0.43 ± 0.31, p = 0.12). A linear mixed-effects model showed no effect of 3D VR and 3D DSA (95% CI = -0.26-0.28, p = 0.96) or 3D VR and 2D DSA (95% CI = -0.02-0.53, p = 0.066) on absolute measurements of the aneurysm in the anteroposterior, mediolateral, and craniocaudal dimensions.Conclusions3D VR technology allows for reproducible, accurate, and reliable measurements comparable to measurements performed on a 2D screen. It may also potentially improve precision for measurements of non-planar aneurysm dimensions.
| S-EPMC11506597 | biostudies-literature
Project description:Recent years have seen active developments of various penalized regression methods, such as LASSO and elastic net, to analyze high dimensional data. In these approaches, the direction and length of the regression coefficients are determined simultaneously. Due to the introduction of penalties, the length of the estimates can be far from being optimal for accurate predictions. We introduce a new framework, regression by projection, and its sparse version to analyze high dimensional data. The unique nature of this framework is that the directions of the regression coefficients are inferred first, and the lengths and the tuning parameters are determined by a cross validation procedure to achieve the largest prediction accuracy. We provide a theoretical result for simultaneous model selection consistency and parameter estimation consistency of our method in high dimension. This new framework is then generalized such that it can be applied to principal components analysis, partial least squares and canonical correlation analysis. We also adapt this framework for discriminant analysis. Compared to the existing methods, where there is relatively little control of the dependency among the sparse components, our method can control the relationships among the components. We present efficient algorithms and related theory for solving the sparse regression by projection problem. Based on extensive simulations and real data analysis, we demonstrate that our method achieves good predictive performance and variable selection in the regression setting, and the ability to control relationships between the sparse components leads to more accurate classification. In supplemental materials available online, the details of the algorithms and theoretical proofs, and R codes for all simulation studies are provided.
| S-EPMC4560121 | biostudies-literature
Project description:To investigate the feasibility of a new two-dimensional (2D) multichannel response (MCR) analysis approach for the detection of clustered microcalcifications (MCs) in digital breast tomosynthesis (DBT).With IRB approval and informed consent, a data set of two-view DBTs from 42 breasts containing biopsy-proven MC clusters was collected in this study. The authors developed a 2D approach for MC detection using projection view (PV) images rather than the reconstructed three-dimensional (3D) DBT volume. Signal-to-noise ratio (SNR) enhancement processing was first applied to each PV to enhance the potential MCs. The locations of MC candidates were then identified with iterative thresholding. The individual MCs were decomposed with Hermite-Gaussian (HG) and Laguerre-Gaussian (LG) basis functions and the channelized Hotelling model was trained to produce the MCRs for each MC on the 2D images. The MCRs from the PVs were fused in 3D by a coincidence counting method that backprojects the MC candidates on the PVs and traces the coincidence of their ray paths in 3D. The 3D MCR was used to differentiate the true MCs from false positives (FPs). Finally a dynamic clustering method was used to identify the potential MC clusters in the DBT volume based on the fact that true MCs of clinical significance appear in clusters. Using two-fold cross validation, the performance of the 3D MCR for classification of true and false MCs was estimated by the area under the receiver operating characteristic (ROC) curve and the overall performance of the MCR approach for detection of clustered MCs was assessed by free response receiver operating characteristic (FROC) analysis.When the HG basis function was used for MCR analysis, the detection of MC cluster achieved case-based test sensitivities of 80% and 90% at the average FP rates of 0.65 and 1.55 FPs per DBT volume, respectively. With LG basis function, the average FP rates were 0.62 and 1.57 per DBT volume at the same sensitivity levels. The difference in the two sets of basis functions for detection of MCs did not show statistical significance.The authors' experimental results indicate that the MCR approach is promising for the detection of MCs on PV images. The HG or LG basis functions are both effective in characterizing the signal response of MCs using the channelized Hotelling model. The coincidence counting method for fusion of the 2D MCR in 3D is an important step for FP reduction. Further study is underway to improve the MCR approach for microcalcification detection in DBT.
| S-EPMC3971829 | biostudies-other
Project description:PurposeTo achieve ultra-high temporal resolution non-contrast 4D angiography with improved spatiotemporal fidelity.MethodsContinuous data acquisition using 3D golden-angle sampling following an arterial spin labeling preparation allows for flexibly reconstructing 4D dynamic angiograms at arbitrary temporal resolutions. However, k-space data is often temporally "binned" before image reconstruction, negatively affecting spatiotemporal fidelity and limiting temporal resolution. In this work, a subspace was extracted by linearly compressing a dictionary constructed from simulated curves of an angiographic kinetic model. The subspace was used to represent and reconstruct the voxelwise signal timecourse at the same temporal resolution as the data acquisition without temporal binning. Physiological parameters were estimated from the resulting images using a Bayesian fitting approach. A group of eight healthy subjects were scanned and the in vivo results reconstructed by different methods were compared. Because of the difficulty of obtaining ground truth 4D angiograms with ultra-high temporal resolution, the in vivo results were cross-validated with numerical simulations.ResultsThe proposed method enables 4D time-resolved angiography with much higher temporal resolution (14.7 ms) than previously reported (˜50 ms) while maintaining high spatial resolution (1.1 mm isotropic). Blood flow dynamics were depicted in greater detail, thin vessel visibility was improved, and the estimated physiological parameters also exhibited more realistic spatial patterns with the proposed method.ConclusionIncorporating a subspace compressed kinetic model into the reconstruction of 4D ASL angiograms notably improved the temporal resolution and spatiotemporal fidelity, which was subsequently beneficial for accurate physiological modeling.
| S-EPMC11893029 | biostudies-literature