Implications of coronary calcification on the assessment of plaque pathology: a comparison of computed tomography and multimodality intravascular imaging.

OBJECTIVES: This study aimed to investigate the impact of calcific (Ca) on the efficacy of coronary computed coronary angiography (CTA) in evaluating plaque burden (PB) and composition with near-infrared spectroscopy-intravascular ultrasound (NIRS-IVUS) serving as the reference standard. MATERIALS AND METHODS: Sixty-four patients (186 vessels) were recruited and underwent CTA and 3-vessel NIRS-IVUS imaging (NCT03556644). Expert analysts matched and annotated NIRS-IVUS and CTA frames, identifying lumen and vessel wall borders. Tissue distribution was estimated using NIRS chemograms and the arc of Ca on IVUS, while in CTA Hounsfield unit cut-offs were utilized to establish plaque composition. Plaque distribution plots were compared at segment-, lesion-, and cross-sectional-levels. RESULTS: Segment- and lesion-level analysis showed no effect of Ca on the correlation of NIRS-IVUS and CTA estimations. However, at the cross-sectional level, Ca influenced the agreement between NIRS-IVUS and CTA for the lipid and Ca components (p-heterogeneity < 0.001). Proportional odds model analysis revealed that Ca had an impact on the per cent atheroma volume quantification on CTA compared to NIRS-IVUS at the segment level (p-interaction < 0.001). At lesion level, Ca affected differences between the modalities for maximum PB, remodelling index, and Ca burden (p-interaction < 0.001, 0.029, and 0.002, respectively). Cross-sectional-level modelling demonstrated Ca's effect on differences between modalities for all studied variables (p-interaction ≤ 0.002). CONCLUSION: Ca burden influences agreement between NIRS-IVUS and CTA at the cross-sectional level and causes discrepancies between the predictions for per cent atheroma volume at the segment level and maximum PB, remodelling index, and Ca burden at lesion-level analysis. CLINICAL RELEVANCE STATEMENT: Coronary calcification affects the quantification of lumen and plaque dimensions and the characterization of plaque composition coronary CTA. This should be considered in the analysis and interpretation of CTAs performed in patients with extensive Ca burden. KEY POINTS: Coronary CT Angiography is limited in assessing coronary plaques by resolution and blooming artefacts. Agreement between dual-source CT angiography and NIRS-IVUS is affected by a Ca burden for the per cent atheroma volume. Advanced CT imaging systems that eliminate blooming artefacts enable more accurate quantification of coronary artery disease and characterisation of plaque morphology.

Cross-sectional angle prediction of lipid-rich and calcified tissue on computed tomography angiography images.

PURPOSE: The assessment of vulnerable plaque characteristics and distribution is important to stratify cardiovascular risk in a patient. Computed tomography angiography (CTA) offers a promising alternative to invasive imaging but is limited by the fact that the range of Hounsfield units (HU) in lipid-rich areas overlaps with the HU range in fibrotic tissue and that the HU range of calcified plaques overlaps with the contrast within the contrast-filled lumen. This paper is to investigate whether lipid-rich and calcified plaques can be detected more accurately on cross-sectional CTA images using deep learning methodology. METHODS: Two deep learning (DL) approaches are proposed, a 2.5D Dense U-Net and 2.5D Mask-RCNN, which separately perform the cross-sectional plaque detection in the Cartesian and polar domain. The spread-out view is used to evaluate and show the prediction result of the plaque regions. The accuracy and F1-score are calculated on a lesion level for the DL and conventional plaque detection methods. RESULTS: For the lipid-rich plaques, the median and mean values of the F1-score calculated by the two proposed DL methods on 91 lesions were approximately 6 and 3 times higher than those of the conventional method. For the calcified plaques, the F1-score of the proposed methods was comparable to those of the conventional method. The median F1-score of the Dense U-Net-based method was 3% higher than that of the conventional method. CONCLUSION: The two methods proposed in this paper contribute to finer cross-sectional predictions of lipid-rich and calcified plaques compared to studies focusing only on longitudinal prediction. The angular prediction performance of the proposed methods outperforms the convincing conventional method for lipid-rich plaque and is comparable for calcified plaque.

Efficacy of human experts and an automated segmentation algorithm in quantifying disease pathology in coronary computed tomography angiography: A head-to-head comparison with intravascular ultrasound imaging.

BACKGROUND: Coronary computed tomography angiography (CCTA) analysis is currently performed by experts and is a laborious process. Fully automated edge-detection methods have been developed to expedite CCTA segmentation however their use is limited as there are concerns about their accuracy. This study aims to compare the performance of an automated CCTA analysis software and the experts using near-infrared spectroscopy-intravascular ultrasound imaging (NIRS-IVUS) as a reference standard. METHODS: Fifty-one participants (150 vessels) with chronic coronary syndrome who underwent CCTA and 3-vessel NIRS-IVUS were included. CCTA analysis was performed by an expert and an automated edge detection method and their estimations were compared to NIRS-IVUS at a segment-, lesion-, and frame-level. RESULTS: Segment-level analysis demonstrated a similar performance of the two CCTA analyses (conventional and automatic) with large biases and limits of agreement compared to NIRS-IVUS estimations for the total atheroma (ICC: 0.55 vs 0.25, mean difference:192 (-102-487) vs 243 (-132-617) and percent atheroma volume (ICC: 0.30 vs 0.12, mean difference: 12.8 (-5.91-31.6) vs 20.0 (0.79-39.2). Lesion-level analysis showed that the experts were able to detect more accurately lesions than the automated method (68.2 â€‹% and 60.7 â€‹%) however both analyses had poor reliability in assessing the minimal lumen area (ICC 0.44 vs 0.36) and the maximum plaque burden (ICC 0.33 vs 0.33) when NIRS-IVUS was used as the reference standard. CONCLUSIONS: Conventional and automated CCTA analyses had similar performance in assessing coronary artery pathology using NIRS-IVUS as a reference standard. Therefore, automated segmentation can be used to expedite CCTA analysis and enhance its applications in clinical practice.

Novel near-infrared spectroscopy-intravascular ultrasound-based deep-learning methodology for accurate coronary computed tomography plaque quantification and characterization.

Aims: Coronary computed tomography angiography (CCTA) is inferior to intravascular imaging in detecting plaque morphology and quantifying plaque burden. We aim to, for the first time, train a deep-learning (DL) methodology for accurate plaque quantification and characterization in CCTA using near-infrared spectroscopy-intravascular ultrasound (NIRS-IVUS). Methods and results: Seventy patients were prospectively recruited who underwent CCTA and NIRS-IVUS imaging. Corresponding cross sections were matched using an in-house developed software, and the estimations of NIRS-IVUS for the lumen, vessel wall borders, and plaque composition were used to train a convolutional neural network in 138 vessels. The performance was evaluated in 48 vessels and compared against the estimations of NIRS-IVUS and the conventional CCTA expert analysis. Sixty-four patients (186 vessels, 22 012 matched cross sections) were included. Deep-learning methodology provided estimations that were closer to NIRS-IVUS compared with the conventional approach for the total atheroma volume (ΔDL-NIRS-IVUS: -37.8 ± 89.0 vs. ΔConv-NIRS-IVUS: 243.3 ± 183.7 mm3, variance ratio: 4.262, P < 0.001) and percentage atheroma volume (-3.34 ± 5.77 vs. 17.20 ± 7.20%, variance ratio: 1.578, P < 0.001). The DL methodology detected lesions more accurately than the conventional approach (Area under the curve (AUC): 0.77 vs. 0.67, P < 0.001) and quantified minimum lumen area (ΔDL-NIRS-IVUS: -0.35 ± 1.81 vs. ΔConv-NIRS-IVUS: 1.37 ± 2.32 mm2, variance ratio: 1.634, P < 0.001), maximum plaque burden (4.33 ± 11.83% vs. 5.77 ± 16.58%, variance ratio: 2.071, P = 0.004), and calcific burden (-51.2 ± 115.1 vs. -54.3 ± 144.4, variance ratio: 2.308, P < 0.001) more accurately than conventional approach. The DL methodology was able to segment a vessel on CCTA in 0.3 s. Conclusions: The DL methodology developed for CCTA analysis from co-registered NIRS-IVUS and CCTA data enables rapid and accurate assessment of lesion morphology and is superior to expert analysts (Clinicaltrials.gov: NCT03556644).

Quantitative error prediction of medical image registration using regression forests.

Predicting registration error can be useful for evaluation of registration procedures, which is important for the adoption of registration techniques in the clinic. In addition, quantitative error prediction can be helpful in improving the registration quality. The task of predicting registration error is demanding due to the lack of a ground truth in medical images. This paper proposes a new automatic method to predict the registration error in a quantitative manner, and is applied to chest CT scans. A random regression forest is utilized to predict the registration error locally. The forest is built with features related to the transformation model and features related to the dissimilarity after registration. The forest is trained and tested using manually annotated corresponding points between pairs of chest CT scans in two experiments: SPREAD (trained and tested on SPREAD) and inter-database (including three databases SPREAD, DIR-Lab-4DCT and DIR-Lab-COPDgene). The results show that the mean absolute errors of regression are 1.07  ±â€¯ 1.86 and 1.76  ±â€¯ 2.59 mm for the SPREAD and inter-database experiment, respectively. The overall accuracy of classification in three classes (correct, poor and wrong registration) is 90.7% and 75.4%, for SPREAD and inter-database respectively. The good performance of the proposed method enables important applications such as automatic quality control in large-scale image analysis.

Robust contour propagation using deep learning and image registration for online adaptive proton therapy of prostate cancer.

PURPOSE: To develop and validate a robust and accurate registration pipeline for automatic contour propagation for online adaptive Intensity-Modulated Proton Therapy (IMPT) of prostate cancer using elastix software and deep learning. METHODS: A three-dimensional (3D) Convolutional Neural Network was trained for automatic bladder segmentation of the computed tomography (CT) scans. The automatic bladder segmentation alongside the computed tomography (CT) scan is jointly optimized to add explicit knowledge about the underlying anatomy to the registration algorithm. We included three datasets from different institutes and CT manufacturers. The first was used for training and testing the ConvNet, where the second and the third were used for evaluation of the proposed pipeline. The system performance was quantified geometrically using the dice similarity coefficient (DSC), the mean surface distance (MSD), and the 95% Hausdorff distance (HD). The propagated contours were validated clinically through generating the associated IMPT plans and compare it with the IMPT plans based on the manual delineations. Propagated contours were considered clinically acceptable if their treatment plans met the dosimetric coverage constraints on the manual contours. RESULTS: The bladder segmentation network achieved a DSC of 88% and 82% on the test datasets. The proposed registration pipeline achieved a MSD of 1.29 ± 0.39, 1.48 ± 1.16, and 1.49 ± 0.44 mm for the prostate, seminal vesicles, and lymph nodes, respectively, on the second dataset and a MSD of 2.31 ± 1.92 and 1.76 ± 1.39 mm for the prostate and seminal vesicles on the third dataset. The automatically propagated contours met the dose coverage constraints in 86%, 91%, and 99% of the cases for the prostate, seminal vesicles, and lymph nodes, respectively. A Conservative Success Rate (CSR) of 80% was obtained, compared to 65% when only using intensity-based registration. CONCLUSION: The proposed registration pipeline obtained highly promising results for generating treatment plans adapted to the daily anatomy. With 80% of the automatically generated treatment plans directly usable without manual correction, a substantial improvement in system robustness was reached compared to a previous approach. The proposed method therefore facilitates more precise proton therapy of prostate cancer, potentially leading to fewer treatment-related adverse side effects.

A deep learning framework for unsupervised affine and deformable image registration.

Image registration, the process of aligning two or more images, is the core technique of many (semi-)automatic medical image analysis tasks. Recent studies have shown that deep learning methods, notably convolutional neural networks (ConvNets), can be used for image registration. Thus far training of ConvNets for registration was supervised using predefined example registrations. However, obtaining example registrations is not trivial. To circumvent the need for predefined examples, and thereby to increase convenience of training ConvNets for image registration, we propose the Deep Learning Image Registration (DLIR) framework for unsupervised affine and deformable image registration. In the DLIR framework ConvNets are trained for image registration by exploiting image similarity analogous to conventional intensity-based image registration. After a ConvNet has been trained with the DLIR framework, it can be used to register pairs of unseen images in one shot. We propose flexible ConvNets designs for affine image registration and for deformable image registration. By stacking multiple of these ConvNets into a larger architecture, we are able to perform coarse-to-fine image registration. We show for registration of cardiac cine MRI and registration of chest CT that performance of the DLIR framework is comparable to conventional image registration while being several orders of magnitude faster.

PelVis: Atlas-based Surgical Planning for Oncological Pelvic Surgery.

Due to the intricate relationship between the pelvic organs and vital structures, such as vessels and nerves, pelvic anatomy is often considered to be complex to comprehend. In oncological pelvic surgery, a trade-off has to be made between complete tumor resection and preserving function by preventing damage to the nerves. Damage to the autonomic nerves causes undesirable post-operative side-effects such as fecal and urinal incontinence, as well as sexual dysfunction in up to 80 percent of the cases. Since these autonomic nerves are not visible in pre-operative MRI scans or during surgery, avoiding nerve damage during such a surgical procedure becomes challenging. In this work, we present visualization methods to represent context, target, and risk structures for surgical planning. We employ distance-based and occlusion management techniques in an atlas-based surgical planning tool for oncological pelvic surgery. Patient-specific pre-operative MRI scans are registered to an atlas model that includes nerve information. Through several interactive linked views, the spatial relationships and distances between the organs, tumor and risk zones are visualized to improve understanding, while avoiding occlusion. In this way, the surgeon can examine surgically relevant structures and plan the procedure before going into the operating theater, thus raising awareness of the autonomic nerve zone regions and potentially reducing post-operative complications. Furthermore, we present the results of a domain expert evaluation with surgical oncologists that demonstrates the advantages of our approach.