Fully automated pipeline for measurement of the thoracic aorta using joint segmentation and localization neural network.

Purpose: Diagnosis and surveillance of thoracic aortic aneurysm (TAA) involves measuring the aortic diameter at various locations along the length of the aorta, often using computed tomography angiography (CTA). Currently, measurements are performed by human raters using specialized software for three-dimensional analysis, a time-consuming process, requiring 15 to 45 min of focused effort. Thus, we aimed to develop a convolutional neural network (CNN)-based algorithm for fully automated and accurate aortic measurements. Approach: Using 212 CTA scans, we trained a CNN to perform segmentation and localization of key landmarks jointly. Segmentation mask and landmarks are subsequently used to obtain the centerline and cross-sectional diameters of the aorta. Subsequently, a cubic spline is fit to the aortic boundary at the sinuses of Valsalva to avoid errors related inclusions of coronary artery origins. Performance was evaluated on a test set of 60 scans with automated measurements compared against expert manual raters. Result: Compared to training separate networks for each task, joint training yielded higher accuracy for segmentation, especially at the boundary (p<0.001), but a marginally worse (0.2 to 0.5 mm) accuracy for landmark localization (p<0.001). Mean absolute error between human and automated was ≤1 mm at six of nine standard clinical measurement locations. However, higher errors were noted in the aortic root and arch regions, ranging between 1.4 and 2.2 mm, although agreement of manual raters was also lower in these regions. Conclusion: Fully automated aortic diameter measurements in TAA are feasible using a CNN-based algorithm. Automated measurements demonstrated low errors that are comparable in magnitude to those with manual raters; however, measurement error was highest in the aortic root and arch.

Label Propagation via Random Walk for Training Robust Thalamus Nuclei Parcellation Model from Noisy Annotations.

Data-driven thalamic nuclei parcellation depends on high-quality manual annotations. However, the small size and low contrast changes among thalamic nuclei, yield annotations that are often incomplete, noisy, or ambiguously labelled. To train a robust thalamic nuclei parcellation model with noisy annotations, we propose a label propagation algorithm based on random walker to refine the annotations before model training. A two-step model was trained to generate first the whole thalamus and then the nuclei masks. We conducted experiments on a mild traumatic brain injury~(mTBI) dataset with noisy thalamic nuclei annotations. Our model outperforms current state-of-the-art thalamic nuclei parcellations by a clear margin. We believe our method can also facilitate the training of other parcellation models with noisy labels.

Recurrent Self Fusion: Iterative Denoising for Consistent Retinal OCT Segmentation.

Optical coherence tomography (OCT) is a valuable imaging technique in ophthalmology, providing high-resolution, cross-sectional images of the retina for early detection and monitoring of various retinal and neurological diseases. However, discrepancies in retinal layer thickness measurements among different OCT devices pose challenges for data comparison and interpretation, particularly in longitudinal analyses. This work introduces the idea of a recurrent self fusion (RSF) algorithm to address this issue. Our RSF algorithm, built upon the self fusion methodology, iteratively denoises retinal OCT images. A deep learning-based retinal OCT segmentation algorithm is employed for downstream analyses. A large dataset of paired OCT scans acquired on both a Spectralis and Cirrus OCT device are used for validation. The results demonstrate that the RSF algorithm effectively reduces speckle contrast and enhances the consistency of retinal OCT segmentation.

Mask removal : Face inpainting via attributes.

Due to the outbreak of the COVID-19 pandemic, wearing masks in public areas has become an effective way to slow the spread of disease. However, it also brings some challenges to applications in daily life as half of the face is occluded. Therefore, the idea of removing masks by face inpainting appeared. Face inpainting has achieved promising performance but always fails to guarantee high-fidelity. In this paper, we present a novel mask removal inpainting network based on face attributes known in advance including nose, chubby, makeup, gender, mouth, beard and young, aiming to ensure the repaired face image is closer to ground truth. To achieve this, a dual pipeline network based on GANs has been proposed, one of which is a reconstructive path used in training that utilizes missing regions in ground truth to get prior distribution, while the other is a generative path for predicting information in the masked region. To establish the process of mask removal, we build a synthetic facial occlusion that mimics the real mask. Experiments show that our method not only generates faces more similarly aligned with real attributes, but also ensures semantic and structural rationality compared with state-of-the-art methods.

Validation of a robust method for quantification of three-dimensional growth of the thoracic aorta using deformable image registration.

PURPOSE: Accurate assessment of thoracic aortic aneurysm (TAA) growth is important for appropriate clinical management. Maximal aortic diameter is the primary metric that is used to assess growth, but it suffers from substantial measurement variability. A recently proposed technique, termed vascular deformation mapping (VDM), is able to quantify three-dimensional aortic growth using clinical computed tomography angiography (CTA) data using an approach based on deformable image registration (DIR). However, the accuracy and robustness of VDM remains undefined given the lack of ground truth from clinical CTA data, and, furthermore, the performance of VDM relative to standard manual diameter measurements is unknown. METHODS: To evaluate the performance of the VDM pipeline for quantifying aortic growth, we developed a novel and systematic evaluation process to generate 76 unique synthetic CTA growth phantoms (based on 10 unique cases) with variable degrees and locations of aortic wall deformation. Aortic deformation was quantified using two metrics: area ratio (AR), defined as the ratio of surface area in triangular mesh elements and the magnitude of deformation in the normal direction (DiN) relative to the aortic surface. Using these phantoms, we further investigated the effects on VDM's measurement accuracy resulting from factors that influence the quality of clinical CTA data such as respiratory translations, slice thickness, and image noise. Lastly, we compare the measurement error of VDM TAA growth assessments against two expert raters performing standard diameter measurements of synthetic phantom images. RESULTS: Across our population of 76 synthetic growth phantoms, the median absolute error was 0.063 (IQR: 0.073-0.054) for AR and 0.181 mm (interquartile range [IQR]: 0.214-0.143 mm) for DiN. Median relative error was 1.4% for AR and 3.3 % $3.3\%$ for DiN at the highest tested noise level (contrast-to-noise ratio [CNR] = 2.66). Error in VDM output increased with slice thickness, with the highest median relative error of 1.5% for AR and 4.1% for DiN at a slice thickness of 2.0 mm. Respiratory motion of the aorta resulted in maximal absolute error of 3% AR and 0.6 mm in DiN, but bulk translations in aortic position had a very small effect on measured AR and DiN values (relative errors < 1 % $< 1\%$ ). VDM-derived measurements of magnitude and location of maximal diameter change demonstrated significantly high accuracy and lower variability compared to two expert manual raters ( p < 0.03 $p<0.03$ across all comparisons). CONCLUSIONS: VDM yields an accurate, three-dimensional assessment of aortic growth in TAA patients and is robust to factors such as image noise, respiration-induced translations, and differences in patient position. Further, VDM significantly outperformed two expert manual raters in assessing the magnitude and location of aortic growth despite optimized experimental measurement conditions. These results support validation of the VDM technique for accurate quantification of aortic growth in patients and highlight several important advantages over diameter measurements.

Vascular Deformation Mapping for CT Surveillance of Thoracic Aortic Aneurysm Growth.

Background Aortic diameter measurements in patients with a thoracic aortic aneurysm (TAA) show wide variation. There is no technique to quantify aortic growth in a three-dimensional (3D) manner. Purpose To validate a CT-based technique for quantification of 3D growth based on deformable registration in patients with TAA. Materials and Methods Patients with ascending and descending TAA with two or more CT angiography studies between 2006 and 2020 were retrospectively identified. The 3D aortic growth was quantified using vascular deformation mapping (VDM), a technique that uses deformable registration to warp a mesh constructed from baseline aortic anatomy. Growth assessments between VDM and clinical CT diameter measurements were compared. Aortic growth was quantified as the ratio of change in surface area at each mesh element (area ratio). Manual segmentations were performed by independent raters to assess interrater reproducibility. Registration error was assessed using manually placed landmarks. Agreement between VDM and clinical diameter measurements was assessed using Pearson correlation and Cohen κ coefficients. Results A total of 38 patients (68 surveillance intervals) were evaluated (mean age, 69 years ± 9 [standard deviation]; 21 women), with TAA involving the ascending aorta (n = 26), descending aorta (n = 10), or both (n = 2). VDM was technically successful in 35 of 38 (92%) patients and 58 of 68 intervals (85%). Median registration error was 0.77 mm (interquartile range, 0.54-1.10 mm). Interrater agreement was high for aortic segmentation (Dice similarity coefficient = 0.97 ± 0.02) and VDM-derived area ratio (bias = 0.0, limits of agreement: -0.03 to 0.03). There was strong agreement (r = 0.85, P < .001) between peak area ratio values and diameter change. VDM detected growth in 14 of 58 (24%) intervals. VDM revealed growth outside the maximally dilated segment in six of 14 (36%) growth intervals, none of which were detected with diameter measurements. Conclusion Vascular deformation mapping provided reliable and comprehensive quantitative assessment of three-dimensional aortic growth and growth patterns in patients with thoracic aortic aneurysms undergoing CT surveillance. Published under a CC BY 4.0 license Online supplemental material is available for this article. See also the editorial by Wieben in this issue.

Removal notice to "VitSeg: Weakly supervised vitiligo segmentation in skin image" [Comput. Med. Imag. Graph. 85 (2020) 101779].

This article has been removed: please see Elsevier Policy on Article Withdrawal (https://www.elsevier.com/about/our-business/policies/article-withdrawal). This article has been removed at the request of the Authors and Editor-in-Chief because complete consent was not obtained by the authors in accordance with journal policy prior to publication. The authors and the journal sincerely apologize for this oversight.

Vascular Deformation Mapping of Abdominal Aortic Aneurysm.

Abdominal aortic aneurysm (AAA) is a complex disease that requires regular imaging surveillance to monitor for aneurysm stability. Current imaging surveillance techniques use maximum diameter, often assessed by computed tomography angiography (CTA), to assess risk of rupture and determine candidacy for operative repair. However, maximum diameter measurements can be variable, do not reliably predict rupture risk and future AAA growth, and may be an oversimplification of complex AAA anatomy. Vascular deformation mapping (VDM) is a recently described technique that uses deformable image registration to quantify three-dimensional changes in aortic wall geometry, which has been previously used to quantify three-dimensional (3D) growth in thoracic aortic aneurysms, but the feasibility of the VDM technique for measuring 3D growth in AAA has not yet been studied. Seven patients with infra-renal AAAs were identified and VDM was used to identify three-dimensional maps of AAA growth. In the present study, we demonstrate that VDM is able to successfully identify and quantify 3D growth (and the lack thereof) in AAAs that is not apparent from maximum diameter. Furthermore, VDM can be used to quantify growth of the excluded aneurysm sac after endovascular aneurysm repair (EVAR). VDM may be a useful adjunct for surgical planning and appears to be a sensitive modality for detecting regional growth of AAAs.

Removed: VitSeg: Weakly supervised vitiligo segmentation in skin image.

This article has been removed: please see Elsevier Policy on Article Withdrawal (https://www.elsevier.com/about/our-business/policies/article-withdrawal). This article has been removed at the request of the Authors and Editor-in-Chief because complete consent was not obtained by the authors in accordance with journal policy prior to publication. The authors and the journal sincerely apologize for this oversight.