Vascular branching geometry relating to portal hypertension: a study of liver microvasculature in cirrhotic rats by X-ray phase-contrast computed tomography

Improving Vessel Segmentation with Multi-Task Learning and Auxiliary Data Available Only During Model Training

Liver vessel segmentation in magnetic resonance imaging data is important for the computational analysis of vascular remodeling, associated with a wide spectrum of diffuse liver diseases. Existing approaches rely on contrast enhanced imaging data, but the necessary dedicated imaging sequences are not uniformly acquired. Images without contrast enhancement are acquired more frequently, but vessel segmentation is challenging, and requires large-scale annotated data. We propose a multi-task learning framework to segment vessels in liver MRI without contrast. It exploits auxiliary contrast enhanced MRI data available only during training to reduce the need for annotated training examples. Our approach draws on paired native and contrast enhanced data with and without vessel annotations for model training. Results show that auxiliary data improves the accuracy of vessel segmentation, even if they are not available during inference. The advantage is most pronounced if only few annotations are available for training, since the feature representation benefits from the shared task structure. A validation of this approach to augment a model for brain tumor segmentation confirms its benefits across different domains. An auxiliary informative imaging modality can augment expert annotations even if it is only available during training.

In silico simulation of hepatic arteries: An open-source algorithm for efficient synthetic data generation

BACKGROUND

In silico testing of novel image reconstruction and quantitative algorithms designed for interventional imaging requires realistic high-resolution modeling of arterial trees with contrast dynamics. Furthermore, data synthesis for training of deep learning algorithms requires that an arterial tree generation algorithm be computationally efficient and sufficiently random.

PURPOSE

The purpose of this paper is to provide a method for anatomically and physiologically motivated, computationally efficient, random hepatic arterial tree generation.

METHODS

The vessel generation algorithm uses a constrained constructive optimization approach with a volume minimization-based cost function. The optimization is constrained by the Couinaud liver classification system to assure a main feeding artery to each Couinaud segment. An intersection check is included to guarantee non-intersecting vasculature and cubic polynomial fits are used to optimize bifurcation angles and to generate smoothly curved segments. Furthermore, an approach to simulate contrast dynamics and respiratory and cardiac motion is also presented.

RESULTS

The proposed algorithm can generate a synthetic hepatic arterial tree with 40 000 branches in 11 s. The high-resolution arterial trees have realistic morphological features such as branching angles (MAD with Murray's law= 1.2 ± 1 . 2 o $ = \;1.2 \pm {1.2^o}$ ), radii (median Murray deviation= 0.08 $ = \;0.08$ ), and smoothly curved, non-intersecting vessels. Furthermore, the algorithm assures a main feeding artery to each Couinaud segment and is random (variability = 0.98 ± 0.01).

CONCLUSIONS

This method facilitates the generation of large datasets of high-resolution, unique hepatic angiograms for the training of deep learning algorithms and initial testing of novel 3D reconstruction and quantitative algorithms designed for interventional imaging.

Quantitative Biomarkers Derived from a Novel Contrast-Free Ultrasound High-Definition Microvessel Imaging for Distinguishing Thyroid Nodules

Low specificity in current ultrasound modalities for thyroid cancer detection necessitates the development of new imaging modalities for optimal characterization of thyroid nodules. Herein, the quantitative biomarkers of a new high-definition microvessel imaging (HDMI) were evaluated for discrimination of benign from malignant thyroid nodules. Without the help of contrast agents, this new ultrasound-based quantitative technique utilizes processing methods including clutter filtering, denoising, vessel enhancement filtering, morphological filtering, and vessel segmentation to resolve tumor microvessels at size scales of a few hundred microns and enables the extraction of vessel morphological features as new tumor biomarkers. We evaluated quantitative HDMI on 92 patients with 92 thyroid nodules identified in ultrasound. A total of 12 biomarkers derived from vessel morphological parameters were associated with pathology results. Using the Wilcoxon rank-sum test, six of the twelve biomarkers were significantly different in distribution between the malignant and benign nodules (all p < 0.01). A support vector machine (SVM)-based classification model was trained on these six biomarkers, and the receiver operating characteristic curve (ROC) showed an area under the curve (AUC) of 0.9005 (95% CI: [0.8279,0.9732]) with sensitivity, specificity, and accuracy of 0.7778, 0.9474, and 0.8929, respectively. When additional clinical data, namely TI-RADS, age, and nodule size were added to the features, model performance reached an AUC of 0.9044 (95% CI: [0.8331,0.9757]) with sensitivity, specificity, and accuracy of 0.8750, 0.8235, and 0.8400, respectively. Our findings suggest that tumor vessel morphological features may improve the characterization of thyroid nodules.

Diagnosis of Clinically Significant Portal Hypertension Using CT- and MRI-based Vascular Model

Background Currently, the hepatic venous pressure gradient (HVPG) remains the reference standard for diagnosis of clinically significant portal hypertension (CSPH) but is limited by its invasiveness and availability. Purpose To investigate a vascular geometric model for noninvasive diagnosis of CSPH (HVPG ≥10 mm Hg) in patients with liver cirrhosis for both contrast-enhanced CT and MRI. Materials and Methods In this retrospective study, consecutive patients with liver cirrhosis who underwent HVPG measurement from August 2016 to April 2019 were included. Patients without hepatic diseases were included and marked as non-CSPH to balance the ratio of CSPH 1:1. A variety of vascular parameters were extracted from the portal vein, hepatic vein, aorta, and inferior vena cava and then entered into a vascular geometric model for identification of CSPH. Diagnostic performance was assessed with the area under the receiver operating characteristic curve (AUC). Results The model was developed and tested with retrospective data from 250 patients with liver cirrhosis and 273 patients without clinical evidence of hepatic disease at contrast-enhanced CT examination, including 213 patients with CSPH (mean age, 49 years ± 12 [SD]; 138 women) and 310 patients without CSPH (mean age, 50 years ± 9; 177 women). For external validation, an MRI data set with 224 patients with cirrhosis (mean age, 49 years ± 10; 158 women) and a CT data set with 106 patients with cirrhosis (mean age, 53 years ± 12; 71 women) were analyzed. Significant reductions in mean whole-vessel volumes were observed in the portal vein (ranging from 36.9 cm³ ± 16.0 to 29.6 cm³ ± 11.1; P < .05) and hepatic vein (ranging from 35.3 cm³ ± 21.5 to 22.4 cm³ ± 15.7; P < .05) when CSPH occurred. Similarly, the mean whole-vessel lengths were shorter in patients with CSPH (portal vein: 1.7 m ± 1.2 vs 3.0 m ± 2.4, P < .05; hepatic vein: 0.9 m ± 1.5 vs 1.8 m ± 1.5, P < .05) than in those without CSPH. The proposed vascular model performed well in the internal test set (mean AUC, 0.90 ± 0.02) and external test sets (mean AUCs, 0.84 ± 0.12 and 0.87 ± 0.11). Conclusion A contrast-enhanced CT- and MRI-based vascular model was proposed with good diagnostic consistency for hepatic venous pressure gradient measurement. ClinicalTrials.gov registration nos. NCT03138915 and NCT03766880 © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Roldán-Alzate and Reeder in this issue.

Insight into microvascular adaptive alterations in the Glisson system of biliary atresia after Kasai portoenterostomy using X-ray phase-contrast CT

OBJECTIVES

To investigate microvascular alterations in the Glisson system of biliary atresia (BA) patients after Kasai portoenterostomy (KP) using three-dimensional (3D) virtual histopathology based on X-ray phase-contrast CT (PCCT).

METHODS

Liver explants from BA patients were imaged using PCCT, and 32 subjects were included and divided into two groups: KP (n = 16) and non-KP (n = 16). Combined with histological analysis and 3D visualization technology, 3D virtual histopathological assessment of the biliary, arterial, and portal venous systems was performed. According to loop volume ratio, 3D spatial density, relative surface area, tortuosity, and other parameters, pathological changes of microvasculature in the Glisson system were investigated.

RESULTS

In the non-KP group, bile ducts mostly manifested as radial multifurcated hyperplasia and twisted into loops. In the KP group, the bile duct hyperplasia was less, and the loop volume ratio of bile ducts decreased by 13.89%. Simultaneously, the arterial and portal venous systems presented adaptive alterations in response to degrees of bile duct hyperplasia. Compared with the non-KP group, the 3D spatial density of arteries in the KP group decreased by 3.53%, and the relative surface area decreased from 0.088 ± 0.035 to 0.039 ± 0.015 (p < .01). Deformed portal branches gradually recovered after KP, with a 2.93% increase in 3D spatial density and a decrease in tortuosity from 1.17 ± 0.06 to 1.14 ± 0.04 (p < .01) compared to the non-KP group.

CONCLUSION

3D virtual histopathology via PCCT clearly reveals the microvascular structures in the Glisson system of BA patients and provides key insights into the morphological mechanism of microvascular adaptation induced by biliary tract dredging after KP in BA disease.

KEY POINTS

• 3D virtual histopathology via X-ray phase-contrast computed tomography clearly presented the morphological structures and pathological changes of microvasculature in the Glisson system of biliary atresia patients. • The morphological alterations of microvasculature in the Glisson system followed the competitive occupancy mechanism in the process of biliary atresia.

Ultrasound high-definition microvasculature imaging with novel quantitative biomarkers improves breast cancer detection accuracy

OBJECTIVES

To overcome the limitations of power Doppler in imaging angiogenesis, we sought to develop and investigate new quantitative biomarkers of a contrast-free ultrasound microvasculature imaging technique for differentiation of benign from malignant pathologies of breast lesion.

METHODS

In this prospective study, a new high-definition microvasculature imaging (HDMI) was tested on 521 patients with 527 ultrasound-identified suspicious breast masses indicated for biopsy. Four new morphological features of tumor microvessels, microvessel fractal dimension (mvFD), Murray's deviation (MD), bifurcation angle (BA), and spatial vascularity pattern (SVP) as well as initial biomarkers were extracted and analyzed, and the results correlated with pathology. Multivariable logistic regression analysis was used to study the performance of different prediction models, initial biomarkers, new biomarkers, and combined new and initial biomarkers in differentiating benign from malignant lesions.

RESULTS

The new HDMI biomarkers, mvFD, BA, MD, and SVP, were statistically significantly different in malignant and benign lesions, regardless of tumor size. Sensitivity and specificity of the new biomarkers in lesions > 20 mm were 95.6% and 100%, respectively. Combining the new and initial biomarkers together showed an AUC, sensitivity, and specificity of 97% (95% CI: 95-98%), 93.8%, and 89.2%, respectively, for all lesions regardless of mass size. The classification was further improved by adding the Breast Imaging Reporting and Data System (BI-RADS) score to the prediction model, showing an AUC, sensitivity, and specificity of 97% (95% CI: 95-98%), 93.8%, and 89.2%, respectively.

CONCLUSION

The addition of new quantitative HDMI biomarkers significantly improved the accuracy in breast lesion characterization when used as a complementary imaging tool to the conventional ultrasound.

KEY POINTS

• Novel quantitative biomarkers extracted from tumor microvessel images increase the sensitivity and specificity in discriminating malignant from benign breast masses. • New HDMI biomarkers Murray's deviation, bifurcation angles, microvessel fractal dimension, and spatial vascularity pattern outperformed the initial biomarkers. • The addition of BI-RADS scores based on US descriptors to the multivariable analysis using all biomarkers remarkably increased the sensitivity, specificity, and AUC in all size groups.

Use of the volume-averaged Murray's deviation method for the characterization of branching geometry in liver fibrosis: a preliminary study on vascular circulation

Background

Vascular changes in liver fibrosis can result in increased intrahepatic vascular resistance and impaired blood circulation. This can hinder the recovery from fibrosis and may eventually lead to portal hypertension, a major cirrhosis complication. This report proposed a volume-averaged Murray's deviation method to characterize intrahepatic circulation in the liver during fibrosis and its subsequent regression via X-ray phase-contrast computed tomography (PCCT).

Methods

Liver fibrosis was induced in 24 Sprague-Dawley rats by exposure to carbon tetrachloride (CCl₄) for up to 10 weeks, after which, spontaneous regression commenced and continued until week 30. High-resolution three-dimensional (3D) imaging of the livers was performed with PCCT. The values of Murray's deviation based on the volume-averaged and the conventional diameter-based methods were compared. After that, the intrahepatic circulation at different stages of fibrosis was evaluated using the volume-averaged method. The increase in collagen during liver fibrosis was assessed by pathological analyses.

Results

A comparison of the 2 methods showed that with an increase in the number of diameter measurements, the value of Murrary's deviation obtained using the diameter-based method gradually approaches those of the volume-averaged method, with minimal variations. The value of Murray's deviation increased with the development of fibrosis. After reversal, the value rapidly decreased and approached that of the normal state in both the main branches (1.05±0.17, 1.17±0.21, 1.34±0.18, and 1.17±0.19 in the normal, moderate, severe, and regressive groups, respectively; P<0.05 between the severe group and other groups) and the small branches (1.05±0.09, 1.42±0.48, 1.79±0.57, and 1.18±0.28 in the normal, moderate, severe, and regressive group, respectively; P<0.05 between adjacent groups). An analysis of Murray's deviation and the pathological results showed that the vascular circulation in this disease model was consistent with the progression and recovery from fibrosis.

Conclusions

This study showed the validity of the volume-averaged method for calculating Murray's deviation and demonstrated that it could accurately evaluate the blood circulation state of the liver during fibrosis and its subsequent regression. Thus, the volume-averaged method of calculating Murray's deviation may be an objective and valuable staging criterion to evaluate intrahepatic circulation during liver fibrosis.

Subcutaneous Low-Density Foreign Bodies Detection via Grating-Based Multimodal X-ray Imaging

Detecting low-density foreign bodies within soft tissues still stands for a serious challenge. Grating-based multimodal X-ray imaging typically has low hardware requirements while simultaneously providing three kinds of imaging information, i.e., absorption, phase-contrast, and dark-field. We aimed to explore the capacity of grating-based multimodal X-ray imaging technology for detecting common foreign bodies within subcutaneous tissues, and to assess the advantages as well as disadvantages of the three kinds of images obtained via grating-based X-ray multimodal technology in relation to diverse kinds of foreign bodies within different tissues. In this study, metal, glass, wood, plastic, graphite, and ceramic foreign bodies were injected into chunks of the pig adipose tissue and chicken thigh muscles. Next, a grating-based multimodal X-ray imaging device developed in our laboratory was used to detect the above foreign bodies within the adipose and muscle tissues. Our results show that grating-based multimodal X-ray imaging clearly revealed the subcutaneous foreign bodies within the adipose and muscle tissues by acquiring complementary absorption, phase-contrast, and dark-field imaging data in a single shot. Grating-based multimodal X-ray imaging has an exciting potential to detect foreign bodies underneath the epidermis.

Quantitative Biomarkers for Cancer Detection Using Contrast-Free Ultrasound High-Definition Microvessel Imaging: Fractal Dimension, Murray's Deviation, Bifurcation Angle & Spatial Vascularity Pattern

A growing body of evidence indicates that there is a strong correlation between microvascular morphological features and malignant tumors. Therefore, quantification of these features might allow more accurate differentiation of benign and malignant tumors. The main objective of this research project is to improve the quantification of microvascular networks depicted in contrast-free ultrasound microvessel images. To achieve this goal, a new series of quantitative microvessel morphological parameters are introduced for differentiation of breast masses using contrast-free ultrasound-based high-definition microvessel imaging (HDMI). Using HDMI, we quantified and analyzed four new parameters: 1) microvessel fractal dimension (mvFD), a marker of tumor microvascular complexity; 2) Murray's deviation (MD), the diameter mismatch, defined as the deviation from Murray's law; 3) bifurcation angle (BA), abnormally decreased angle; and 4) spatial vascular pattern (SVP), indicating tumor vascular distribution pattern, either intratumoral or peritumoral. The new biomarkers have been tested on 60 patients with breast masses. Validation of the feature's extraction algorithm was performed using a synthetic data set. All the proposed parameters had the power to discriminate the breast lesion malignancy (p < 0.05), displaying BA as the most sensitive test, with a sensitivity of 90.6%, and mvFD as the most specific test, with a specificity of 92%. The results of all four new biomarkers showed an AUC = 0.889, sensitivity of 80% and specificity of 91.4% In conclusion, the added value of the proposed quantitative morphological parameters, as new biomarkers of angiogenesis within breast masses, paves the way for more accurate breast cancer detection with higher specificity.