New methods for the geometrical analysis of tubular organs

Skeleton-guided multi-scale dual-coordinate attention aggregation network for retinal blood vessel segmentation

Deep learning plays a pivotal role in retinal blood vessel segmentation for medical diagnosis. Despite their significant efficacy, these techniques face two major challenges. Firstly, they often neglect the severe class imbalance in fundus images, where thin vessels in the foreground are proportionally minimal. Secondly, they are susceptible to poor image quality and blurred vessel edges, resulting in discontinuities or breaks in vascular structures. In response, this paper proposes the Skeleton-guided Multi-scale Dual-coordinate Attention Aggregation (SMDAA) network for retinal vessel segmentation. SMDAA comprises three innovative modules: Dual-coordinate Attention (DCA), Unbalanced Pixel Amplifier (UPA), and Vessel Skeleton Guidance (VSG). DCA, integrating Multi-scale Coordinate Feature Aggregation (MCFA) and Scale Coordinate Attention Decoding (SCAD), meticulously analyzes vessel structures across various scales, adept at capturing intricate details, thereby significantly enhancing segmentation accuracy. To address class imbalance, we introduce UPA, dynamically allocating more attention to misclassified pixels, ensuring precise extraction of thin and small blood vessels. Moreover, to preserve vessel structure continuity, we integrate vessel anatomy and develop the VSG module to connect fragmented vessel segments. Additionally, a Feature-level Contrast (FCL) loss is introduced to capture subtle nuances within the same category, enhancing the fidelity of retinal blood vessel segmentation. Extensive experiments on three public datasets (DRIVE, STARE, and CHASE_DB1) demonstrate superior performance in comparison to current methods. The code is available at https://github.com/wangwxr/SMDAA_NET.

3D airway geometry analysis of factors in airway navigation failure for lung nodules

BACKGROUND

This study aimed to quantitatively reveal contributing factors to airway navigation failure during radial probe endobronchial ultrasound (R-EBUS) by using geometric analysis in a three-dimensional (3D) space and to investigate the clinical feasibility of prediction models for airway navigation failure.

METHODS

We retrospectively reviewed patients who underwent R-EBUS between January 2017 and December 2018. Geometric quantification was analyzed using in-house software built with open-source python libraries including the Vascular Modeling Toolkit ( http://www.vmtk.org ), simple insight toolkit ( https://sitk.org ), and sci-kit image ( https://scikit-image.org ). We used a machine learning-based approach to explore the utility of these significant factors.

RESULTS

Of the 491 patients who were eligible for analysis (mean age, 65 years +/- 11 [standard deviation]; 274 men), the target lesion was reached in 434 and was not reached in 57. Twenty-seven patients in the failure group were matched with 27 patients in the success group based on propensity scores. Bifurcation angle at the target branch, the least diameter of the last section, and the curvature of the last section are the most significant and stable factors for airway navigation failure. The support vector machine can predict airway navigation failure with an average area under the curve of 0.803.

CONCLUSIONS

Geometric analysis in 3D space revealed that a large bifurcation angle and a narrow and tortuous structure of the closest bronchus from the lesion are associated with airway navigation failure during R-EBUS. The models developed using quantitative computer tomography scan imaging show the potential to predict airway navigation failure.

Predicting cerebrovascular age and its clinical relevance: Modeling using 3D morphological features of brain vessels

Aging manifests as many phenotypes, among which age-related changes in brain vessels are important, but underexplored. Thus, in the present study, we constructed a model to predict age using cerebrovascular morphological features, further assessing their clinical relevance using a novel pipeline. Age prediction models were first developed using data from a normal cohort (n = 1181), after which their relevance was tested in two stroke cohorts (n = 564 and n = 455). Our novel pipeline adapted an existing framework to compute generic vessel features for brain vessels, resulting in 126 morphological features. We further built various machine learning models to predict age using only clinical factors, only brain vessel features, and a combination of both. We further assessed deviation from healthy aging using the age gap and explored its clinical relevance by correlating the predicted age and age gap with various risk factors. The models constructed using only brain vessel features and those combining clinical factors with vessel features were better predictors of age than the clinical factor-only model (r = 0.37, 0.48, and 0.26, respectively). Predicted age was associated with many known clinical factors, and the associations were stronger for the age gap in the normal cohort. The age gap was also associated with important factors in the pooled cohort atherosclerotic cardiovascular disease risk score and white matter hyperintensity measurements. Cerebrovascular age, computed using the morphological features of brain vessels, could serve as a potential individualized marker for the early detection of various cerebrovascular diseases.

Landmark-Assisted Anatomy-Sensitive Retinal Vessel Segmentation Network

Automatic retinal vessel segmentation is important for assisting clinicians in diagnosing ophthalmic diseases. The existing deep learning methods remain constrained in instance connectivity and thin vessel detection. To this end, we propose a novel anatomy-sensitive retinal vessel segmentation framework to preserve instance connectivity and improve the segmentation accuracy of thin vessels. This framework uses TransUNet as its backbone and utilizes self-supervised extracted landmarks to guide network learning. TransUNet is designed to simultaneously benefit from the advantages of convolutional and multi-head attention mechanisms in extracting local features and modeling global dependencies. In particular, we introduce contrastive learning-based self-supervised extraction anatomical landmarks to guide the model to focus on learning the morphological information of retinal vessels. We evaluated the proposed method on three public datasets: DRIVE, CHASE-DB1, and STARE. Our method demonstrates promising results on the DRIVE and CHASE-DB1 datasets, outperforming state-of-the-art methods by improving the F1 scores by 0.36% and 0.31%, respectively. On the STARE dataset, our method achieves results close to the best-performing methods. Visualizations of the results highlight the potential of our method in maintaining topological continuity and identifying thin blood vessels. Furthermore, we conducted a series of ablation experiments to validate the effectiveness of each module in our model and considered the impact of image resolution on the results.

Graph Convolution Based Cross-Network Multiscale Feature Fusion for Deep Vessel Segmentation

Vessel segmentation is widely used to help with vascular disease diagnosis. Vessels reconstructed using existing methods are often not sufficiently accurate to meet clinical use standards. This is because 3D vessel structures are highly complicated and exhibit unique characteristics, including sparsity and anisotropy. In this paper, we propose a novel hybrid deep neural network for vessel segmentation. Our network consists of two cascaded subnetworks performing initial and refined segmentation respectively. The second subnetwork further has two tightly coupled components, a traditional CNN-based U-Net and a graph U-Net. Cross-network multi-scale feature fusion is performed between these two U-shaped networks to effectively support high-quality vessel segmentation. The entire cascaded network can be trained from end to end. The graph in the second subnetwork is constructed according to a vessel probability map as well as appearance and semantic similarities in the original CT volume. To tackle the challenges caused by the sparsity and anisotropy of vessels, a higher percentage of graph nodes are distributed in areas that potentially contain vessels while a higher percentage of edges follow the orientation of potential nearby vessels. Extensive experiments demonstrate our deep network achieves state-of-the-art 3D vessel segmentation performance on multiple public and in-house datasets.

Precision MRI phenotyping enables detection of small changes in body composition for longitudinal cohorts

Longitudinal studies provide unique insights into the impact of environmental factors and lifespan issues on health and disease. Here we investigate changes in body composition in 3088 free-living participants, part of the UK Biobank in-depth imaging study. All participants underwent neck-to-knee MRI scans at the first imaging visit and after approximately two years (second imaging visit). Image-derived phenotypes for each participant were extracted using a fully-automated image processing pipeline, including volumes of several tissues and organs: liver, pancreas, spleen, kidneys, total skeletal muscle, iliopsoas muscle, visceral adipose tissue (VAT), abdominal subcutaneous adipose tissue, as well as fat and iron content in liver, pancreas and spleen. Overall, no significant changes were observed in BMI, body weight, or waist circumference over the scanning interval, despite some large individual changes. A significant decrease in grip strength was observed, coupled to small, but statistically significant, decrease in all skeletal muscle measurements. Significant increases in VAT and intermuscular fat in the thighs were also detected in the absence of changes in BMI, waist circumference and ectopic-fat deposition. Adjusting for disease status at the first imaging visit did not have an additional impact on the changes observed. In summary, we show that even after a relatively short period of time significant changes in body composition can take place, probably reflecting the obesogenic environment currently inhabited by most of the general population in the United Kingdom.

Overview of radiomics in prostate imaging and future directions

Recent advancements in imaging technology and analysis methods have led to an analytic framework known as radiomics. This framework extracts comprehensive high-dimensional features from imaging data and performs data mining to build analytical models for improved decision-support. Its features include many categories spanning texture and shape; thus, it can provide abundant information for precision medicine. Many studies of prostate radiomics have shown promising results in the assessment of pathological features, prediction of treatment response, and stratification of risk groups. Herein, we aimed to provide a general overview of radiomics procedures, discuss technical issues, explain various clinical applications, and suggest future research directions, especially for prostate imaging.

Automatic Aortic Dissection Centerline Extraction Via Morphology-Guided CRN Tracker

Aortic dissection (AD) centerline extraction has important clinical value in the quantitative diagnosis and treatment of AD disease. However, AD centerline extraction is a difficult task and quantitative evaluation is rarely studied. In this work, we propose a fully automatic algorithm to extract AD centerline based on a convolutional regression network (CRN) and the morphological properties of AD. To this end, we first design a topological model to describe the complex topology of AD. With this model, CRNs are trained to estimate the position, tangential vector, and scale of the centerline. The tracking accuracy is further improved by centerline continuity and a gradient-based penalty function. In addition, seed points are extracted on the basis of random regression and line clustering to ensure automated vessel tracking. The proposed method has been evaluated on an AD database and a public aortic database, and achieved high overlapping ratios of 0.9610 and 1.0000, respectively. The tracked centerline is very close to the ground truth and shows good stability, with low average distance errors of 1.4720 mm and 1.8748 mm, respectively.

Shape-to-graph mapping method for efficient characterization and classification of complex geometries in biological images

With the ever-increasing quality and quantity of imaging data in biomedical research comes the demand for computational methodologies that enable efficient and reliable automated extraction of the quantitative information contained within these images. One of the challenges in providing such methodology is the need for tailoring algorithms to the specifics of the data, limiting their areas of application. Here we present a broadly applicable approach to quantification and classification of complex shapes and patterns in biological or other multi-component formations. This approach integrates the mapping of all shape boundaries within an image onto a global information-rich graph and machine learning on the multidimensional measures of the graph. We demonstrated the power of this method by (1) extracting subtle structural differences from visually indistinguishable images in our phenotype rescue experiments using the endothelial tube formations assay, (2) training the algorithm to identify biophysical parameters underlying the formation of different multicellular networks in our simulation model of collective cell behavior, and (3) analyzing the response of U2OS cell cultures to a broad array of small molecule perturbations.

In this paper, we present a methodology that is based on mapping an arbitrary set of outlines onto a complete, strictly defined structure, in which every point representing the shape becomes a terminal point of a global graph. Because this mapping preserves the whole complexity of the shape, it allows for extracting the full scope of geometric features of any scale. Importantly, an extensive set of graph-based metrics in each image makes integration with machine learning routines highly efficient even for a small data sets and provide an opportunity to backtrack the subtle morphological features responsible for the automated distinction into image classes. The resulting tool provides efficient, versatile, and robust quantification of complex shapes and patterns in experimental images.

Marginal radiomics features as imaging biomarkers for pathological invasion in lung adenocarcinoma

OBJECTIVES

Lung adenocarcinomas which manifest as ground-glass nodules (GGNs) have different degrees of pathological invasion and differentiating among them is critical for treatment. Our goal was to evaluate the addition of marginal features to a baseline radiomics model on computed tomography (CT) images to predict the degree of pathologic invasiveness.

METHODS

We identified 236 patients from two cohorts (training, n = 189; validation, n = 47) who underwent surgery for GGNs. All GGNs were pathologically confirmed as adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA), or invasive adenocarcinoma (IA). The regions of interest were semi-automatically annotated and 40 radiomics features were computed. We selected features using L1-norm regularization to build the baseline radiomics model. Additional marginal features were developed using the cumulative distribution function (CDF) of intratumoral intensities. An improved model was built combining the baseline model with CDF features. Three classifiers were tested for both models.

RESULTS

The baseline radiomics model included five features and resulted in an average area under the curve (AUC) of 0.8419 (training) and 0.9142 (validation) for the three classifiers. The second model, with the additional marginal features, resulted in AUCs of 0.8560 (training) and 0.9581 (validation). All three classifiers performed better with the added features. The support vector machine showed the most performance improvement (AUC improvement = 0.0790) and the best performance was achieved by the logistic classifier (validation AUC = 0.9825).

CONCLUSION

Our novel marginal features, when combined with a baseline radiomics model, can help differentiate IA from AIS and MIA on preoperative CT scans.

KEY POINTS

• Our novel marginal features could improve the existing radiomics model to predict the degree of pathologic invasiveness in lung adenocarcinoma.

Reproducibility of an airway tapering measurement in computed tomography with application to bronchiectasis

We propose a pipeline to acquire a scalar tapering measurement from the carina to the most distal point of an individual airway visible on computed tomography (CT). We show the applicability of using tapering measurements on clinically acquired data by quantifying the reproducibility of the tapering measure. We generate a spline from the centerline of an airway to measure the area and arclength at contiguous intervals. The tapering measurement is the gradient of the linear regression between area in log space and arclength. The reproducibility of the measure was assessed by analyzing different radiation doses, voxel sizes, and reconstruction kernel on single timepoint and longitudinal CT scans and by evaluating the effect of airway bifurcations. Using 74 airways from 10 CT scans, we show a statistical difference, p = 3.4 × 10 - 4 , in tapering between healthy airways ( n = 35 ) and those affected by bronchiectasis ( n = 39 ). The difference between the mean of the two populations is 0.011    mm - 1 , and the difference between the medians of the two populations was 0.006    mm - 1 . The tapering measurement retained a 95% confidence interval of ± 0.005    mm - 1 in a simulated 25 mAs scan and retained a 95% confidence of ± 0.005    mm - 1 on simulated CTs up to 1.5 times the original voxel size. We have established an estimate of the precision of the tapering measurement and estimated the effect on precision of the simulated voxel size and CT scan dose. We recommend that the scanner calibration be undertaken with the phantoms as described, on the specific CT scanner, radiation dose, and reconstruction algorithm that are to be used in any quantitative studies.