Volumetric Segmentation Method Research Articles

Dense-sparse representation matters: A point-based method for volumetric medical image segmentation

Deep learning methods utilizing Convolutional Neural Networks (CNNs) and Transformers have achieved remarkable success in volumetric medical image analysis. While successful, the symmetrical structure of numerous networks pays insufficient attention to the encoding phase, and the large amount of memory occupied by voxels leads to unnecessary redundancy in the network. In this paper, we present a novel approach to handle volumetric medical images by converting them into point cloud and introduce a new asymmetrical segmentation architecture. We propose a dual-path encoder that fully captures both dense and sparse representations of the input point cloud sampled from volumes. Moreover, the two obtained representations are subtracted at the skip connection as a complementary feature during the decoding stage. Experimental results on the Brain Tumor Segmentation (BraTS) and the Multi-sequence Cardiac MR Segmentation tasks demonstrate the great potential of our point-based method for volumetric medical image segmentation.

A combined encoder–transformer–decoder network for volumetric segmentation of adrenal tumors

BackgroundThe morphology of the adrenal tumor and the clinical statistics of the adrenal tumor area are two crucial diagnostic and differential diagnostic features, indicating precise tumor segmentation is essential. Therefore, we build a CT image segmentation method based on an encoder–decoder structure combined with a Transformer for volumetric segmentation of adrenal tumors.MethodsThis study included a total of 182 patients with adrenal metastases, and an adrenal tumor volumetric segmentation method combining encoder–decoder structure and Transformer was constructed. The Dice Score coefficient (DSC), Hausdorff distance, Intersection over union (IOU), Average surface distance (ASD) and Mean average error (MAE) were calculated to evaluate the performance of the segmentation method.ResultsAnalyses were made among our proposed method and other CNN-based and transformer-based methods. The results showed excellent segmentation performance, with a mean DSC of 0.858, a mean Hausdorff distance of 10.996, a mean IOU of 0.814, a mean MAE of 0.0005, and a mean ASD of 0.509. The boxplot of all test samples' segmentation performance implies that the proposed method has the lowest skewness and the highest average prediction performance.ConclusionsOur proposed method can directly generate 3D lesion maps and showed excellent segmentation performance. The comparison of segmentation metrics and visualization results showed that our proposed method performed very well in the segmentation.

A statistical deformation model-based data augmentation method for volumetric medical image segmentation

The accurate delineation of organs-at-risk (OARs) is a crucial step in treatment planning during radiotherapy, as it minimizes the potential adverse effects of radiation on surrounding healthy organs. However, manual contouring of OARs in computed tomography (CT) images is labor-intensive and susceptible to errors, particularly for low-contrast soft tissue. Deep learning-based artificial intelligence algorithms surpass traditional methods but require large datasets. Obtaining annotated medical images is both time-consuming and expensive, hindering the collection of extensive training sets. To enhance the performance of medical image segmentation, augmentation strategies such as rotation and Gaussian smoothing are employed during preprocessing. However, these conventional data augmentation techniques cannot generate more realistic deformations, limiting improvements in accuracy. To address this issue, this study introduces a statistical deformation model-based data augmentation method for volumetric medical image segmentation. By applying diverse and realistic data augmentation to CT images from a limited patient cohort, our method significantly improves the fully automated segmentation of OARs across various body parts. We evaluate our framework on three datasets containing tumor OARs from the head, neck, chest, and abdomen. Test results demonstrate that the proposed method achieves state-of-the-art performance in numerous OARs segmentation challenges. This innovative approach holds considerable potential as a powerful tool for various medical imaging-related sub-fields, effectively addressing the challenge of limited data access.

Quantitative 3D Investigation of Nanoparticle Assemblies by Volumetric Segmentation of Electron Tomography Data Sets

Morphological characterization of nanoparticle assemblies and hybrid nanomaterials is critical in determining their structure-property relationships as well as in the development of structures with desired properties. Electron tomography has become a widely utilized technique for the three-dimensional characterization of nanoparticle assemblies. However, the extraction of quantitative morphological parameters from the reconstructed volume can be a complex and labor-intensive task. In this study, we aim to overcome this challenge by automating the volumetric segmentation process applied to three-dimensional reconstructions of nanoparticle assemblies. The key to enabling automated characterization is to assess the performance of different volumetric segmentation methods in accurately extracting predefined quantitative descriptors for morphological characterization. In our methodology, we compare the quantitative descriptors obtained through manual segmentation with those obtained through automated segmentation methods, to evaluate their accuracy and effectiveness. To show generality, our study focuses on the characterization of assemblies of CdSe/CdS quantum dots, gold nanospheres and CdSe/CdS encapsulated in polymeric micelles, and silica-coated gold nanorods decorated with both CdSe/CdS or PbS quantum dots. We use two unsupervised segmentation algorithms: the watershed transform and the spherical Hough transform. Our results demonstrate that the choice of automated segmentation method is crucial for accurately extracting the predefined quantitative descriptors. Specifically, the spherical Hough transform exhibits superior performance in accurately extracting quantitative descriptors, such as particle size and interparticle distance, thereby allowing for an objective, efficient, and reliable volumetric segmentation of complex nanoparticle assemblies.

An unsupervised machine learning algorithm for in-situ defect-detection in laser powder-bed fusion

This research studies the development of a machine-learning algorithm to detect porosity induced during laser powder bed fusion (LPBF) due to the lack of fusion (LoF) phenomenon. The detection algorithm is based on analyzing in-situ light intensity data emitted from the melt pool during LPBF. A Self-Organizing Map (SOM), an unsupervised machine learning algorithm, is thoroughly customized to classify disturbances in the light intensity signal, where the clustered disturbances are mapped with the geometrical feature and position of defects. The defects' position and size are correlated with the actual geometrical feature and position of defects identified through a post-processing computed tomography (CT) scanning. To this end, two sets of samples are designed, including 1) samples with intentional micro-voids to mimic the lack of fusion in the printed parts for assessing the sensor response and calibrating/optimizing the SOM algorithm, and 2) samples with randomized pores induced normally during the process to evaluate the proposed SOM algorithm at different process parameters. Along with the SOM approach, a volumetric segmentation method and confusion matrix are incorporated into the SOM-based algorithm to examine the sensitivity (true positive rates) and specificity (true negative rates) of defect prediction. The comparison between the outcome of the prediction algorithm and experimental CT data indicates that the sensitivity and specificity are from 61 % to 94 % and 69 % to 93 %, respectively, where the prediction percentage is highly dependent on the process parameters. In comparison with conventional prediction algorithms (e.g., Absolute Limits), the proposed SOM algorithm has a higher prediction rate when it is faster and can rapidly identify the location of defects that may open up an opportunity to develop intermittent controllers for LPBF.

Volumetric quantification of choroid and Haller's sublayer using OCT scans: An accurate and unified approach based on stratified smoothing.

The choroid, a dense vascular structure in the posterior segment of the eye, maintains the health of the retina by supplying oxygen and nutrients, and assumes clinical significance in screening ocular diseases including age-related macular degeneration (AMD) and central serous chorioretinopathy (CSCR). As a technological assist, algorithmic estimation of choroidal biomarkers has been suggested based on sectional (B-scan) optical coherence tomography (OCT) images. However, most such 2D estimation techniques are compute-intensive, yet enjoy limited accuracy and have only been validated on OCT image datasets of healthy eyes. Not surprisingly, fine-scale analyses, including those involving Haller's sublayer, remain relatively rare and unsophisticated. Against this backdrop, we propose an efficient algorithm to quantify desired biomarkers with improved accuracy based on volume OCT scans. Specifically, we attempted an accurate, computationally light volumetric segmentation method involving stratified smoothing to detect choroid and Haller's sublayer. For detecting the various boundaries of the choroid and the Haller's sublayer, we propose a common volumetric method that performs suitable exponential enhancement and maintains smooth spatial continuity across 2D B-scans. Further, we achieve suitable volumetric smoothing by primarily deploying light-duty linear regression, and sparingly using compute-intensive tensor voting, and hence significantly reduce overall complexity. The proposed methodology is tested on five health and five diseased OCT volumes considering various metrics including volumetric Dice coefficient and corresponding quotient measures to facilitate comparison vis-à-vis intra-observer repeatability. On five healthy and five diseased OCT volumes, respectively, the proposed method for choroid segmentation recorded volumetric Dice coefficients of 93.53 % and 93.30 %, which closely approximate the respective reference observer repeatability values of 95.60 % and 95.49 %. In terms of related quotient measures, our method achieved more than 50 % improvement over a recently reported method. In detecting Haller's sublayer as well, our algorithm records statistical performance closely matching that of reference manual method. Advancing the state-of-the-art, the proposed volumetric segmentation, tested on both healthy and diseased datasets, demonstrated close match with the manual reference. Our method assumes significance in accurate screening of chorioretinal diseases including AMD, CSCR and pachychoroid. Further, it enables generating accurate training data for developing deep learning models for improved detection of choroid and Haller's sublayer.

Three-Dimensional Vessel Segmentation in Whole-Tissue and Whole-Block Imaging Using a Deep Neural Network: Proof-of-Concept Study

In the field of pathology, micro-computed tomography (micro-CT) has become an attractive imaging modality because it enables full analysis of the three-dimensional characteristics of a tissue sample or organ in a noninvasive manner. However, because of the complexity of the three-dimensional information, understanding would be improved by development of analytical methods and software such as those implemented for clinical CT. As the accurate identification of tissue components is critical for this purpose, we have developed a deep neural network (DNN) to analyze whole-tissue images (WTIs) and whole-block images (WBIs) of neoplastic cancer tissue using micro-CT. The aim of this study was to segment vessels from WTIs and WBIs in a volumetric segmentation method using DNN. To accelerate the segmentation process while retaining accuracy, a convolutional block in DNN was improved by introducing a residual inception block. Three colorectal tissue samples were collected and one WTI and 70 WBIs were acquired by a micro-CT scanner. The implemented segmentation method was then tested on the WTI and WBIs. As a proof-of-concept study, our method successfully segmented the vessels on all WTI and WBIs of the colorectal tissue sample. In addition, despite the large size of the images for analysis, all segmentation processes were completed in 10 minutes.

Correction

Correction

CT-based radiomics for prediction of histologic subtype and metastatic disease in primary malignant lung neoplasms.

As some of the most important factors for treatment decision of lung cancer (which is the deadliest neoplasm) are staging and histology, this work aimed to associate quantitative contrast-enhanced computed tomography (CT) features from malignant lung tumors with distant and nodal metastases (according to clinical TNM staging) and histopathology (according to biopsy and surgical resection) using radiomics assessment. A local cohort of 85 patients were retrospectively (2010-2017) analyzed after approval by the institutional research review board. CT images acquired with the same protocol were semiautomatically segmented by a volumetric segmentation method. Tumors were characterized by quantitative CT features of shape, first-order, second-order, and higher-order textures. Statistical and machine learning analyses assessed the features individually and combined with clinical data. Univariate and multivariate analyses identified 40, 2003, and 45 quantitative features associated with distant metastasis, nodal metastasis, and histopathology (adenocarcinoma and squamous cell carcinoma), respectively. A machine learning model yielded the highest areas under the receiver operating characteristic curves of 0.92, 0.84, and 0.88 to predict the same previous patterns. Several radiomic features (including wavelet energies, information measures of correlation and maximum probability from co-occurrence matrix, busyness from neighborhood intensity-difference matrix, directionalities from Tamura's texture, and fractal dimension estimation) significantly associated with distant metastasis, nodal metastasis, and histology were discovered in this work, presenting great potential as imaging biomarkers for pathological diagnosis and target therapy decision.

Optimized High Resolution 3D Dense-U-Net Network for Brain and Spine Segmentation

The 3D image segmentation is the process of partitioning a digital 3D volumes into multiple segments. This paper presents a fully automatic method for high resolution 3D volumetric segmentation of medical image data using modern supervised deep learning approach. We introduce 3D Dense-U-Net neural network architecture implementing densely connected layers. It has been optimized for graphic process unit accelerated high resolution image processing on currently available hardware (Nvidia GTX 1080ti). The method has been evaluated on MRI brain 3D volumetric dataset and CT thoracic scan dataset for spine segmentation. In contrast with many previous methods, our approach is capable of precise segmentation of the input image data in the original resolution, without any pre-processing of the input image. It can process image data in 3D and has achieved accuracy of 99.72% on MRI brain dataset, which outperformed results achieved by human expert. On lumbar and thoracic vertebrae CT dataset it has achieved the accuracy of 99.80%. The architecture proposed in this paper can also be easily applied to any task already using U-Net network as a segmentation algorithm to enhance its results. Complete source code was released online under open-source license.

Volumetric gray matter measures of amygdala and accumbens in childhood overweight/obesity.

ObjectivesNeuroimaging data suggest that pediatric overweight and obesity are associated with morphological alterations in gray matter (GM) brain structures, but previous studies using mainly voxel-based morphometry (VBM) showed inconsistent results. Here, we aimed to examine the relationship between youth obesity and the volume of predefined reward system structures using magnetic resonance (MR) volumetry. We also aimed to complement volumetry with VBM-style analysis.MethodsFifty-one Caucasian young subjects (32 females; mean age: 13.8±1.9, range: 10.2–16.5 years) were included. Subjects were selected from a subsample of the I.Family study examined in the Hungarian center. A T1-weighted 1 mm3 isotropic resolution image was acquired. Age- and sex-standardized body mass index (zBMI) was assessed at the day of MRI and ~1.89 years (mean±SD: 689±188 days) before the examination. Obesity related GM alterations were investigated using MR volumetry in five predefined brain structures presumed to play crucial roles in body weight regulation (hippocampus, amygdala, accumbens, caudate, putamen), as well as whole-brain and regional VBM.ResultsThe volumes of accumbens and amygdala showed significant positive correlations with zBMI, while their GM densities were inversely related to zBMI. Voxel-based GM mass also showed significant negative correlation with zBMI when investigated in the predefined amygdala region, but this relationship was mediated by GM density.ConclusionsOverweight/obesity related morphometric brain differences already seem to be present in children/adolescents. Our work highlights the disparity between volume and VBM-derived measures and that GM mass (combination of volume and density) is not informative in the context of obesity related volumetric changes. To better characterize the association between childhood obesity and GM morphometry, a combination of volumetric segmentation and VBM methods, as well as future longitudinal studies are necessary. Our results suggest that childhood obesity is associated with enlarged structural volumes, but decreased GM density in the reward system.

Testing associations between cannabis use and subcortical volumes in two large population-based samples.

Disentangling the putative impact of cannabis on brain morphology from other comorbid substance use is critical. After controlling for the effects of nicotine, alcohol and multi-substance use, this study aimed to determine whether frequent cannabis use is associated with significantly smaller subcortical grey matter volumes. Exploratory analyses using mixed linear models, one per region of interest (ROI), were performed whereby individual differences in volume (outcome) at seven subcortical ROIs were regressed onto cannabis and comorbid substance use (predictors). Two large population-based twin samples from the United States and Australia. A total of 622 young Australian adults [66% female; μage =25.9, standard deviation SD)=3.6] and 474 middle-aged US males (μage =56.1SD=2.6 ) of predominately Anglo-Saxon ancestry with complete substance use and imaging data. Subjects with a history of stroke or traumatic brain injury were excluded. Magnetic resonance imaging (MRI) and volumetric segmentation methods were used to estimate volume in seven subcortical ROIs: thalamus, caudate nucleus, putamen, pallidum, hippocampus, amygdala and nucleus accumbens. Substance use measurements included maximum nicotine and alcohol use, total life-time multi-substance use, maximum cannabis use in the young adults and regular cannabis use in the middle-aged males. After correcting for multiple testing (P=0.007), cannabis use was unrelated to any subcortical ROI. However, maximum nicotine use was associated with significantly smaller thalamus volumes in middle-aged males. In exploratory analyses based on young adult and middle-aged samples, normal variation in cannabis use is unrelated statistically to individual differences in brain morphology as measured by subcortical volume.

An Automated Approach for Kidney Segmentation in Three-Dimensional Ultrasound Images.

Automated segmentation of kidneys in three-dimensional (3-D) abdominal ultrasound volumes is a task of paramount importance in automated diagnosis of abdominal trauma. However, ultrasound speckle noise, low-contrast boundaries, partial kidney occlusion, and probe misalignment restrict the utility of the solution, especially when it is used in emergency rooms and Focused Assessment with Sonography Trauma applications. This paper introduces a systematic and cost-effective method capable of detecting and segmenting the kidney's shape in acquired 3-D ultrasound volumes, using off-line training datasets. This paper offers a new shape model representation, called the complex-valued implicit shape model, to generate a 3-D kidney shape model by combining prior knowledge of training shapes and anatomical knowledge. We apply shape-to-volume registration, based on a new similarity metric, to detect the kidney shape by fitting the 3-D shape model on 3-D ultrasound volumes. Upon kidney detection, the fitted shape model is used to initialize and evolve a new level-set function, called complex-valued rational level-set with shape prior, to segment the kidney's shape. Experimentation using both simulated and actual ultrasound volumes indicate that the proposed solution provides a better performance over the state-of-the-art volumetric ultrasound segmentation methods.

A computerized volumetric segmentation method applicable to multi-centre MRI data to support computer-aided breast tissue analysis, density assessment and lesion localization

Density assessment and lesion localization in breast MRI require accurate segmentation of breast tissues. A fast, computerized algorithm for volumetric breast segmentation, suitable for multi-centre data, has been developed, employing 3D bias-corrected fuzzy c-means clustering and morphological operations. The full breast extent is determined on T1-weighted images without prior information concerning breast anatomy. Left and right breasts are identified separately using automatic detection of the midsternum. Statistical analysis of breast volumes from eighty-two women scanned in a UK multi-centre study of MRI screening shows that the segmentation algorithm performs well when compared with manually corrected segmentation, with high relative overlap (RO), high true-positive volume fraction (TPVF) and low false-positive volume fraction (FPVF), and has an overall performance of RO 0.94 ± 0.05, TPVF 0.97 ± 0.03 and FPVF 0.04 ± 0.06, respectively (training: 0.93 ± 0.05, 0.97 ± 0.03 and 0.04 ± 0.06; test: 0.94 ± 0.05, 0.98 ± 0.02 and 0.05 ± 0.07).

Automated volumetric segmentation of retinal fluid on optical coherence tomography.

We propose a novel automated volumetric segmentation method to detect and quantify retinal fluid on optical coherence tomography (OCT). The fuzzy level set method was introduced for identifying the boundaries of fluid filled regions on B-scans (x and y-axes) and C-scans (z-axis). The boundaries identified from three types of scans were combined to generate a comprehensive volumetric segmentation of retinal fluid. Then, artefactual fluid regions were removed using morphological characteristics and by identifying vascular shadowing with OCT angiography obtained from the same scan. The accuracy of retinal fluid detection and quantification was evaluated on 10 eyes with diabetic macular edema. Automated segmentation had good agreement with manual segmentation qualitatively and quantitatively. The fluid map can be integrated with OCT angiogram for intuitive clinical evaluation.

Patch-based label fusion segmentation of brainstem structures with dual-contrast MRI for Parkinson's disease.

Parkinson's disease (PD) is a neurodegenerative disorder that impairs the motor functions. Both surgical treatment and study of PD require delineation of basal ganglia nuclei morphology. While many automatic volumetric segmentation methods have been proposed for the lentiform nucleus, few have attempted to identify the key brainstem substructures including the subthalamic nucleus (STN), substantia nigra (SN), and red nucleus (RN) due to their small size and poor contrast in conventional T1W MRI. A dual-contrast patch-based label fusion method was developed to segment the SN, STN, and RN using multivariate cross-correlation. Two different MRI contrasts (T2*w and phase) are produced from a multi-contrast multi-echo FLASH MRI sequence, enabling visualization of these nuclei. T1-T2* fusion MRI was used to resolve the issue of poor nuclei (i.e., the STN, SN, and RN) contrast on T1w MRI, and to mitigate susceptibility artifacts that may hinder accurate nonlinear registration on T2*w MRI. Unbiased group-wise registration was used for anatomical normalization between the atlas library and the target subject. The performance of the proposed method was compared with a state-of-the-art single-contrast label fusion technique. The proposed method outperformed a state-of-the-art single-contrast patch-based method in segmenting the STN, RN and SN, and the results were better than those reported in previous literature. Our dual-contrast patch-based label fusion method was superior to a single-contrast method for segmenting brainstem nuclei using a multi-contrast multi-echo FLASH MRI sequence. The method is promising for the treatment and research of Parkinson's disease. This method can be extended for multiple alternative image contrasts and other fields of applications.

Generic fitted shapes (GFS): Volumetric object segmentation in service robotics

In this paper, a simultaneous 3D volumetric segmentation and reconstruction method, based on the so-called Generic Fitted Shapes (GFS) is proposed. The aim of this work is to cope with the lack of volumetric information encountered in visually controlled mobile manipulation systems equipped with stereo or RGB-D cameras. Instead of using primitive volumes, such as cuboids or cylinders, for approximating objects in point clouds, their volumetric structure has been estimated based on fitted generic shapes. The proposed GFSs can capture the shapes of a broad range of object classes without the need of large a-priori shape databases. The fitting algorithm, which aims at determining the particular geometry of each object of interest, is based on a modified version of the active contours approach extended to the 3D Cartesian space. The proposed volumetric segmentation system produces comprehensive closed object surfaces which can be further used in mobile manipulation scenarios. Within the experimental setup, the proposed technique has been evaluated against two state-of-the-art methods, namely superquadrics and 3D Object Retrieval (3DOR) engines.

SEGMENTATION OF THE COMPLETE SUPERIOR CEREBELLAR PEDUNCLES USING A MULTI-OBJECT GEOMETRIC DEFORMABLE MODEL.

The superior cerebellar peduncles (SCPs) are white matter tracts that serve as the major efferent pathways from the cerebellum to the thalamus. With diffusion tensor images (DTI), tractography algorithms or volumetric segmentation methods have been able to reconstruct part of the SCPs. However, when the fibers cross, the primary eigenvector (PEV) no longer represents the primary diffusion direction. Therefore, at the crossing of the left and right SCP, known as the decussation of the SCPs (dSCP), fiber tracts propagate incorrectly. To our knowledge, previous methods have not been able to segment the SCPs correctly. In this work, we explore the diffusion properties and seek to volumetrically segment the complete SCPs. The non-crossing SCPs and dSCP are modeled as different objects. A multi-object geometric deformable model is employed to define the boundaries of each piece of the SCPs, with the forces derived from diffusion properties as well as the PEV. We tested our method on a software phantom and real subjects. Results indicate that our method is able to the resolve the crossing and segment the complete SCPs with repeatability.

Towards a universal error measure in connectomics using the variation of information

Towards a universal error measure in connectomics using the variation of information

Assessment of global cardiac I-123 MIBG uptake and washout using volumetric quantification of SPECT acquisitions

BackgroundAssessment of cardiac innervation using single-photon emission computer tomography (SPECT) is less established than planar imaging, but may be more suitable for quantification. Therefore, a volumetric quantification of I-123 MIBG SPECT acquisitions was performed. Reproducibility, the effects of extra cardiac I-123 MIBG uptake and the relation with conventional planar indices were evaluated. Methods54 patients referred for planar and SPECT I-123 MIBG acquisitions were included. Ellipsoidal or box-shaped volumes of interest were placed on the left ventricle, cardiac lumen, mediastinum, lung and liver. SPECT segmentation was performed twice in all patients. Indices were determined based on the heart-to-mediastinum (HM), myocardial wall-to-mediastinum and myocardial wall-to-lumen regions. HM ratios and washout rates were also determined based on anterior planar images. ResultsCardiac count densities were highly reproducible (CV 1.5-5.4, ICC 0.96-0.99) and inter-rater variability was low (CV 1.8-6.8, ICC 0.94-0.99). Mediastinal uptake was an important explanatory variable of uptake in the entire heart (early R2 = 0.36; delayed R2 =0.43) and myocardial wall (early R2 = 0.28; delayed R2 = 0.37). Lung washout was an explanatory variable of organ washout of the heart (heart R2 = 0.38; myocardial wall R2 = 0.33). In general, SPECT indices showed moderate-to-good correlations with the planar uptake (PCC 0.497-0.851). ConclusionBy applying a volumetric segmentation method we were able to segment the heart in all patients. SPECT I-123 MIBG quantification was found to be highly reproducible and had a moderate to good correlation with the planar indices.