Threshold-based Segmentation Method Research Articles

Land Survey and Mapping for Agricultural Purpose using UAV

Unmanned Aerial Vehicles (UAVs) are a group of aerial vehicles that include drones. Agriculture has always been the backbone of India’s sustained growth. Land surveying is the most challenging task in the agricultural field in terms of accuracy and time required for surveying using the traditional method. We have built a fully autonomous UAV and calibrated it for agricultural land survey purposes. The navigation and mapping process, in which a predefined path is provided to the UAV flight controller. During its autonomous flight, the drone's camera captures multiple images. Using a threshold-based segmentation method, the cultivated land and uncultivated land areas are determined using this system. Pre-processing of the images, land area segmentation, and post-processing of the images are carried out. The pre-processed image is divided into several distinct regions by applying a colour thresholding technique, which serves as the foundation for the segmentation of land. The results illustrate that the proposed method outperforms the existing method.

Vessel Density Features of Optical Coherence Tomography Angiography for Classification of Glaucoma Using Machine Learning.

Machine learning (ML) based on the optical coherence tomography angiography vessel density features with different thresholds using a support vector machine (SVM) model provides excellent performance for glaucoma detection. To assess the classification performance of ML based on the 4 vessel density features of peripapillary optical coherence tomography angiography for glaucoma detection. Images from 119 eyes of 119 glaucoma patients and 76 eyes of 76 healthy individuals were included. Four vessel density features of optical coherence tomography angiography images were developed using a threshold-based segmentation method and were integrated into 3 models of machine learning classifiers. Images were divided into 70% training set and 30% test set. Classification performances of SVM, random forest, and Gaussian Naive Bayes models were evaluated with the area under the receiver operating characteristic curve (AUC) and accuracy. Glaucoma eyes had lower vessel densities at different thresholds. For differentiating glaucoma eyes, the best results were achieved with 70% and 100% thresholds, in which SVM classifier discriminated glaucoma from healthy eyes with an AUC of 1 and accuracy of 1. After SVM, the random forest classifier with 100% thresholds showed an AUC of 0.993 and an accuracy of 0.994. Furthermore, the AUC of our ML performance (SVM) was 0.96 in a subgroup analysis of mild and moderate glaucoma eyes. ML based on the combined peripapillary vessel density features of total vessels and capillaries in the whole image and ring image could provide excellent performance for glaucoma detection.

Vessel Density Features of Optical Coherence Tomography Angiography for Classification of Optic Neuropathies Using Machine Learning.

To evaluate the classification performance of machine learning based on the 4 vessel density features of peripapillary optical coherence tomography angiography (OCT-A) for classifying healthy, nonarteritic anterior ischemic optic neuropathy (NAION), and optic neuritis (ON) eyes. Forty-five eyes of 45 NAION patients, 32 eyes of 32 ON patients, and 76 eyes of 76 healthy individuals with optic nerve head OCT-A were included. Four vessel density features of OCT-A images were developed using a threshold-based segmentation method and were integrated in 3 models of machine learning classifiers. Classification performances of support vector machine (SVM), random forest, and Gaussian Naive Bayes (GNB) models were evaluated with the area under the receiver-operating-characteristic curve (AUC) and accuracy. We divided 121 images into a 70% training set and 30% test set. For ON-NAION classification, best results were achieved with 50% threshold, in which 3 classifiers (SVM, RF, and GNB) discriminated ON from NAION with an AUC of 1 and accuracy of 1. For ON-Normal classification, with 100% threshold, SVM and RF classifiers were able to discriminate normal from ON with AUCs of 1 and accuracies of 1. For NAION-normal classification, with 50% threshold, the SVM and RF classified the NAION from normal with AUC and accuracy of 1. ML based on the combined peripapillary vessel density features of total vessels and capillaries in the whole image and ring image could provide excellent performance for NAION and ON distinction.

SIFT-GVF-based lung edge correction method for correcting the lung region in CT images.

Juxtapleural nodules were excluded from the segmented lung region in the Hounsfield unit threshold-based segmentation method. To re-include those regions in the lung region, a new approach was presented using scale-invariant feature transform and gradient vector flow models in this study. First, the scale-invariant feature transform method was utilized to detect all scale-invariant points in the binary lung region. The boundary points in the neighborhood of a scale-invariant point were collected to form the supportive boundary lines. Then, we utilized a Fourier descriptor to obtain a character representation of each supportive boundary line. Spectrum energy recognizes supportive boundaries that must be corrected. Third, the gradient vector flow-snake method was presented to correct the recognized supportive borders with a smooth profile curve, giving an ideal correction edge in those regions. Finally, the performance of the proposed method was evaluated through experiments on multiple authentic computed tomography images. The perfect results and robustness proved that the proposed method could correct the juxtapleural region precisely.

Improved Seismic Residual Diffracted Multiple Suppression Method Based on Object Detection and Image Segmentation

Seismic multiple is one of the most common noises in marine seismic data, which heavily affects subsequent processing and interpretation. To eliminate the influence of seismic multiples, many methods have been developed, while surface-related multiple elimination (SRME) is one of the most widely deployed methods. However, results of SRME always contain a few strong residual diffracted multiples (RDMs) in practice because of the unprecise prediction of diffracted multiples compared to reflection multiples. If we try to apply further multiple suppression methods to SRME results, it not only tends to damage the signals, but also spends lots of unnecessary computations where there is no RDM. In this article, we propose an improved RDM suppression method based on object detection and image segmentation. First, we employ an object detection network to locate bounding boxes containing RDMs in the SRME results. Then a threshold-based image segmentation method is utilized to identify regions of strong RDMs in the detected boxes. According to the segmentation results, parameters for weak multiples and strong multiples are provided for the adaptive multiple subtraction (AMS) in different regions to generate different results. At last, we combine the suppression results of strong RDMs and weak RDMs as the final results. Application on field data demonstrates that our method is able to suppress RDMs with little loss of signal.

A Novel Threshold-Based Segmentation Method for Quantification of COVID-19 Lung Abnormalities.

Since December 2019, the novel coronavirus disease 2019 (COVID-19) has claimed the lives of more than 3.75 million people worldwide. Consequently, methods for accurate COVID-19 diagnosis and classification are necessary to facilitate rapid patient care and terminate viral spread. Lung infection segmentations are useful to identify unique infection patterns that may support rapid diagnosis, severity assessment, and patient prognosis prediction, but manual segmentations are time-consuming and depend on radiologic expertise. Deep learning-based methods have been explored to reduce the burdens of segmentation; however, their accuracies are limited due to the lack of large, publicly available annotated datasets that are required to establish ground truths. For these reasons, we propose a semi-automatic, threshold-based segmentation method to generate region of interest (ROI) segmentations of infection visible on lung computed tomography (CT) scans. Infection masks are then used to calculate the percentage of lung abnormality (PLA) to determine COVID-19 severity and to analyze the disease progression in follow-up CTs. Compared with other COVID-19 ROI segmentation methods, on average, the proposed method achieved improved precision (47.49%) and specificity (98.40%) scores. Furthermore, the proposed method generated PLAs with a difference of pm 3.89% from the ground-truth PLAs. The improved ROI segmentation results suggest that the proposed method has potential to assist radiologists in assessing infection severity and analyzing disease progression in follow-up CTs.

View-Consistent Metal Segmentation in the Projection Domain for Metal Artifact Reduction in CBCT – An Investigation of Potential Improvement

The positive outcome of a trauma intervention depends on an intraoperative evaluation of inserted metallic implants. Due to occurring metal artifacts, the quality of this evaluation heavily depends on the performance of so-called Metal Artifact Reduction methods (MAR). The majority of these MAR methods require prior segmentation of the inserted metal objects. Therefore, typically a rather simple thresholding-based segmentation method in the reconstructed 3D volume is applied, despite some major disadvantages. With this publication, the potential of shifting the segmentation task to a learning-based, view-consistent 2D projection-based method on the downstream MAR's outcome is investigated. For segmenting the present metal, a rather simple learning-based 2D projection-wise segmentation network that is trained using real data acquired during cadaver studies, is examined. To overcome the disadvantages that come along with a 2D projection-wise segmentation, a Consistency Filter is proposed. The influence of the shifted segmentation domain is investigated by comparing the results of the standard fsMAR with a modified fsMAR version using the new segmentation masks. With a quantitative and qualitative evaluation on real cadaver data, the investigated approach showed an increased MAR performance and a high insensitivity against metal artifacts. For cases with metal outside the reconstruction's FoV or cases with vanishing metal, a significant reduction in artifacts could be shown. Thus, increases of up to roughly 3 dB w.r.t. the mean PSNR metric over all slices and up to 9 dB for single slices were achieved. The shown results reveal a beneficial influence of the shift to a 2D-based segmentation method on real data for downstream use with a MAR method, like the fsMAR. The nature of the method further suggests the same beneficial behavior for all (also recent data-driven) MAR methods, that for now comprise a 3D-volume-based segmentation step for subsequent inpainting.

Li-ion distribution and diffusion-induced stress calculations of particles using an image-based finite element method

The realistic particles in lithium batteries have unique microstructures, which should be determined in the analysis of chemo-mechanical behaviors during charge/discharge cycles. In this paper, an image-based finite element method (IBFEM) is developed to investigate the volume expansion, Li-ion distribution, and diffusion-induced stress (DIS) of a realistic particle. A threshold-based segmentation method is proposed to improve the segmentation accuracy based on the improved threshold of each phase. Based on the IBFEM, the active material phase, conductive additive and binder phase, and pore space are segmented and reconstructed using microstructural images. The volume expansion, Li-ion distribution, and DIS of an image-based SnO2 nanoparticle are simulated under galvanostatic/potentiostatic charging process. The predicted lithiation that induced the volume expansion of the SnO2 nanoparticle is approximately coincided with the experiment result observed by TEM. The realistic particle shape and size have a significant impact on the Li-ion distribution and DIS. The concentration profile is analogous to the boundary of the realistic particle at the early stage of lithiation and gradually disappears with further lithiation. The von Mises stress near a concave region is higher than that near a convex region. The maximum von Mises stress in the realistic particle with a larger equivalent radius is higher than that in the smaller particle with the same state of charge. By considering the comprehensive effect of particle shape and size, the size can be increased appropriately while guaranteeing the particle mechanical integrity. These results will contribute to understanding DIS evolution in a realistic particle and can be used to guide the selection of active particles for electrode preparation.

Evaluating Performance of Microwave Image Reconstruction Algorithms: Extracting Tissue Types with Segmentation Using Machine Learning.

Evaluating the quality of reconstructed images requires consistent approaches to extracting information and applying metrics. Partitioning medical images into tissue types permits the quantitative assessment of regions that contain a specific tissue. The assessment facilitates the evaluation of an imaging algorithm in terms of its ability to reconstruct the properties of various tissue types and identify anomalies. Microwave tomography is an imaging modality that is model-based and reconstructs an approximation of the actual internal spatial distribution of the dielectric properties of a breast over a reconstruction model consisting of discrete elements. The breast tissue types are characterized by their dielectric properties, so the complex permittivity profile that is reconstructed may be used to distinguish different tissue types. This manuscript presents a robust and flexible medical image segmentation technique to partition microwave breast images into tissue types in order to facilitate the evaluation of image quality. The approach combines an unsupervised machine learning method with statistical techniques. The key advantage for using the algorithm over other approaches, such as a threshold-based segmentation method, is that it supports this quantitative analysis without prior assumptions such as knowledge of the expected dielectric property values that characterize each tissue type. Moreover, it can be used for scenarios where there is a scarcity of data available for supervised learning. Microwave images are formed by solving an inverse scattering problem that is severely ill-posed, which has a significant impact on image quality. A number of strategies have been developed to alleviate the ill-posedness of the inverse scattering problem. The degree of success of each strategy varies, leading to reconstructions that have a wide range of image quality. A requirement for the segmentation technique is the ability to partition tissue types over a range of image qualities, which is demonstrated in the first part of the paper. The segmentation of images into regions of interest corresponding to various tissue types leads to the decomposition of the breast interior into disjoint tissue masks. An array of region and distance-based metrics are applied to compare masks extracted from reconstructed images and ground truth models. The quantitative results reveal the accuracy with which the geometric and dielectric properties are reconstructed. The incorporation of the segmentation that results in a framework that effectively furnishes the quantitative assessment of regions that contain a specific tissue is also demonstrated. The algorithm is applied to reconstructed microwave images derived from breasts with various densities and tissue distributions to demonstrate the flexibility of the algorithm and that it is not data-specific. The potential for using the algorithm to assist in diagnosis is exhibited with a tumor tracking example. This example also establishes the usefulness of the approach in evaluating the performance of the reconstruction algorithm in terms of its sensitivity and specificity to malignant tissue and its ability to accurately reconstruct malignant tissue.

Machine Learning for Microstructure Quantification of Different Material Classes

Abstract Material characterization is one of the major challenges faced in the field of materials research. The general approach is the assessment of quantitative properties, which are dependent on the utilization of destructive/non-destructive techniques. Conventional methods require the user to manually assess the obtained micrographs to identify the microstructural patterns followed by physical tests to quantify properties and characterization. A recent development in this area is the use of the concept of machine learning (ML) in image segmentation and analysis. Over the years, research in this area has resulted in the development of stable, robust and reliable systems, which yield consistently good results. This paper is aimed at introducing the use of one such machine learning approach based on Artificial Neural Networks (ANN) for image segmentation and quantification of material properties and discussion of some use cases. The results of the ML based method are compared with the results obtained from the traditional threshold based segmentation method.

Left Ventricular Global Function Analysis by Multi-Detector CT Threshold Based Three Dimension Segmentation Method in Patients with Coronary Artery Disease: A Comparative Study with Two Dimensional Echocardiography

AbstractBackground: Evaluation of global left ventricular function in patients with Coronary Artery Disease (CAD) is important for clinical diagnosis, risk assessment, therapeutic decisions, and prognosis. Because MDCT is a noninvasive imaging modality frequently used for the assessment of coronary artery disease, its ability to assess the LV volumes and EF has to be evaluated and compared with Two Dimensional Standard Echocardiography (2DSE).Aim of Work: To evaluate the role of MDCT threshold based segmentation method in quantification of global left ventricular function in patients with coronary artery disease and compare MDCT data with 2D-echocardiography as the standard of reference.Patients and Methods: Sixty patients with suspected coronary artery disease underwent contrast enhanced MDCT using retrospective gating. Ten phases of cardiac cycle were processed to assess end-systolic and end-diastolic phase at LV short-axis view. 2DSE was performed within two weeks before MDCT. Left Ventricular Ejection Fraction (LVEF), Left Ventricular End-Diastolic Volume (LVDV) and Left Ventricular End-Systolic Volume (LVSV) were calculated using the three dimensional threshold based segmentation method (in MDCT) and modified Simpson method (in echocar-diography).Results: LVEF, LVDV and LVSV were 51.9±10.9, 182.5± 45.2 and 89.8±40.2 and 51.6±10.5, 175.4±32.3 and 87.2±30.6 on MDCT and echocardiography respectively with excellent correlation between the two modalities (p;0.001) using Pear-son's correlationcoefficient. A Bland-Altman analysis showed that MDCT had slightly higher LVEF, LVDV and LVSV values with mean value of differences of 0.6%, 7ml and 2.5ml respectively, MDCT segmentation method had an accuracy of 91.6%, 95% and 98.3% respectively.Conclusion: MDCT threshold based 3D segmentation method is an accurate method for quantitative evaluation of global left ventricular function compared to 2D echocardiog-raphy.

Improving Brain Tumor Diagnosis Using MRI Segmentation Based on Collaboration of Beta Mixture Model and Learning Automata

In this paper, an automatic brain tumor diagnosis system is presented using a new threshold-based segmentation method. The proposed segmentation method is based on collaboration of beta mixture model and learning automata (LA). This segmentation method approximates histogram of a given MR image through mixture of Beta functions whose optimized parameters are determined through LA tool. Each Beta function shows one pixel class, and the threshold values which used for segmentation are obtained via intersection of two adjacent Beta functions. Feature extraction is based on statistical features. Moreover, support vector machine (SVM), K-nearest neighbor (KNN) and decision tree (DT) are applied as binary classifiers. In order to evaluate the performance, the proposed method is analyzed on set of 79 MR images from TCIA dataset and Harvard Medical School. The results of the experiments show that the proposed segmentation method presents superior average values for image similarity indices Dice and Jaccard. Moreover, the best accuracy of the presented brain tumor diagnosis system is obtained using SVM classifier with linear kernel, which is more than 98% in 10-fold cross-validation.

Quantitative Measurement of CSF in Patients with Spontaneous Intracranial Hypotension.

CSF hypovolemia is a core feature of spontaneous intracranial hypotension. Spontaneous intracranial hypotension is characterized by orthostatic headache and radiologic manifestations, including CSF along the neural sleeves, diffuse pachymeningeal enhancement, and/or venous engorgement. However, these characteristics are only qualitative. Quantifying intraspinal CSF volumes could improve spontaneous intracranial hypotension diagnosis and evaluation of hypovolemic statuses in patients with spontaneous intracranial hypotension. The purpose of this study was to compare intraspinal CSF volumes across spontaneous intracranial hypotension stages and to test the clinical applicability of these measures. A cohort of 23 patients with spontaneous intracranial hypotension and 32 healthy controls was subjected to brain MR imaging and MR myelography with 1.5T imaging. An automatic threshold-based segmentation method was used to calculate intraspinal CSF volumes at initial hospitalization (spontaneous intracranial hypotension-initial), partial improvement (spontaneous intracranial hypotension-intermediate), and complete recovery (spontaneous intracranial hypotension-recovery) stages. The mean intraspinal CSF volumes observed were the following: 95.31 mL for healthy controls, 72.31 mL for spontaneous intracranial hypotension-initial, 81.15 mL for spontaneous intracranial hypotension-intermediate, and 93.74 mL for spontaneous intracranial hypotension-recovery. Increased intraspinal CSF volumes were related to disease recovery (P < .001). The intraspinal CSF volumes of patients before complete recovery were significantly lower than those of healthy controls. With the estimated intradural CSF volumes as a reference, the intraspinal CSF volume percentage was lower in patients with spontaneous intracranial hypotension with venous engorgement than in those without it (P = .058). With a threshold-based segmentation method, we found that spinal CSF hypovolemia is fundamentally related to spontaneous intracranial hypotension. Intraspinal CSF volumes could be a sensitive parameter for the evaluation of treatment response and follow-up monitoring in patients with spontaneous intracranial hypotension.

Contrast-enhanced micro-CT imaging in murine carotid arteries: a new protocol for computing wall shear stress.

BackgroundWall shear stress (WSS) is involved in the pathophysiology of atherosclerosis. The correlation between WSS and atherosclerosis can be investigated over time using a WSS-manipulated atherosclerotic mouse model. To determine WSS in vivo, detailed 3D geometry of the vessel network is required. However, a protocol to reconstruct 3D murine vasculature using this animal model is lacking. In this project, we evaluated the adequacy of eXIA 160, a small animal contrast agent, for assessing murine vascular network on micro-CT. Also, a protocol was established for vessel geometry segmentation and WSS analysis.MethodsA tapering cast was placed around the right common carotid artery (RCCA) of ApoE−/− mice (n = 8). Contrast-enhanced micro-CT was performed using eXIA 160. An innovative local threshold-based segmentation procedure was implemented to reconstruct 3D geometry of the RCCA. The reconstructed RCCA was compared to the vessel geometry using a global threshold-based segmentation method. Computational fluid dynamics was applied to compute the velocity field and WSS distribution along the RCCA.ResultseXIA 160-enhanced micro-CT allowed clear visualization and assessment of the RCCA in all eight animals. No adverse biological effects were observed from the use of eXIA 160. Segmentation using local threshold values generated more accurate RCCA geometry than the global threshold-based approach. Mouse-specific velocity data and the RCCA geometry generated 3D WSS maps with high resolution, enabling quantitative analysis of WSS. In all animals, we observed low WSS upstream of the cast. Downstream of the cast, asymmetric WSS patterns were revealed with variation in size and location between animals.ConclusionseXIA 160 provided good contrast to reconstruct 3D vessel geometry and determine WSS patterns in the RCCA of the atherosclerotic mouse model. We established a novel local threshold-based segmentation protocol for RCCA reconstruction and WSS computation. The observed differences between animals indicate the necessity to use mouse-specific data for WSS analysis. For our future work, our protocol makes it possible to study in vivo WSS longitudinally over a growing plaque.

Lung surface deformation prediction from spirometry measurement and chest wall surface motion.

The authors have developed and evaluated a method to predict lung surface motion based on spirometry measurements, and chest and abdomen motion at selected locations. A patient-specific 3D triangular surface mesh of the lung region was obtained at the end expiratory phase by the threshold-based segmentation method. Lung flow volume changes were recorded with a spirometer for each patient. A total of 192 selected points at a regular spacing of 2 × 2 cm matrix points were used to detect chest wall motion over a total area of 32 × 24 cm covering the chest and abdomen surfaces. QR factorization with column pivoting was employed to remove redundant observations of the chest and abdominal areas. To create a statistical model between the lung surface and the corresponding surrogate signals, the authors developed a predictive model based on canonical ridge regression. Two unique weighting vectors were selected for each vertex on the lung surface; they were optimized during the training process using all other 4D-CT phases except for the test inspiration phase. These parameters were employed to predict the vertex locations of a testing data set. The position of each lung surface mesh vertex was estimated from the motion at selected positions within the chest wall surface and from spirometry measurements in ten lung cancer patients. The average estimation of the 98th error percentile for the end inspiration phase was less than 1 mm (AP = 0.9 mm, RL = 0.6 mm, and SI = 0.8 mm). The vertices located at the lower region of the lung had a larger estimation error as compared with those within the upper region of the lung. The average landmark motion errors, derived from the biomechanical modeling using real surface deformation vector fields (SDVFs), and the predicted SDVFs were 3.0 and 3.1 mm, respectively. Our newly developed predictive model provides a noninvasive approach to derive lung boundary conditions. The proposed system can be used with personalized biomechanical respiration modeling to derive lung tumor motion during radiation therapy from noninvasive measurements.

TH-CD-207A-05: Lung Surface Deformation Vector Fields Prediction by Monitoring Respiratory Surrogate Signals

Purpose: In this study, we developed and evaluated a method for predicting lung surface deformation vector fields (SDVFs) based on surrogate signals such as chest and abdomen motion at selected locations and spirometry measurements. Methods: A Patient-specific 3D triangular surface mesh of the lung region at end-expiration (EE) phase was obtained by threshold-based segmentation method. For each patient, a spirometer recorded the flow volume changes of the lungs; and 192 selected points at a regular spacing of 2cm X 2cm matrix points over a total area of 34cm X 24cm on the surface of chest and abdomen was used to detect chest wall motions. Preprocessing techniques such as QR factorization with column pivoting (QRCP) were employed to remove redundant observations of the chest and abdominal area. To create a statistical model between the lung surface and the corresponding surrogate signals, we developed a predictive model based on canonical ridge regression (CRR). Two unique weighting vectors were selected for each vertex on the surface of the lung, and they were optimized during the training process using the all other phases of 4D-CT except the end-inspiration (EI) phase. These parameters were employed to predict the vertices locations of a testing data set, which was the EI phase of 4D-CT. Results: For ten lung cancer patients, the deformation vector field of each vertex of lung surface mesh was estimated from the external motion at selected positions on the chest wall surface plus spirometry measurements. The average estimation of 98th percentile of error was less than 1 mm (AP= 0.85, RL= 0.61, and SI= 0.82). Conclusion: The developed predictive model provides a non-invasive approach to derive lung boundary condition. Together with personalized biomechanical respiration modelling, the proposed model can be used to derive the lung tumor motion during radiation therapy accurately from non-invasive measurements.

A fully automatic, threshold-based segmentation method for the estimation of the Metabolic Tumor Volume from PET images: validation on 3D printed anthropomorphic oncological lesions

18F-Fluorodeoxyglucose (18F-FDG) Positron Emission Tomography (PET) is a standard functional diagnostic technique to in vivo image cancer. Different quantitative paramters can be extracted from PET images and used as in vivo cancer biomarkers. Between PET biomarkers Metabolic Tumor Volume (MTV) has gained an important role in particular considering the development of patient-personalized radiotherapy treatment for non-homogeneous dose delivery. Different imaging processing methods have been developed to define MTV. The different proposed PET segmentation strategies were validated in ideal condition (e.g. in spherical objects with uniform radioactivity concentration), while the majority of cancer lesions doesn't fulfill these requirements. In this context, this work has a twofold objective: 1) to implement and optimize a fully automatic, threshold-based segmentation method for the estimation of MTV, feasible in clinical practice 2) to develop a strategy to obtain anthropomorphic phantoms, including non-spherical and non-uniform objects, miming realistic oncological patient conditions. The developed PET segmentation algorithm combines an automatic threshold-based algorithm for the definition of MTV and a k-means clustering algorithm for the estimation of the background. The method is based on parameters always available in clinical studies and was calibrated using NEMA IQ Phantom. Validation of the method was performed both in ideal (e.g. in spherical objects with uniform radioactivity concentration) and non-ideal (e.g. in non-spherical objects with a non-uniform radioactivity concentration) conditions. The strategy to obtain a phantom with synthetic realistic lesions (e.g. with irregular shape and a non-homogeneous uptake) consisted into the combined use of standard anthropomorphic phantoms commercially and irregular molds generated using 3D printer technology and filled with a radioactive chromatic alginate. The proposed segmentation algorithm was feasible in a clinical context and showed a good accuracy both in ideal and in realistic conditions.

River Channel Extraction From SAR Images by Combining Gray and Morphological Features

In remote sensing images acquired by the synthetic aperture radar (SAR), the backscattering coefficient is the function of the gray level. Since the gray of the pixels in the water area is relatively low, the gray level is commonly regarded as a valid feature for river channel identification and extraction. However, the gray threshold method is challenging mostly due to the speckles and false objects with similar gray value to the interest objects. Combining gray and morphological features, a novel multistage method for river channel extraction is proposed in this paper. First, the gray threshold-based image segmentation method is applied for initially removing the background noises. Next, a novel morphological model is proposed for identifying the river channel. After extracting the rough area of the river channel, the gray threshold-based image segmentation is reused for pruning results. Finally, the morphological filter is employed for correcting results. The benefit of the proposed method lies in its ability for properly combining the gray and morphological features, and the advantages of these features are fully exploited. The proposed method was evaluated both objectively and subjectively by utilizing a series of SAR images and quantified criterions. For the overall dataset, the missed detection rate, false alarm rate, and the correctness were statistically calculated, respectively. All experimental results proved that in comparision to the classic Otsu gray threshold method and its updated editions, the proposed approach is fast, robust, and effective for river channel extraction.

Semi-automatic region-of-interest segmentation based computer-aided diagnosis of mass lesions from dynamic contrast-enhanced magnetic resonance imaging based breast cancer screening.

Cancer screening with magnetic resonance imaging (MRI) is currently recommended for very high risk women. The high variability in the diagnostic accuracy of radiologists analyzing screening MRI examinations of the breast is due, at least in part, to the large amounts of data acquired. This has motivated substantial research towards the development of computer-aided diagnosis (CAD) systems for breast MRI which can assist in the diagnostic process by acting as a second reader of the examinations. This retrospective study was performed on 184 benign and 49 malignant lesions detected in a prospective MRI screening study of high risk women at Sunnybrook Health Sciences Centre. A method for performing semi-automatic lesion segmentation based on a supervised learning formulation was compared with the enhancement threshold based segmentation method in the context of a computer-aided diagnostic system. The results demonstrate that the proposed method can assist in providing increased separation between malignant and radiologically suspicious benign lesions. Separation between malignant and benign lesions based on margin measures improved from a receiver operating characteristic (ROC) curve area of 0.63 to 0.73 when the proposed segmentation method was compared with the enhancement threshold, representing a statistically significant improvement. Separation between malignant and benign lesions based on dynamic measures improved from a ROC curve area of 0.75 to 0.79 when the proposed segmentation method was compared to the enhancement threshold, also representing a statistically significant improvement. The proposed method has potential as a component of a computer-aided diagnostic system.

A dual model HU conversion from MRI intensity values within and outside of bone segment for MRI-based radiotherapy treatment planning of prostate cancer

The lack of electron density information in magnetic resonance images (MRI) poses a major challenge for MRI-based radiotherapy treatment planning (RTP). In this study the authors convert MRI intensity values into Hounsfield units (HUs) in the male pelvis and thus enable accurate MRI-based RTP for prostate cancer patients with varying tissue anatomy and body fat contents. T1/T2*-weighted MRI intensity values and standard computed tomography (CT) image HUs in the male pelvis were analyzed using image data of 10 prostate cancer patients. The collected data were utilized to generate a dual model HU conversion technique from MRI intensity values of the single image set separately within and outside of contoured pelvic bones. Within the bone segment local MRI intensity values were converted to HUs by applying a second-order polynomial model. This model was tuned for each patient by two patient-specific adjustments: MR signal normalization to correct shifts in absolute intensity level and application of a cutoff value to accurately represent low density bony tissue HUs. For soft tissues, such as fat and muscle, located outside of the bone contours, a threshold-based segmentation method without requirements for any patient-specific adjustments was introduced to convert MRI intensity values into HUs. The dual model HU conversion technique was implemented by constructing pseudo-CT images for 10 other prostate cancer patients. The feasibility of these images for RTP was evaluated by comparing HUs in the generated pseudo-CT images with those in standard CT images, and by determining deviations in MRI-based dose distributions compared to those in CT images with 7-field intensity modulated radiation therapy (IMRT) with the anisotropic analytical algorithm and 360° volumetric-modulated arc therapy (VMAT) with the Voxel Monte Carlo algorithm. The average HU differences between the constructed pseudo-CT images and standard CT images of each test patient ranged from -2 to 5 HUs and from 22 to 78 HUs in soft and bony tissues, respectively. The average local absolute value differences were 11 HUs in soft tissues and 99 HUs in bones. The planning target volume doses (volumes 95%, 50%, 5%) in the pseudo-CT images were within 0.8% compared to those in CT images in all of the 20 treatment plans. The average deviation was 0.3%. With all the test patients over 94% (IMRT) and 92% (VMAT) of dose points within body (lower than 10% of maximum dose suppressed) passed the 1 mm and 1% 2D gamma index criterion. The statistical tests (t- and F-tests) showed significantly improved (p ≤ 0.05) HU and dose calculation accuracies with the soft tissue conversion method instead of homogeneous representation of these tissues in MRI-based RTP images. This study indicates that it is possible to construct high quality pseudo-CT images by converting the intensity values of a single MRI series into HUs in the male pelvis, and to use these images for accurate MRI-based prostate RTP dose calculations.