Intensity-based Algorithms Research Articles

Empirical Evaluation of Coding and Inter Pixel Redundancy in still Image Compression

The main aim of this research work is to compress grayscale images efficiently using prediction and intensity-based image compression algorithms. Image compression is useful for removing the duplication in an image to store and transmit the data in an efficient form. This research work analyzes four new schemes for gray scale lossy image compression. Among the four schemes considered, two compressive approaches are designed for Prediction Based Image Compression (PBIC) level implementation. Third approach is designed for Intensity Based Image Compression (IBIC). Finally, the previously designed PBIC and IBIC schemes lead to an Integrated Encoder. All the considered method performances are analyzed using the performance metrics. These results are compared with JPEG 2000 which is a extensively used benchmark compression encoder. The outcome of all the proposed methods is also compared with modern encoders.

Unleashing the power of optical attenuation coefficients to facilitate segmentation strategies in OCT imaging of age-related macular degeneration: perspective.

The use of optical attenuation coefficients (OAC) in optical coherence tomography (OCT) imaging of the retina has improved the segmentation of anatomic layers compared with traditional intensity-based algorithms. Optical attenuation correction has improved our ability to measure the choroidal thickness and choroidal vascularity index using dense volume scans. Algorithms that combine conventional intensity-based segmentation with depth-resolved OAC OCT imaging have been used to detect elevations of the retinal pigment epithelium (RPE) due to drusen and basal laminar deposits, the location of hyperpigmentation within the retina and along the RPE, the identification of macular atrophy, the thickness of the outer retinal (photoreceptor) layer, and the presence of calcified drusen. OAC OCT algorithms can identify the risk-factors that predict disease progression in age-related macular degeneration.

Automated scoring of nematode nictation on a textured background.

Entomopathogenic nematodes, including Steinernema spp., play an increasingly important role as biological alternatives to chemical pesticides. The infective juveniles of these worms use nictation-a behavior in which animals stand on their tails-as a host-seeking strategy. The developmentally-equivalent dauer larvae of the free-living nematode Caenorhabditis elegans also nictate, but as a means of phoresy or "hitching a ride" to a new food source. Advanced genetic and experimental tools have been developed for C. elegans, but time-consuming manual scoring of nictation slows efforts to understand this behavior, and the textured substrates required for nictation can frustrate traditional machine vision segmentation algorithms. Here we present a Mask R-CNN-based tracker capable of segmenting C. elegans dauers and S. carpocapsae infective juveniles on a textured background suitable for nictation, and a machine learning pipeline that scores nictation behavior. We use our system to show that the nictation propensity of C. elegans from high-density liquid cultures largely mirrors their development into dauers, and to quantify nictation in S. carpocapsae infective juveniles in the presence of a potential host. This system is an improvement upon existing intensity-based tracking algorithms and human scoring which can facilitate large-scale studies of nictation and potentially other nematode behaviors.

Improved FAST algorithm for non-uniform rotational distortion correction in OCT endoscopic imaging.

Optical Coherence Tomography (OCT) is widely used for endoscopic imaging in endoluminal organs because of its high imaging accuracy and resolution. However, OCT endoscopic imaging suffers from Non-Uniform Rotational Distortion (NURD), which can be caused by many factors, such as irregular motor rotation and changes in friction between the probe and the sheath. Correcting this distortion is essential to obtaining high-quality Optical Coherence Tomography Angiography (OCTA) images. There are two main approaches for correcting NURD: hardware-based methods and algorithm-based methods. Hardware-based methods can be costly, challenging to implement, and may not eliminate NURD. Algorithm-based methods, such as image registration, can be effective for correcting NURD but can also be prone to the problem of NURD propagation. To address this issue, we process frames by coarse and fine registration, respectively. The new reference frame is generated by filtering out the A-scan that may have the NURD problem by coarse registration. And the fine registration uses this frame to achieve the final NURD correction. In addition, we have improved the Features from Accelerated Segment Test (FAST) algorithm and put it into coarse and fine registration process. Four evaluation functions were used for the experimental results, including signal-to-noise ratio (SNR), peak signal-to-noise ratio (PSNR), mean squared error (MSE), and structural similarity index measure (SSIM). By comparing with Scale-invariant feature transform (SIFT), Speeded up robust features (SURF), Oriented FAST and Rotated BRIEF (ORB), intensity-based (Cross-correlation), and Optical Flow algorithms, our algorithm has a higher similarity between the corrected frames. Moreover, the noise in the OCTA data is better suppressed, and the vascular information is well preserved. Our image registration-based algorithm reduces the problem of NURD propagation between B-scan frames and improves the imaging quality of OCT endoscopic images.

Interfractional dose accumulation for MR-guided liver SBRT: Variation among algorithms is highly patient- and fraction-dependent

Background and purposeDaily plan adaptations could take the dose delivered in previous fractions into account. Due to high dose delivered per fraction, low number of fractions, steep dose gradients, and large interfractional organ deformations, this might be particularly important for liver SBRT. This study investigates inter-algorithm variation of interfractional dose accumulation for MR-guided liver SBRT. Materials and methodsWe assessed 27 consecutive MR-guided liver SBRT treatments of 67.5 Gy in three (n = 15) or 50 Gy in five fractions (n = 12), both prescribed to the GTV. We calculated fraction doses on daily patient anatomy, warped these doses to the simulation MRI using seven different algorithms, and accumulated the warped doses. Thus, we obtained differences in planned doses and warped or accumulated doses for each algorithm. This enabled us to calculate the inter-algorithm variations in warped doses per fraction and in accumulated doses per treatment course. ResultsThe four intensity-based algorithms were more consistent with planned PTV dose than affine or contour-based algorithms. The mean (range) variation of the dose difference for PTV D95% due to dose warping by these intensity-based algorithms was 10.4 percentage points (0.3 to 43.7) between fractions and 8.6 (0.3 to 24.9) between accumulated treatment doses. As seen by these ranges, the variation was very dependent on the patient and the fraction being analyzed. Nevertheless, no correlations between patient or plan characteristics on the one hand and inter-algorithm dose warping variation on the other hand was found. ConclusionInter-algorithm dose accumulation variation is highly patient- and fraction-dependent for MR-guided liver SBRT. We advise against trusting a single algorithm for dose accumulation in liver SBRT.

Geometric Analysis and Dosimetric Evaluation of Various Deformable Image Registration (DIR) Algorithms for Abdominal Images with 3D Biomechanically Modelled Deformation

<h3>Purpose/Objective(s)</h3> The accuracy of DIR algorithms available on commercial radiotherapy planning platforms were examined and their clinical applicability geometrically and dosimetrically for abdominal sites were evaluated using biomechanically modelled deformation. <h3>Materials/Methods</h3> 5 pancreatic patients' planning CT were used in this study. Following the guideline of AAPM TG-132, a patient-specific QA workflow was developed and implemented to evaluate DIR for abdomen using digital phantoms, created with a commercial virtual phantom software (VPS). First, the planning CT images were deformed in superior-inferior direction to simulation respiratory motion in lung and abdomen, using the biomechanical model embedded in VPS. In addition, 5 mm motion in both anterior-posterior and lateral directions were added to region of interest (ROI), including stomach and duodenum, and small bowel. The added ROI deformations are consistent with the abdomen data observed in daily CBCT. The resulting 3D deformation vector field (DVF) and the corresponding deformed CT was then obtained, serving as ground truth. Second, the original planning CT and the deformed CT were imported to two commercial DIR systems (DS1 and DS2), to perform DIR with intensity-based algorithms, and derive DVFs. Quantitative differences between ground truth DVF and DS1 or DS2's DVFs, such as mean, max, and the distribution of DIR errors were tabulated. Then the original plan was re-calculated on the deformed CT, which was generated based on DS1 or DS2's DVF. The dose errors were then calculated on a pixel-to-pixel basis. The relationship between the DIR errors and the dose error ranges (25%-75% on the dose error distribution histogram) were investigated. As a pilot study, comprehensive reports were generated to show the DIR performances as patient specific QA records. <h3>Results</h3> The displacements on the ground truth DVF range from 0 to 37 mm. Geometric analysis showed DIR errors as tabulated below. Dosimetrically, as the DIR error increases, the dose error range also increase, for both DS1 and DS2. Dosimetric evaluation of the dose errors for ROIs were tabulated as below. <h3>Conclusion</h3> Preliminary evaluation of geometric and dosimetric accuracy of DIR on DS1 and DS2 were performed, with a realistic biomechanical model derived DVF serving as ground truth. A patient specific DIR QA workflow was developed for a pancreatic patient cohort, and it has potential to be implemented as a clinical workflow.

A WEIGHTED COHERENCE ESTIMATOR FOR COHERENT CHANGE DETECTION IN SYNTHETIC APERTURE RADAR IMAGES

Abstract. Synthetic aperture radar (SAR) coherent change detection (CCD) often utilizes the degree of coherence to detect changes that have occurred between two data collections. Although having shown some performances in change detection, many existing coherence estimators are still relatively limited because the change areas do not stand out well from all decorrelation areas due to the low cluster-to-noise ratio (CNR) and volume scattering. Moreover, many estimators require the equal-variance assumption between two SAR images of the same scene. However, the assumption is less likely to be met in regions of significant differences in intensity, such as the change areas. To address these problems, we proposed an improved coherence estimator that introduces the parameters about the true-variance ratio as the weights. Since these parameters are closely related to the ratio-change statistic in intensity-based change detection algorithms, their introduction frees the estimator from the need for the equal-variance assumption and enables detection results to largely combine the advantages of intensity-based and CCD methods. Experiments on simulated and real SAR image pairs demonstrate the effectiveness of the proposed estimator in highlighting the change, obviously improving the contrast between the change and the background.

Semi-Automated, Object-Based Tomography of Dislocation Structures.

The characterization of the three-dimensional arrangement of dislocations is important for many analyses in materials science. Dislocation tomography in transmission electron microscopy is conventionally accomplished through intensity-based reconstruction algorithms. Although such methods work successfully, a disadvantage is that they require many images to be collected over a large tilt range. Here, we present an alternative, semi-automated object-based approach that reduces the data collection requirements by drawing on the prior knowledge that dislocations are line objects. Our approach consists of three steps: (1) initial extraction of dislocation line objects from the individual frames, (2) alignment and matching of these objects across the frames in the tilt series, and (3) tomographic reconstruction to determine the full three-dimensional configuration of the dislocations. Drawing on innovations in graph theory, we employ a node-line segment representation for the dislocation lines and a novel arc-length mapping scheme to relate the dislocations to each other across the images in the tilt series. We demonstrate the method for a dataset collected from a dislocation network imaged by diffraction-contrast scanning transmission electron microscopy. Based on these results and a detailed uncertainty analysis for the algorithm, we discuss opportunities for optimizing data collection and further automating the method.

Normalization in dynamic speckle analysis for non-destructive monitoring of speed of processes

The paper is dedicated to analysis of normalized intensity-based pointwise algorithms for processing dynamic speckle images with spatially varying speckle statistics in non-destructive visualization of regions of faster or slower changes across an object. Both existing and newly proposed algorithms are analyzed. Extraction of speed of changes is done by acquiring correlated in time speckle images formed on the object surface under laser illumination. The studied algorithms have been applied to simulated low and high contrast speckle data. Their performance has been compared to processing of binary patterns as another approach for dealing with varying speckle statistics in the acquired images. The efficiency of the algorithms have been checked on the experimental data, including data in a compressed format. We have proven that the algorithms with normalization at successive instants by a sum of two intensities or a single intensity outperform as a whole the algorithms which apply the time-averaged estimates of the mean value and the variance of speckle intensity.

Liver DCE-MRI registration based on sparse recovery of contrast agent curves.

Dynamic contrast-enhanced MRI (DCE-MRI) registration is a challenging task because of the effect of remarkable intensity changes caused by contrast agent injections. Unrealistic deformation usually occurs by using traditional intensity-based algorithms. To alleviate the effect of contrast agent on registration, we proposed a DCE-MRI registration strategy and investigated the registration performance on the clinical DCE-MRI time series of liver. We reconstructed the time-intensity curves of the contrast agent through sparse representation with a predefined dictionary whose columns were the time-intensity curves with high correlations with respect to a preselected contrast agent curve. After reshaping 1D-reconstructed contrast agent time-intensity curves into a 4D contrast agent time series, we aligned the original time series to the reconstructed contrast agent time series through traditional free-form deformation (FFD) registration scheme combined with a residual complexity (RC) similarity and an iterative registration strategy. This study included the DCE-MRI time series of 20 patients with liver cancer. Qualitatively, the time-cut images and subtraction images of different registration methods did not obviously differ. Quantitatively, the mean (standard deviation) of temporal intensity smoothness of all the patients achieved 54.910 (18.819), 54.609 (18.859), and 53.391 (19.031) in FFD RC, RDDR, Zhou etal.'s method and the proposed method, respectively. The mean (standard deviation) of changes in the lesion volume were 0.985 (0.041), 0.983 (0.041), 0.981 (0.046), and 0.989 (0.036) in FFD RC, RDDR, Zhou etal.'s method and the proposed method. Our proposed method would be an effective registration strategy for DCE-MRI time series, and its performance was comparable with that of three advanced registration methods.

Supervised Learning With Perceptual Similarity for Multimodal Gene Expression Registration of a Mouse Brain Atlas.

The acquisition of high quality maps of gene expression in the rodent brain is of fundamental importance to the neuroscience community. The generation of such datasets relies on registering individual gene expression images to a reference volume, a task encumbered by the diversity of staining techniques employed, and by deformations and artifacts in the soft tissue. Recently, deep learning models have garnered particular interest as a viable alternative to traditional intensity-based algorithms for image registration. In this work, we propose a supervised learning model for general multimodal 2D registration tasks, trained with a perceptual similarity loss on a dataset labeled by a human expert and augmented by synthetic local deformations. We demonstrate the results of our approach on the Allen Mouse Brain Atlas (AMBA), comprising whole brain Nissl and gene expression stains. We show that our framework and design of the loss function result in accurate and smooth predictions. Our model is able to generalize to unseen gene expressions and coronal sections, outperforming traditional intensity-based approaches in aligning complex brain structures.

Deep-learning-based image registration and automatic segmentation of organs-at-risk in cone-beam CT scans from high-dose radiation treatment of pancreatic cancer.

Accurate deformable registration between computed tomography (CT) and cone-beam CT (CBCT) images of pancreatic cancer patients treated with high biologically effective radiation doses is essential to assess changes in organ-at-risk (OAR) locations and shapes and to compute delivered dose. This study describes the development and evaluation of a deep-learning (DL) registration model to predict OAR segmentations on the CBCT derived from segmentations on the planning CT. The DL model is trained with CT-CBCT image pairs of the same patient, on which OAR segmentations of the small bowel, stomach, and duodenum have been manually drawn. A transformation map is obtained, which serves to warp the CT image and segmentations. In addition to a regularity loss and an image similarity loss, an OAR segmentation similarity loss is also used during training, which penalizes the mismatch between warped CT segmentations and manually drawn CBCT segmentations. At test time, CBCT segmentations are not required as they are instead obtained from the warped CT segmentations. In an IRB-approved retrospective study, a dataset consisting of 40 patients, each with one planning CT and two CBCT scans, was used in a fivefold cross-validation to train and evaluate the model, using physician-drawn segmentations as reference. Images were preprocessed to remove gas pockets. Network performance was compared to two intensity-based deformable registration algorithms (large deformation diffeomorphic metric mapping [LDDMM] and multimodality free-form [MMFF]) as baseline. Evaluated metrics were Dice similarity coefficient (DSC), change in OAR volume within a volume of interest (enclosing the low-dose PTV plus 1cm margin) from planning CT to CBCT, and maximum dose to 5 cm3 of the OAR [D(5cc)]. Processing time for one CT-CBCT registration with the DL model at test time was less than 5seconds on a GPU-based system, compared to an average of 30minutes for LDDMM optimization. For both small bowel and stomach/duodenum, the DL model yielded larger median DSC and smaller interquartile variation than either MMFF (paired t-test P<10-4 for both type of OARs) or LDDMM (P<10-3 and P=0.03 respectively). Root-mean-square deviation (RMSD) of DL-predicted change in small bowel volume relative to reference was 22% less than for MMFF (P=0.007). RMSD of DL-predicted stomach/duodenum volume change was 28% less than for LDDMM (P=0.0001). RMSD of DL-predicted D(5cc) in small bowel was 39% less than for MMFF (P=0.001); in stomach/duodenum, RMSD of DL-predicted D(5cc) was 18% less than for LDDMM (P<10-3 ). The proposed deep network CT-to-CBCT deformable registration model shows improved segmentation accuracy compared to intensity-based algorithms and achieves an order-of-magnitude reduction in processing time.

Assessing Protein Interactions in Live-Cells with FRET-Sensitized Emission.

Förster Resonance Energy Transfer (FRET) is the radiationless transfer of energy from an excited donor to an acceptor molecule and depends upon the distance and orientation of the molecules as well as the extent of overlap between the donor emission and acceptor absorption spectra. FRET permits to study the interaction of proteins in the living cell over time and in different subcellular compartments. Different intensity-based algorithms to measure FRET using microscopy have been described in the literature. Here, a protocol and an algorithm are provided to quantify FRET efficiency based on measuring both the sensitized emission of the acceptor and quenching of the donor molecule. The quantification of ratiometric FRET in the living cell not only requires the determination of the crosstalk (spectral spill-over, or bleed-through) of the fluorescent proteins but also the detection efficiency of the microscopic setup. The protocol provided here details how to assess these critical parameters.

A level set method based on domain transformation and bias correction for MRI brain tumor segmentation

BackgroundIntensity inhomogeneity is one of the common artifacts in image processing. This artifact makes image segmentation more challenging and adversely affects the performance of intensity-based image processing algorithms. New methodIn this paper, a novel region-based level set method is proposed for segmenting the images with intensity inhomogeneity with applications to brain tumor segmentation in magnetic resonance imaging (MRI) scans. For this purpose, the inhomogeneous regions are first modeled as Gaussian distributions with different means and variances, and then transferred into a new domain, where preserves the Gaussian intensity distribution of each region but with better separation. Moreover, our method can perform bias field correction. To this end, the bias field is represented by a linear combination of smooth base functions that enables better intensity inhomogeneity modeling. Therefore, level set fundamental formulation and bias field are modified in the proposed approach. ResultsTo assess the performance of the proposed method, different inhomogeneous images, including synthetic images as well as real brain magnetic resonance images from BraTS 2017 dataset are segmented. Being evaluated by Dice, Jaccard, Sensitivity, and Specificity metrics, the results show that the proposed method suppresses the side effect of the over-smoothing object boundary and it has good accuracy in the segmentation of images with extreme intensity non-uniformity. The mean values of these metrics in brain tumor segmentation are 0.86 ± 0.03, 0.77 ± 0.05, 0.94 ± 0.04, 0.99 ± 0.003, respectively. Comparison with existing method(s)Our method were compared with six state-of-the-art image segmentation methods: Chan-Vese (CV), Local Intensity Clustering (LIC), Local iNtensity Clustering (LINC), Global inhomogeneous intensity clustering (GINC), Multiplicative Intrinsic Component Optimization (MICO), and Local Statistical Active Contour Model (LSACM) models. We used qualitative and quantitative comparison methods for segmenting synthetic and real images. Experiments indicate that our proposed method is robust to noise and intensity non-uniformity and outperforms other state-of-the-art segmentation methods in terms of bias field correction, noise resistance, and segmentation accuracy. ConclusionsExperimental results show that the proposed model is capable of accurate segmentation and bias field estimation simultaneously. The proposed model suppresses the side effect of the over-smoothing object boundary. Moreover, our model has good accuracy in the segmentation of images with extreme intensity non-uniformity.

Image registration comparative analysis: normalized correlation versus SIFT-based registration

The paper compares the image registration algorithms: the classical normalized correlation (as a representative of intensity-based algorithms) and the SIFT-based algorithm (feature-based registration). A gradient subpixel correction algorithm was also used for normalized correlation. We compared the effectiveness of their work on real images (including a terrain map) when modeling artificial distortions. The accuracy of determining the position (shift) of one image relative to another in the presence of rotation and scale changes was studied. The experiment was carried out using a simulation model created in the Python programming language using the OpenCV computer vision library.&#x0D; The results of the experiments show that in the absence of rotation and scale changes between the registered images the normalized correlation provides a slightly smaller root-mean-square error. At the same time, if there are even small such distortions, for example, a rotation of more than 2 degrees and a scale change of more than 2 percent, the probability of correct registration for the normalized correlation drops sharply. It was also noted that the advantages of normalized correlation are almost 5 times higher speed and the possibility of using it for small fragments (50x50 or less), where it is problematic for the SIFT algorithm to allocate a sufficient number of keypoints.&#x0D; It was also shown that the use of a two-stage algorithm (SIFT-based registration at the first stage, and optimization with normalized correlation as a criterion at the second) allows you to get both high accuracy and stability to rotation and scale change, but this will be accompanied by high computational costs.

Generalized and Mechanistic PV Module Performance Prediction From Computer Vision and Machine Learning on Electroluminescence Images

Electroluminescence (EL) imaging of photovoltiac (PV) modules offers high-speed, high-resolution information about device performance, affording opportunities for greater insight and efficiency in module characterization across manufacturing, research and development, and power plant operations and management. Predicting module electrical properties from EL image features is a critical step toward these applications. In this article, we demonstrate quantification of both generalized and performance mechanism-specific EL image features, using pixel intensity-based and machine learning classification algorithms. From EL image features, we build predictive models for PV module power and series resistance, using time-series current–voltage ( I–V ) and EL data obtained stepwise on five brands of modules spanning three Si cell types through two accelerated exposures: damp heat (DH) (85 $^\circ$ C/85% RH) and thermal cycling (TC) (IEC 61215). In total, 195 pairs of EL images and I–V characteristics were analyzed, yielding 11 700 individual PV cell images. A convolutional neural network was built to classify cells by the severity of busbar corrosion with high accuracy (95%). Generalized power predictive models estimated the maximum power of PV modules from EL images with high confidence and an adjusted- $R^2$ of 0.88, across all module brands and cell types in extended DH and TC exposures. Mechanistic degradation prediction was demonstrated by quantification of busbar corrosion in EL images of three module brands in DH, and subsequent modeling of series resistance using these mechanism-specific EL image features. For modules exhibiting busbar corrosion, we demonstrated series resistance predictive models with adjusted- $R^2$ of up to 0.73.

A New Approach for Automated Image Segmentation of Organs at Risk in Cervical Cancer

HomeRadiology: Imaging CancerVol. 2, No. 2 Previous Research HighlightsFree AccessA New Approach for Automated Image Segmentation of Organs at Risk in Cervical CancerMeredith Jones, Lacey R. McNallyMeredith Jones, Lacey R. McNallyMeredith JonesLacey R. McNallyPublished Online:Mar 27 2020https://doi.org/10.1148/rycan.2020204010MoreSectionsPDF ToolsImage ViewerAdd to favoritesCiteTrack CitationsPermissionsReprints ShareShare onFacebookTwitterLinked InEmail Take-Away Points■ Major Focus: Development of an automated multiclass image segmentation model to identify organs at risk (OARs) on CT scans in patients with locally advanced cervical cancer, prior to radiation therapy.■ Key Results: The convolutional neural network model outperformed model-based, intensity-based, and atlas-based segmentation algorithms, with comparable results to other deep learning segmentation models while also decreasing the time needed for image segmentation by several thousand-fold.■ Impact: Reduction of intra- and interuser variability imposed by manual OAR delineation can lead to more accurate and efficient radiation therapy planning.The recent article by Liu et al developed an automated multiclass image segmentation model to identify organs at risk (OARs) on CT scans. To plan effective and safe radiation therapy in the context of cervical cancer, a CT scan is often acquired prior to the first treatment. Currently, a radiation oncologist will segment the CT images by hand, which is a time-consuming process, subject to intra- and interobserver variability, to identify OARs. To overcome these potential limitations and biases, automated image segmentation techniques can be utilized to improve accuracy and speed of annotating each image. A major problem in creation of deep learning image segmentation models for medical images is the lack of available datasets for model training and validation. Liu et al developed a two-dimensional image segmentation model based off the established convolutional neural network architecture, U-Net. U-Net was utilized because it is feasible to train from scratch with a small training set.This study looked at the CT scans of 105 patients with locally advanced cervical cancer. OAR segmentation for cervical cancer is a particularly challenging problem due to variability in organ size and frequent unclear boundaries. Even an experienced radiologist will have trouble. Seven organs were identified as OARs for cervical cancer radiation therapy: bladder, bone marrow, femoral head left, femoral head right, rectum, small intestine, and spinal cord.The model uses context aggregation blocks where U-Net has convolution layers. These context aggregation blocks use dilation convolution layers with different dilated rates and normal convolution layers with different kernel sizes to yield a low false-positive rate and accurate feature extraction in a large receptive field. A squeeze-extract block was used to ensure proper consideration of organs that are small. A class-weighted cross-entropy loss function was also used to overcome the imbalance in the datasets created by the differences in size and shape of each OAR.A ground truth was available based on the radiation oncologist’s annotations, allowing a Dice similarity coefficient (DSC) and 95th Hausdorff distance (HD) to be calculated and used as an evaluation metric. The DSC and the HD metrics indicated a comparable performance to other deep learning image segmentation models in most of the organs analyzed; inferior results were seen in the rectum and superior results in the intestines. All results were analyzed by an oncologist who determined that greater than 90% of results produced by this model were highly acceptable to be used during radiation therapy planning. To make the model more robust and generalizable, it is necessary to acquire a larger dataset with scans from different sources. Translation of this model to three dimensional may ultimately improve clinical translation.Highlighted ArticleLiu Z, Liu X, Xiao B, et al. Segmentation of organs-at-risk in cervical cancer CT images with a convolutional neural network. Physica Medica 2020;69:184-191. doi: https://doi.org/10.1016/j.ejmp.2019.12.008Highlighted ArticleLiu Z, Liu X, Xiao B, et al. Segmentation of organs-at-risk in cervical cancer CT images with a convolutional neural network. Physica Medica 2020;69:184-191. doi: https://doi.org/10.1016/j.ejmp.2019.12.008 Crossref, Medline, Google ScholarArticle HistoryPublished online: Mar 27 2020 FiguresReferencesRelatedDetailsCited ByJournal of Biomimetics, Biomaterials and Biomedical Engineering, Vol. 49Mathematical Problems in Engineering, Vol. 2020Recommended Articles nnU-Net: Further Automating Biomedical Image AutosegmentationRadiology: Imaging Cancer2021Volume: 3Issue: 1Deep Learning: A Primer for RadiologistsRadioGraphics2017Volume: 37Issue: 7pp. 2113-2131Generative Adversarial Networks: A Primer for RadiologistsRadioGraphics2021Volume: 41Issue: 3pp. 840-857Current Applications and Future Impact of Machine Learning in RadiologyRadiology2018Volume: 288Issue: 2pp. 318-328Deep Generative Adversarial Networks: Applications in Musculoskeletal ImagingRadiology: Artificial Intelligence2021Volume: 3Issue: 3See More RSNA Education Exhibits A Multidisciplinary Approach for Program Development with Artificial Intelligence in Pancreatic Cancer: How We Fit InDigital Posters2019Abnormal Changes After Radiation Therapy: What Radiologists Should KnowDigital Posters2018Artificial Intelligence in Diagnostic Imaging: Current Applications and Future PerspectiveDigital Posters2019 RSNA Case Collection Acute Radiation EnteropathyRSNA Case Collection2021Small Bowel LymphomaRSNA Case Collection2021Hirschsprung DiseaseRSNA Case Collection2021 Vol. 2, No. 2 Metrics Downloaded 310 times Altmetric Score

HeadLocNet: Deep convolutional neural networks for accurate classification and multi-landmark localization of head CTs.

Cochlear implants (CIs) are used to treat subjects with hearing loss. In a CI surgery, an electrode array is inserted into the cochlea to stimulate auditory nerves. After surgery, CIs need to be programmed. Studies have shown that the cochlea-electrode spatial relationship derived from medical images can guide CI programming and lead to significant improvement in hearing outcomes. We have developed a series of algorithms to segment the inner ear anatomy and localize the electrodes. But, because clinical head CT images are acquired with different protocols, the field of view and orientation of the image volumes vary greatly. As a consequence, visual inspection and manual image registration to an atlas image are needed to document their content and to initialize intensity-based registration algorithms used in our processing pipeline. For large-scale evaluation and deployment of our methods these steps need to be automated. In this article we propose to achieve this with a deep convolutional neural network (CNN) that can be trained end-to-end to classify a head CT image in terms of its content and to localize landmarks. The detected landmarks can then be used to estimate a point-based registration with the atlas image in which the same landmark set's positions are known. We achieve 99.5% classification accuracy and an average localization error of 3.45 mm for 7 landmarks located around each inner ear. This is better than what was achieved with earlier methods we have proposed for the same tasks.

Using Image Registration and Machine Learning to Develop a Workstation Tool for Rapid Analysis of Glomeruli in Medical Renal Biopsies

Background: Prescreening of biopsies has the potential to improve pathologists’ workflow. Tools that identify features and display results in a visually thoughtful manner can enhance efficiency, accuracy, and reproducibility. Machine learning for detection of glomeruli ensures comprehensive assessment and registration of four different stains allows for simultaneous navigation and viewing. Methods: Medical renal core biopsies (4 stains each) were digitized using a Leica SCN400 at *40 and loaded into the Corista Quantum research platform. Glomeruli were manually annotated by pathologists. The tissue on the 4 stains was registered using a combination of keypoint- and intensity-based algorithms, and a 4-panel simultaneous viewing display was created. Using a training cohort, machine learning convolutional neural net (CNN) models were created to identify glomeruli in all stains, and merged into composite fields of views (FOVs). The sensitivity and specificity of glomerulus detection, and FOV area for each detection were calculated. Results: Forty-one biopsies were used for training (28) and same-batch evaluation (6). Seven additional biopsies from a temporally different batch were also evaluated. A variant of AlexNet CNN, used for object recognition, showed the best result for the detection of glomeruli with same-batch and different-batch evaluation: Same-batch sensitivity 92%, “modified” specificity 89%, average FOV size represented 0.8% of the total slide area; different-batch sensitivity 90%, “modified” specificity 98% and average FOV size 1.6% of the total slide area. Conclusions: Glomerulus detection in the best CNN model shows that machine learning algorithms may be accurate for this task. The added benefit of biopsy registration with simultaneous display and navigation allows reviewers to move from one machine-generated FOV to the next in all 4 stains. Together these features could increase both efficiency and accuracy in the review process.

Motion correction of chemical exchange saturation transfer MRI series using robust principal component analysis (RPCA) and PCA.

Chemical exchange saturation transfer (CEST) MRI requires the acquisition of multiple saturation-weighted images and can last several minutes. Misalignments among these images, which are often due to the inevitable motion of the subject, will corrupt CEST contrast maps and result in large quantification errors. Therefore, the registration of the CEST series is critical. However, registration is challenging since common intensity-based registration algorithms may fail to differentiate CEST signals from motion artifacts. Herein, we studied how different patterns of motion affect CEST quantification and proposed a cascaded two-step registration scheme by utilizing features extracted from the entire Z-spectral image series instead of direct registration to a single image. The proposed approach is conducted in two stages: during the first coarse registration, the Z-spectral image series is decomposed by robust principal component analysis (RPCA) to separate CEST contrast from motion. The recomposed image series using only the low-rank component, which contains minimized motion, are averaged to generate a reference for the alignment of the image series. To further remove residual misalignments, the coarse registration is followed by a refinement stage, which uses PCA iteratively to generate motionless synthetic reference series with the first few principal components (PCs) that correspond to CEST contrast. In the end, the quality check is performed to exclude the images with unsuccessful registration. The proposed registration scheme (RPCA + PCA_R) was assessed by both phantom experiments and in vivo data of tumor-bearing mouse brain, with simulated random rigid motion in different patterns applied to the acquired static Z-spectral image series. For comparison, previous correction schemes using an explicit image [either S0 or Ssat(∆ω)] as registration reference were also performed, named as S0_R and Ssat_R respectively. To illustrate the advantage of combination of RPCA and PCA, registration was also exploited using either only the RPCA-based method (RPCA_R) or only the PCA-based one (PCA_R). Compared with the above four methods, RPCA + PCA_R allowed for more accurate correction of the corrupted Z-spectral images, exhibiting smaller MTRasym(∆ω) error maps and lower residual Z-spectra referring to the static data. Among all the five correction methods, the corrected Z-spectral image series by RPCA + PCA_R and the resulting MTRasym(∆ω) maps achieved the highest correlation coefficients (CC) with respect to the static ones. The registration scheme of RPCA + PCA_R provides robust motion correction between two specific Z-spectral images and among an entire image series, through extraction of the static component from the entire Z-spectra set and inclusion of a PCA-based refinement step. Therefore, this method can help improve CEST acquisition and quantification.