Detection Of Abnormal Regions Research Articles

Radiomic detection of abnormal brain regions in tuberous sclerosis complex.

Radiomics refers to the extraction of quantitative information from medical images and is most commonly utilized in oncology to provide ancillary information for solid tumor diagnosis, prognosis, and treatment response. The traditional radiomic pipeline involves segmentation of volumes of interest with comparison to normal brain. In other neurologic disorders, such as epilepsy, lesion delineation may be difficult or impossible due to poor anatomic definition, small size, and multifocal or diffuse distribution. Tuberous sclerosis complex (TSC) is a rare genetic disease in which brain magnetic resonance imaging (MRI) demonstrates multifocal abnormalities with variable imaging and epileptogenic features. The purpose of this study was to develop a radiomic workflow for identification of abnormal brain regions in TSC, using a whole-brain atlas-based approach with generation of heatmaps based on signal deviation from normal controls. This was a retrospective pilot study utilizing high-resolution whole-brain 3D FLAIR MRI datasets from retrospective enrollment of tuberous sclerosis complex (TSC) patients and normal controls. Subjects underwent MRI including high-resolution 3D FLAIR sequences. Preprocessing included skull stripping, coregistration, and intensity normalization. Using the Brainnetome and Harvard-Oxford atlases, brain regions were parcellated into 318 discrete regions. Expert neuroradiologists spatially labeled all tubers in TSC patients using ITK-SNAP. The pyradiomics toolbox was used to extract 88 radiomic features based on IBSI guidelines, comparing tuber-affected and non-tuber-affected parenchyma in TSC patients, as well as normal brain tissue in control patients. For model training and validation, regions with tubers from 20 TSC patients and 30 normal control subjects were randomly divided into two training sets (80%) and two validation sets (20%). Additional model testing was performed on a separate group of 20 healthy controls. LASSO (least absolute shrinkage and selection operator) was used to perform variable selection and regularization to identify regions containing tubers. Relevant radiomic features selected by LASSO were combined to produce a radiomic score ω, defined as the sum of squared differences from average control group values. Region-specific ω scores were converted to heat maps and spatially coregistered with brain MRI to reflect overall radiomic deviation from normal. The proposed radiomic workflow allows for quantification of deviation from normal in 318 regions of the brain with the use of a summative radiomic score ω. This score can be used to generate spatially registered heatmaps to identify brain regions with radiomic abnormalities. The pilot study of TSC showed radiomic scores ω that were statistically different in regions containing tubers from regions without tubers/normal brain (p<0.0001). Our model exhibits an AUC of 0.81 (95% confidence interval: 0.78-0.84) on the testing set, and the best threshold obtained on the training set, when applied to the testing set, allows us to identify regions with tubers with a specificity of 0.91 and a sensitivity of 0.60. We describe a whole-brain atlas-based radiomic approach to identify abnormal brain regions in TSC patients. This approach may be helpful for identifying specific regions of interest based on relatively greater signal deviation, particularly in clinical scenarios with numerous or poorly defined anatomic lesions.

Robust anomaly detection in industrial images by blending global–local features

AbstractIndustrial image anomaly detection achieves automated detection and localization of defects or abnormal regions in images through image processing and deep learning techniques. Currently, utilizing the approach of reverse knowledge distillation has yielded favourable outcomes. However, it is still a challenge in terms of the feature extraction capability of the image and the robustness of the decoding of the student network. This study first addresses the issue that the teacher network has not been able to extract global information more effectively. To acquire more global information, a vision transformer network is introduced to enhance the model's global information extraction capability, obtaining better features to further assist the student network in decoding. Second, for anomalous samples, to address the low similarity between features extracted by the teacher network and features restored by the student network, Gaussian noise is introduced. This further increases the probability that the features decoded by the student model match normal sample features, enhancing the robustness of the student model. Extensive experiments were conducted on industrial image datasets AeBAD, MvtecAD, and BTAD. In the AeBAD dataset, under the PRO performance metric, the result is 89.83%, achieving state‐of‐the‐art performance. Under the AUROC performance metric, it reaches 83.35%. Similarly, good results were achieved on the MvtecAD and BTAD datasets. The proposed method's effectiveness and performance advantages were validated across multiple industrial datasets, providing a valuable reference for the application of industrial image anomaly detection methods.

Automated tumor segmentation and brain tissue extraction from multiparametric MRI of pediatric brain tumors: A multi-institutional study.

Brain tumors are the most common solid tumors and the leading cause of cancer-related death among all childhood cancers. Tumor segmentation is essential in surgical and treatment planning, and response assessment and monitoring. However, manual segmentation is time-consuming and has high interoperator variability. We present a multi-institutional deep learning-based method for automated brain extraction and segmentation of pediatric brain tumors based on multi-parametric MRI scans. Multi-parametric scans (T1w, T1w-CE, T2, and T2-FLAIR) of 244 pediatric patients ( n = 215 internal and n = 29 external cohorts) with de novo brain tumors, including a variety of tumor subtypes, were preprocessed and manually segmented to identify the brain tissue and tumor subregions into four tumor subregions, i.e., enhancing tumor (ET), non-enhancing tumor (NET), cystic components (CC), and peritumoral edema (ED). The internal cohort was split into training ( n = 151), validation ( n = 43), and withheld internal test ( n = 21) subsets. DeepMedic, a three-dimensional convolutional neural network, was trained and the model parameters were tuned. Finally, the network was evaluated on the withheld internal and external test cohorts. Dice similarity score (median ± SD) was 0.91 ± 0.10/0.88 ± 0.16 for the whole tumor, 0.73 ± 0.27/0.84 ± 0.29 for ET, 0.79 ± 19/0.74 ± 0.27 for union of all non-enhancing components (i.e., NET, CC, ED), and 0.98 ± 0.02 for brain tissue in both internal/external test sets. Our proposed automated brain extraction and tumor subregion segmentation models demonstrated accurate performance on segmentation of the brain tissue and whole tumor regions in pediatric brain tumors and can facilitate detection of abnormal regions for further clinical measurements.

Study on Significance Enhancement Algorithm of Abnormal Features of Urban Road Ground Penetrating Radar Images

Currently, ground penetrating radar is the major technology for the detection of urban road collapses and disaster sources. Vehicle-mounted GPR collects tens of GB of data on site every day, but the present interpretation of abnormal regions detected by radar relies on manual interpretation with low process efficiency. The abnormal region images of GPR are different from the surrounding normal images. In terms of the features of abnormal regions in GPR images with an obvious brightness change and obvious directional characteristics, an abnormal region detection algorithm based on visual attention mechanism is proposed. Firstly, the complex background noise in the GPR images is suppressed by wavelet denoising and gamma transform, and the brightness and directional characteristics of the abnormal regions are enhanced. Secondly, by building a multi-scale image brightness and orientation feature pyramid model, the features of abnormal regions of interest are continuously enhanced, and the rapid screening of abnormal regions has been achieved. The effectiveness of the algorithm has been verified by actual tests on different types of abnormal radar image data.

Architectural Distortion-Based Digital Mammograms Classification Using Depth Wise Convolutional Neural Network.

Simple SummaryBreast cancer is leading cancer increases the death rate in women. Early diagnosis of breast cancer in women can save their lives. The current study proposed a novel scheme to detect architectural distortion from mammogram images to predict breast cancer using a deep learning approach. Results are evaluated on a public and a private dataset which may help to improve the diagnostic ability of breast cancer of radiologists and doctors in daily clinical routines. Furthermore, the proposed method achieved maximum accuracy as compared with previous approaches. This study can be interesting and valuable in the healthcare predictive modeling domain and will add a real contribution to society.Architectural distortion is the third most suspicious appearance on a mammogram representing abnormal regions. Architectural distortion (AD) detection from mammograms is challenging due to its subtle and varying asymmetry on breast mass and small size. Automatic detection of abnormal ADs regions in mammograms using computer algorithms at initial stages could help radiologists and doctors. The architectural distortion star shapes ROIs detection, noise removal, and object location, affecting the classification performance, reducing accuracy. The computer vision-based technique automatically removes the noise and detects the location of objects from varying patterns. The current study investigated the gap to detect architectural distortion ROIs (region of interest) from mammograms using computer vision techniques. Proposed an automated computer-aided diagnostic system based on architectural distortion using computer vision and deep learning to predict breast cancer from digital mammograms. The proposed mammogram classification framework pertains to four steps such as image preprocessing, augmentation and image pixel-wise segmentation. Architectural distortion ROI’s detection, training deep learning, and machine learning networks to classify AD’s ROIs into malignant and benign classes. The proposed method has been evaluated on three databases, the PINUM, the CBIS-DDSM, and the DDSM mammogram images, using computer vision and depth-wise 2D V-net 64 convolutional neural networks and achieved 0.95, 0.97, and 0.98 accuracies, respectively. Experimental results reveal that our proposed method outperforms as compared with the ShuffelNet, MobileNet, SVM, K-NN, RF, and previous studies.

Detecting, extracting and classifying foreign objects in inter-basin channels to ensure water supply safety

Abstract Foreign objects (e.g., livestock, rafting, and vehicles) intruded into inter-basin channels pose threats to water quality and water supply safety. Timely detection of the foreign objects and acquiring relevant information (e.g., quantities, geometry, and types) is a premise to enforce proactive measures to control potential loss. Large-scale water channels usually span a long distance and hence are difficult to be efficiently covered by manual inspection. Applying unmanned aerial vehicles for inspection can provide time-sensitive aerial images, from which intrusion incidents can be visually pinpointed. To automate the processing of such aerial images, this paper aims to propose a method based on computer vision to detect, extract, and classify foreign objects in water channels. The proposed approach includes four steps, i.e., aerial image preprocessing, abnormal region detection, instance extraction, and foreign object classification. Experiments demonstrate the efficacy of the approach, which can recognize three typical foreign objects (i.e., livestock, rafting, and vehicle) with a robust performance. The proposed approach can raise early awareness of intrusion incidents in water channels for water quality assurance.

An Intuitionistic Fuzzy Clustering Approach for Detection of Abnormal Regions in Mammogram Images.

Breast cancer is one of the leading causes of mortality in the world and it occurs in high frequency among women that carries away many lives. To detect cancer, extraction or segmentation of lesions/tumors is required. Segmentation process is very crucial if the mammogram images are blurred or low contrast. This paper suggests a novel clustering approach for segmenting lesions/tumors in the mammogram images using Atanassov's intuitionistic fuzzy set theory. The algorithm initially converts an image to an intuitionistic fuzzy image using a novel intuitionistic fuzzy generator. From the intuitionistic fuzzy image, two membership intervals are computed. Then, using Zadeh's min t-conorm, a new membership function is computed. Using the new membership function, an interval type 2 fuzzy image is constructed. Two types of distance functions are used in clustering-intuitionistic fuzzy divergence and a fuzzy exponential type distance function. Further, in each iteration, membership matrix is updated using a hesitation degree and a clustered image is obtained. Tumors/lesions are then segmented from the clustered image. The proposed method is compared with existing methods both quantitatively and qualitatively and it is observed that the proposed method performs better than the existing methods.

Regional abnormality of functional connectivity is associated with clinical manifestations in individuals with intractable focal epilepsy

We explored regional functional connectivity alterations in intractable focal epilepsy brains using resting-state functional MRI. Distributions of the network parameters (corresponding to degree and eigenvector centrality) measured at each brain region for all 25 patients were significantly different from age- and sex-matched control data that were estimated by a healthy control dataset (n = 582, 18–84 years old). The number of abnormal regions whose parameters exceeded the mean + 2 SD of age- and sex-matched data for each patient were associated with various clinical parameters such as the duration of illness and seizure severity. Furthermore, abnormal regions for each patient tended to have functional connections with each other (mean ± SD = 58.6 ± 20.2%), the magnitude of which was negatively related to the quality of life. The abnormal regions distributed within the default mode network with significantly higher probability (p < 0.05) in 7 of 25 patients. We consider that the detection of abnormal regions by functional connectivity analysis using a large number of control datasets is useful for the numerical assessment of each patient’s clinical conditions, although further study is necessary to elucidate etiology-specific abnormalities.

Quantitative Investigation of Tomographic Effects in Abnormal Regions of Complex Structures

The detection of abnormal regions in complex structures is one of the most challenging targets for underground space engineering. Natural or artificial geologic variations reduce the effectiveness of conventional exploration methods. With the emergence of real-time monitoring, seismic wave velocity tomography allows the detection and imaging of abnormal regions to be accurate, intuitive, and quantitative. Since tomographic results are affected by multiple factors in practical small-scale applications, it is necessary to quantitatively investigate those influences. We adopted an improved three-dimensional (3D) tomography method combining passive acoustic emission acquisition and active ultrasonic measurements. By varying individual parameters (i.e., prior model, sensor configuration, ray coverage, event distributions, and event location errors), 37 comparative tests were conducted. The quantitative impact of different factors was obtained. Synthetic experiments showed that the method could effectively adapt to complex structures. The optimal input parameters based on quantization results can significantly improve the detection reliability in abnormal regions.

Detection of Abnormal Tumor Regions in Ultrasonic Thyroid Images using SVM Classification Method

Detection of tumor or abnormal regions in thyroid gland is difficult task in human. The following methods are presently used for detecting the abnormal regions in thyroid gland as blood test, sample testing from thyroid gland and image processing method. This paper develops a methodology to detect the tumor regions in thyroid images using image registration and image enhancement technique. The Support Vector Machine (SVM) classifier is operated in two modes as training pattern generation and testing mode. The generation of training pattern from both normal and abnormal ultrasonic thyroid images. This proposed method for thyroid tumor region detection obtains 96.54% of sensitivity, 97.57% of specificity and 98.56% of average tumor segmentation accuracy.

Recognition of Different Types of Leukocytes Using YOLOv2 and Optimized Bag-of-Features

White blood cells (WBCs) protect human body against different types of infections including fungal, parasitic, viral, and bacterial. The detection of abnormal regions in WBCs is a difficult task. Therefore a method is proposed for the localization of WBCs based on YOLOv2-Nucleus-Cytoplasm, which contains darkNet-19 as a basenetwork of the YOLOv2 model. In this model features are extracted from LeakyReLU-18 of darkNet-19 and supplied as an input to the YOLOv2 model. The YOLOv2-Nucleus-Cytoplasm model localizes and classifies the WBCs with maximum score labels. It also localize the WBCs into the blast and non-blast cells. After localization, the bag-of-features are extracted and optimized by using particle swarm optimization(PSO). The improved feature vector is fed to classifiers i.e., optimized naive Bayes (O-NB) & optimized discriminant analysis (O-DA) for WBCs classification. The experiments are performed on LISC, ALL-IDB1, and ALL-IDB2 datasets.

Abnormal region detection in cervical smear images based on fully convolutional network

Automation-assisted cervical screening via liquid-based cytology has achieved great success using segmentation and classification methods. This work tries to do abnormal region detection on field of view cervical cell images based on deep learning, which is a novel way to solve cervical cytological screening problem. Since some abnormal nuclei gather in groups, the proposed method chooses abnormal regions instead of abnormal nuclei as the detection targets in order to locate the abnormal regions for the further diagnosis of the pathologists. In this study, a novel abnormal region detection approach for cervical screening is proposed based on a size-sensitive fully convolutional network (R-FCN). Due to the regular feature distribution, a fewer-layer convolutional neural backbone network is designed for more efficient feature extraction and less running time. In addition, a new measure named hit degree is defined to describe the degree how closely each detected region and the corresponding ground truth matches up. Experimental results show that an average precision of 93.2% is achieved for abnormal region detection in cervical smear images. The proposed method is promising for the development of computer-aided systems in clinical cervical cytological screening.

A graph-based approach to automated EUS image layer segmentation and abnormal region detection

Endoscopic ultrasonography (EUS) has shown great advantages in the diagnosis and staging of gastrointestinal malignant tumors. However, EUS based diagnosis is limited by variability in the examiner’s subjective interpretation to differentiate between normal and early esophageal carcinoma. In this paper, we propose a novel approach aiming at automatic abnormal region detection from the esophageal EUS images; the contribution is three-fold: first, we present a series of preprocessing strategies developed specifically for the enhancement of EUS images to aid the estimation in the subsequent works. Second, we provide an automatic layer segmentation method based on the multiple surface graph searching approach with incorporation of geometric constraints, which is applied to segment the EUS images into five discernible layers. Third, we introduce the novel feature extraction strategy by utilizing the features from each column in the segmented layers. The SVM classifier is then applied to fulfill the normal and early esophageal carcinoma classification. Subsequently, a clustering method is used to assemble the abnormal columns together so as to detect the abnormal region. Experimental results show that our method has demonstrated its robustness even facing noisy EUS images, and has achieved high accuracy in detecting abnormal region.

Tinnitus Abnormal Brain Region Detection Based on Dynamic Causal Modeling and Exponential Ranking

Dynamic Causal Modeling (DCM) has been extended for the analysis of electroencephalography (EEG) based on a specific biophysical and neurobiological generative model for EEG. Comparing to methods that summarize neural activities with linear relationships, the generative model enables DCM to better describe how signals are generated and better reveal the underlying mechanism of the activities occurring in human brains. Since DCM provides us with an approach to the effective connectivity between brain areas, with exponential ranking, the abnormality of the observed signals can be further located to a specific brain region. In this paper, a combination of DCM and exponential ranking is proposed as a new method aiming at searching for the abnormal brain regions which are associated with chronic tinnitus.

Automated Detection of Facial Disorders (ADFD): a novel approach based-on digital photographs

ABSTRACTAbnormal regions on a face are often indicative of facial dermatological diseases, such as rosacea, eczema, allergy, burn injury, facial rash and acne. Identification and quantitative evaluation of these facial disorders is often subjective and currently done by expert dermatologists. However, with the advances in image processing techniques, it is now possible to consistently identify, classify and objectively quantify facial skin disorders using digital photographs. Furthermore, digital photographs can be used to assess progression of the diseases by applying time-series analysis on infected (abnormal) facial regions of the photograph. In this paper, we propose a novel real-time approach for detection and segmentation of abnormal facial regions. Experimental results showed that the proposed method is efficient for segmentation of abnormal regions, which were caused by facial fever or disease, from digital photographs in terms of accuracy and required processing time.

Computer-based classification of chromoendoscopy images using homogeneous texture descriptors

Computer-aided analysis of clinical pathologies is a challenging task in the field of medical imaging. Specifically, the detection of abnormal regions in the frames collected during an endoscopic session is difficult. The variations in the conditions of image acquisition, such as field of view or illumination modification, make it more demanding. Therefore, the design of a computer-assisted diagnostic system for the recognition of gastric abnormalities requires features that are robust to scale, rotation, and illumination variations of the images. Therefore, this study focuses on designing a set of texture descriptors based on the Gabor wavelets that will cope with certain image dynamics. The proposed features are extracted from the images and utilized for the classification of the chromoendoscopy (CH) frames into normal and abnormal categories. Moreover, to attain a higher accuracy, an optimized subset of descriptors is selected through the genetic algorithm. The results obtained using the proposed features are compared with other existing texture descriptors (e.g., local binary pattern and homogeneous texture descriptors). Furthermore, the selected features are used to train the support vector machine (SVM), naive Bayes (NB) algorithm, k-nearest neighbor algorithm, linear discriminant analysis, and ensemble tree classifier. The performance of these state-of-the-art classifiers for different texture descriptors is compared based on the accuracy, sensitivity, specificity, and area under the curve (AUC) derived by using the CH images. The classification results reveal that the SVM classifier achieves 90.0% average accuracy and 0.93 AUC when it is employed with an optimized set of features obtained by using a genetic algorithm.

SU‐F‐J‐226: Structural Similarity‐Based Ultrasound Image Similarity Measurement

Purpose:We explore a method to measure the global ultrasound image similarity measurement, with the aim to facilitate abnormal region detection and automatic tracking failure re‐initialization in the ultrasound image sequences.Methods:Structural similarity is used in our similarity measurement framework. The classical structural similarity assesses the visual impact of three characteristics of the ultrasound imaging region from three aspects: luminance, contrast and structure. Unlike the approach to calculate the local means, standard deviations and cross‐covariance for adjacent two ultrasound images directly, we propose to compute the three aspects based on the local speckle statistics, in which locally adaptive Nakagami distribution‐based measurements are used, resulting in significantly increasing the robustness against the speckle noise.Results:As the tongue can stay in the rest position for long duration and it's easy to conduct the experiment, the proposed similarity measurement is firstly tested on the ultrasound tongue image. The performance demonstrates that the proposed method can provide an accurate similarity description between the frames. Moreover, the potential application of the similarity measurement‐based automatic contour tracking failure re‐initialization is also explored in our experiment. And the results show that using the proposed similarity‐based automatic tracking failure re‐initialization method can improve the accuracy of the tracking. And the tracking error reduce from about 2–3 mm to about 1.5 mm using Euclidean distance measurement.Conclusion:The performance demonstrates that the proposed method can provide an accurate similarity description between the frames. And, the potential applications of the similarity measurement seem evident in several aspects.

Abnormality Detection via Iterative Deformable Registration and Basis-Pursuit Decomposition.

We present a generic method for automatic detection of abnormal regions in medical images as deviations from a normative data base. The algorithm decomposes an image, or more broadly a function defined on the image grid, into the superposition of a normal part and a residual term. A statistical model is constructed with regional sparse learning to represent normative anatomical variations among a reference population (e.g., healthy controls), in conjunction with a Markov random field regularization that ensures mutual consistency of the regional learning among partially overlapping image blocks. The decomposition is performed in a principled way so that the normal part fits well with the learned normative model, while the residual term absorbs pathological patterns, which may then be detected through a statistical significance test. The decomposition is applied to multiple image features from an individual scan, detecting abnormalities using both intensity and shape information. We form an iterative scheme that interleaves abnormality detection with deformable registration, gradually improving robustness of the spatial normalization and precision of the detection. The algorithm is evaluated with simulated images and clinical data of brain lesions, and is shown to achieve robust deformable registration and localize pathological regions simultaneously. The algorithm is also applied on images from Alzheimer's disease patients to demonstrate the generality of the method.

Suggestions for automatic quantitation of endoscopic image analysis to improve detection of small intestinal pathology in celiac disease patients

Although many groups have attempted to develop an automated computerized method to detect pathology of the small intestinal mucosa caused by celiac disease, the efforts have thus far failed. This is due in part to the occult presence of the disease. When pathological evidence of celiac disease exists in the small bowel it is visually often patchy and subtle. Due to presence of extraneous substances such as air bubbles and opaque fluids, the use of computerized automation methods have only been partially successful in detecting the hallmarks of the disease in the small intestine—villous atrophy, fissuring, and a mottled appearance. By using a variety of computerized techniques and assigning a weight or vote to each technique, it is possible to improve the detection of abnormal regions which are indicative of celiac disease, and of treatment progress in diagnosed patients. Herein a paradigm is suggested for improving the efficacy of automated methods for measuring celiac disease manifestation in the small intestinal mucosa. The suggestions are applicable to both standard and videocapsule endoscopic imaging, since both methods could potentially benefit from computerized quantitation to improve celiac disease diagnosis.

Application of neural networks in detection of abnormal brain perfusion regions

In this paper we modify the existing image processing schema for computed tomography perfusion (CTP) maps analysis in order to increase its efficiency in detection and classification of perfusion abnormalities by adding multilayer perceptron neural network diagnostic module and new neural network based image processing procedure. The main new contribution of this paper is description of self-organizing Kohonen's map (SOM) architecture that is used in the image processing step. CTP is based on generation of time series images that shows the flow of contrast material in vascular system of the brain. Despite the fact that SOM was previously reported as a reliable tool for time series classification we decided to utilize it as the extension of our existing methodology. We also present methodology of CTP – based diagnostic system for brain stroke diagnosis updated with our latest results and never before presented validation results. Our proposition is consisted of two steps: initial image processing after which potential asymmetry regions between hemispheres are detected and the second step during which position, type and prognostic map for potentially infarcted tissues are generated. We also made comparison of efficiency of our new approach with existing one. The test set for our algorithm validation was consisted of 75 CTP images triplets (one cerebral blood flow, cerebral blood volume and CT in each) from 30 patients (both man and woman) with suspicion of ischemia/stroke. The maps were previously diagnosed by an expert. Forty two CTP maps were described as normal (no abnormalities), 33 maps showed syndromes of ischemic strokes with various severities. Finally over 82% of CTP maps from our test set were rightly segmented and classified.