scholarly journals Detection and Classification of Cancer from Microscopic Biopsy Images Using Clinically Significant and Biologically Interpretable Features

2015 ◽  
Vol 2015 ◽  
pp. 1-14 ◽  
Author(s):  
Rajesh Kumar ◽  
Rajeev Srivastava ◽  
Subodh Srivastava
Keyword(s):  
Texture Features ◽  
Histogram Equalization ◽  
Gray Level ◽  
Segmentation Methods ◽  
Segmented Images ◽  
Clinically Significant ◽  
Classification Of Images ◽  
Nearest Neighborhood ◽  
Neighborhood Method

A framework for automated detection and classification of cancer from microscopic biopsy images using clinically significant and biologically interpretable features is proposed and examined. The various stages involved in the proposed methodology include enhancement of microscopic images, segmentation of background cells, features extraction, and finally the classification. An appropriate and efficient method is employed in each of the design steps of the proposed framework after making a comparative analysis of commonly used method in each category. For highlighting the details of the tissue and structures, the contrast limited adaptive histogram equalization approach is used. For the segmentation of background cells, k-means segmentation algorithm is used because it performs better in comparison to other commonly used segmentation methods. In feature extraction phase, it is proposed to extract various biologically interpretable and clinically significant shapes as well as morphology based features from the segmented images. These include gray level texture features, color based features, color gray level texture features, Law’s Texture Energy based features, Tamura’s features, and wavelet features. Finally, the K-nearest neighborhood method is used for classification of images into normal and cancerous categories because it is performing better in comparison to other commonly used methods for this application. The performance of the proposed framework is evaluated using well-known parameters for four fundamental tissues (connective, epithelial, muscular, and nervous) of randomly selected 1000 microscopic biopsy images.

scholarly journals A Comparative Study on Application of Computer Vision and Fluorescence Imaging Spectroscopy for Detection of Huanglongbing Citrus Disease in the USA and Brazil

2013 ◽  
Vol 2013 ◽  
pp. 1-6 ◽  
Author(s):  
Caio B. Wetterich ◽  
Ratnesh Kumar ◽  
Sindhuja Sankaran ◽  
José Belasque Junior ◽  
Reza Ehsani ◽  
...  
Keyword(s):  
Computer Vision ◽  
Fluorescence Imaging ◽  
Imaging Spectroscopy ◽  
Texture Features ◽  
Support Vector ◽  
Infected Leaf ◽  
Segmented Images ◽  
The Usa ◽  
Citrus Disease

The overall objective of this work was to develop and evaluate computer vision and machine learning technique for classification of Huanglongbing-(HLB)-infected and healthy leaves using fluorescence imaging spectroscopy. The fluorescence images were segmented using normalized graph cut, and texture features were extracted from the segmented images using cooccurrence matrix. The extracted features were used as an input into the classifier, support vector machine (SVM). The classification results were evaluated based on classification accuracies and number of false positives and false negatives. The results indicated that the SVM could classify HLB-infected leaf fluorescence intensities with up to 90% classification accuracy. Though the fluorescence intensities from leaves collected in Brazil and the USA were different, the method shows potential for detecting HLB.

Recognition and Classification of Food Grains, Fruits and Flowers Using Machine Vision

2009 ◽  
Vol 5 (4) ◽  
Author(s):  
Dayanand G Savakar ◽  
Basavaraj S Anami
Keyword(s):  
Machine Vision ◽  
Vision System ◽  
Texture Features ◽  
Morphological Features ◽  
Machine Vision System ◽  
Food Grains ◽  
Colour Features ◽  
Classification Of Images ◽  
Color Texture

In this paper, we have presented different methodologies devised for recognition and classification of images of agricultural/horticultural produce. A classifier based on BPNN is developed which uses the color, texture and morphological features to recognize and classify the different agricultural/horticultural produce. Even though these features have given different accuracies in isolation for varieties of food grains, mangoes and jasmine flowers, the combination of features proved to be very effective. The average recognition and classification accuracies using colour features are 87.5%, 78.4% and 75.7% for food grains, mango and jasmine flowers, respectively, and the average accuracies have increased to 90.8%, 80.2% and 85.8% for food grains, mangoes and jasmine flowers ,respectively, using texture features. The average accuracies have increased to 94.1%, 84.0% and 90.1% for food grains, mangoes and jasmine flowers, respectively. The results are encouraging and promise a good machine vision system in the area of recognition and classification of agricultural/horticultural produce.

scholarly journals Prediction of Clinically Significant Cancer Using Radiomics Features of Pre-Biopsy of Multiparametric MRI in Men Suspected of Prostate Cancer

Cancers ◽  
2021 ◽  
Vol 13 (24) ◽  
pp. 6199
Author(s):  
Chidozie N. Ogbonnaya ◽  
Xinyu Zhang ◽  
Basim S. O. Alsaedi ◽  
Norman Pratt ◽  
Yilong Zhang ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Quantitative Imaging ◽  
Spatial Relationship ◽  
Region Of Interest ◽  
Texture Features ◽  
Gray Level ◽  
Discriminative Ability ◽  
Significant Difference ◽  
Clinically Significant ◽  
Occurrence Matrix

Background: Texture features based on the spatial relationship of pixels, known as the gray-level co-occurrence matrix (GLCM), may play an important role in providing the accurate classification of suspected prostate cancer. The purpose of this study was to use quantitative imaging parameters of pre-biopsy multiparametric magnetic resonance imaging (mpMRI) for the prediction of clinically significant prostate cancer. Methods: This was a prospective study, recruiting 200 men suspected of having prostate cancer. Participants were imaged using a protocol-based 3T MRI in the pre-biopsy setting. Radiomics parameters were extracted from the T2WI and ADC texture features of the gray-level co-occurrence matrix were delineated from the region of interest. Radical prostatectomy histopathology was used as a reference standard. A Kruskal–Wallis test was applied first to identify the significant radiomic features between the three groups of Gleason scores (i.e., G1, G2 and G3). Subsequently, the Holm–Bonferroni method was applied to correct and control the probability of false rejections. We compared the probability of correctly predicting significant prostate cancer between the explanatory GLCM radiomic features, PIRADS and PSAD, using the area under the receiver operation characteristic curves. Results: We identified the significant difference in radiomic features between the three groups of Gleason scores. In total, 12 features out of 22 radiomics features correlated with the Gleason groups. Our model demonstrated excellent discriminative ability (C-statistic = 0.901, 95%CI 0.859–0.943). When comparing the probability of correctly predicting significant prostate cancer between explanatory GLCM radiomic features (Sum Variance T2WI, Sum Entropy T2WI, Difference Variance T2WI, Entropy ADC and Difference Variance ADC), PSAD and PIRADS via area under the ROC curve, radiomic features were 35.0% and 34.4% more successful than PIRADS and PSAD, respectively, in correctly predicting significant prostate cancer in our patients (p < 0.001). The Sum Entropy T2WI score had the greatest impact followed by the Sum Variance T2WI. Conclusion: Quantitative GLCM texture analyses of pre-biopsy MRI has the potential to be used as a non-invasive imaging technique to predict clinically significant cancer in men suspected of having prostate cancer.

Bridge Target Detection in SAR Images Based on Texture Feature

2013 ◽  
Vol 347-350 ◽  
pp. 3634-3638 ◽  
Author(s):  
Nan Zheng ◽  
Wei Zheng ◽  
Zhong Lin Xu ◽  
Da Cheng Wang
Keyword(s):  
False Alarm ◽  
Target Detection ◽  
Texture Feature ◽  
Texture Features ◽  
Histogram Equalization ◽  
Experimental Results ◽  
Gray Level ◽  
Contrastive Analysis ◽  
Sar Images ◽  
Occurrence Matrix

This paper carries out an algorithm research on bridge target detection in SAR images and presents a method that combines both texture features and correlation features. The method firstly extracts initial targets by using the algorithm of histogram equalization segmentation, and then conducts a contrastive analysis for targets and their surrounding background textures by using the gray level co-occurrence matrix to get rid of the false alarm target. The experimental results show that the method is simple, effective and has certain algorithm robustness.

scholarly journals Segmentation and Classification of Mammogram into Normal and Abnormal using Texture Features

2020 ◽  
Vol 9 (1) ◽  
pp. 1005-1008
Keyword(s):  
Breast Cancer ◽  
Early Stage ◽  
Signal To Noise Ratio ◽  
Texture Features ◽  
Histogram Equalization ◽  
K Nearest Neighbors ◽  
Region Extraction ◽  
Analysis Society ◽  
Occurrence Matrix

Breast cancer is known to be a fatal disease since decades in women worldwide. Mammography is an effective tool used for the detection of breast cancer in the early stage. Computer aided tools helps medical field by ruling out the false identification of cancer cells in mammograms. Breast region extraction and classification of the extracted region into normal and abnormal is a crucial step in mammographic based diagnosis of breast cancer. Hence, in the proposed paper a method for segmentation of breast region and classification of breast region is presented. Breast region extraction is performed using Otsu’s thresholding method and intensity adjustments, enhancement is performed by Contrast Limited Adaptive Histogram Equalization (CLAHE). Gray Level Co-Occurrence Matrix (GLCM), Histogram of Oriented Gradients (HOG), Local Binary Pattern (LBP) features are extracted to classify the breast region using K-Nearest Neighbors (KNN) classifier. The proposed algorithm is tested on Mammographic Image Analysis Society (MIAS) dataset, obtained minimum Root Mean Square Error (RMSE) and maximum Peak Signal-to-Noise Ratio (PSNR). For classification, 80.12% of accuracy is obtained with TPR and FPR of about 0.8317 and 0.3412 respectively

scholarly journals GENERALIZATION OF THE COOCCURRENCE MATRIX FOR COLOUR IMAGES: APPLICATION TO COLOUR TEXTURE CLASSIFICATION

2011 ◽  
Vol 23 (1) ◽  
pp. 63 ◽  
Author(s):  
Vincent Arvis ◽  
Christophe Debain ◽  
Michel Berducat ◽  
Albert Benassi
Keyword(s):  
Texture Classification ◽  
Computation Time ◽  
Texture Features ◽  
Grey Scale ◽  
Colour Image ◽  
Good Classification ◽  
Real Time Applications ◽  
Classification Of Images ◽  
Multispectral Method

Three different approaches to colour texture analysis are tested on the classification of images from the VisTex and Outex databases. All the methods tested are based on extensions of the cooccurrence matrix method. The first method is a multispectral extension since cooccurrence matrices are computed both between and within the colour bands. The second uses joint colour-texture features: colour features are added to grey scale texture features in the entry of the classifier. The last uses grey scale texture features computed on a previously quantized colour image. Results show that the multispectral method gives the best percentages of good classification (VisTex: 97.9%, Outex: 94.9%). The joint colour-texture method is not far from it (VisTex: 96.8%, Outex: 91.0%), but the quantization method is not very good (VisTex:83.6%, Outex:68.4%). Each method is decomposed to try to understand each one deeper, and computation time is estimated to show that multispectral method is fast enough to be used in most real time applications.

scholarly journals Detección de Neumonía viral y bacteriana en imágenes de rayos x utilizando redes neuronales artificiales

2019 ◽  
pp. 9-16
Author(s):  
Itzel Guadalupe Guerrero-Gasca ◽  
Juan Israel Yañez-Vargas ◽  
Joel Quintanilla-Domínguez ◽  
Luis Rey Lara-González ◽  
Arturo Gasca-Ortega
Keyword(s):  
Image Processing ◽  
Bacterial Pneumonia ◽  
Confusion Matrix ◽  
X Rays ◽  
Segmented Images ◽  
Artificial Neural ◽  
Classification Of Images ◽  
Occurrence Matrix ◽  
Processing Techniques

This paper presents the experiments and results on the viral and bacterial pneumonia identification, which were obtained by means of image processing techniques and artificial neural networks. The objective of this research is to reduce the patient’s waiting time to obtain the result of the x-rays diagnosis of a pulmonary disease of pneumonia. At the time of this writing, pneumonia is considered the most common cause of infant mortality in the world, responsible for 15% of all deaths in children under 5 years. To obtain the classifier model we start from the detection in the pulmonary region through digital image processing and obtaining the characteristics in the segmented images, discriminating against those that provide a diagnosis through Gray Level Co-occurrence Matrix (GLCM). Finally, those features are used as the description in the classification of images such as: healthy, viral pneumonia and bacterial pneumonia. We use a total of eight features: autocorrelation, contrast, cluster prominence, variance cluster shade, sum of entropy, difference of entropy and number of pixels. These characteristics were used to model and train an artificial neural network Backpropagation, obtaining results that are presented in their confusion matrix along with the accuracy percentage obtained.

DEEP CONVOLUTIONAL NEURAL STRATEGY FOR DETECTION AND PREDICTION OF MELANOMA SKIN CANCER

2021 ◽  
Vol 33 (01) ◽  
pp. 2050045
Author(s):  
Ravi Dandu ◽  
Jayakameshwaraiah ◽  
Y. B. Ravi Kumar
Keyword(s):  
Learning Strategy ◽  
Research Work ◽  
Texture Features ◽  
Representation Learning ◽  
Features Extraction ◽  
Blue Channel ◽  
Segmentation Accuracy ◽  
Channel Information ◽  
Classification Of Images

The research work has focused on detection and prediction of melanoma which is done by subjecting to features extraction, where the features of an image consisting of melanoma regions are detected by analysis and this analysis is done by considering the features like color and texture-based features learning strategy. These features are extracted by combining color and texture-based features extraction with deep convolutional features representation learning strategy. The colors of images are extracted by representing the colors of different channels into red, green and blue channel information. The combination of texture features extraction with color-based features extraction in addition to Alex net features extraction learning has made the system more robust and efficient toward the segmentation and classification of images. Further, the erected method involves convoluting the features of extracted information with color and texture-based method which has led our system to full convolution neural networks with images features extraction. The melanoma is detected and segmented with watershed segmentation, these segmented features are subjected to the proposed features extraction method, where the features are extracted by combining the methods of texture with color-based information. These colors are made available to the proposed method by analyzing the regions of melanoma images. The erected method does the task of features extraction by Weber law descriptors in combination with red, green, blue channels information extracted from features representation learning. The proposed method has yielded an accuracy of 94.12% of segmentation accuracy and a classification accuracy of 94.32% with respect to various other classification techniques.

An Enhanced Technique for Classification of Fruits using Shape Color and Texture Features

2017 ◽  
Vol 7 (7) ◽  
pp. 408
Author(s):  
Yashpal Jitarwal ◽  
Tabrej Ahamad Khan ◽  
Pawan Mangal
Keyword(s):  
Image Processing ◽  
Texture Features ◽  
Automatic Classification ◽  
Processing Technique ◽  
Image Processing Technique ◽  
Advance Technique ◽  
Time Required ◽  
Fruit Sorting ◽  
Human Inspection

In earlier times fruits were sorted manually and it was very time consuming and laborious task. Human sorted the fruits of the basis of shape, size and color. Time taken by human to sort the fruits is very large therefore to reduce the time and to increase the accuracy, an automatic classification of fruits comes into existence.To improve this human inspection and reduce time required for fruit sorting an advance technique is developed that accepts information about fruits from their images, and is called as Image Processing Technique.

Sign in / Sign up

Export Citation Format

Share Document