Mammogram Classification with HanmanNets Using Hanman Transform Classifier

Breast cancer is a deadly disease and radiologists recommend mammography to detect it at the early stages. This paper presents two types of HanmanNets using the information set concept for the derivation of deep information set features from ResNet by modifying its kernel functions to yield Type-1 HanmanNets and then AlexNet, GoogLeNet and VGG-16 by changing their feature maps to yield Type-2 HanmanNets. The two types of HanmanNets exploit the final feature maps of these architectures in the generation of deep information set features from mammograms for their classification using the Hanman Transform Classifier. In this work, the characteristics of the abnormality present in the mammograms are captured using the above network architectures that help derive the features of HanmanNets based on information set concept and their performance is compared via the classification accuracies. The highest accuracy of 100% is achieved for the multi-class classifications on the mini-MIAS database thus surpassing the results in the literature. Validation of the results is done by the expert radiologists to show their clinical relevance.

Cancer Regions in Mammogram Images Using ANFIS Classifier Based Probability Histogram Segmentation Algorithm

Every year,the number of women affected by breast tumors is increasing worldwide.Hence,detecting and segmenting the cancer regions in mammogram images is important to prevent death in women patients due to breast cancer.The conventional methods obtained low sensitivity and specificity with cancer region segmentation accuracy.The high-resolution standard mammogram images were supported by conventional methods as one of the main drawbacks.The conventional methods mostly segmented the cancer regions in mammogram images concerning their exterior pixel boundaries.These drawbacks are resolved by the proposed cancer region detection methods stated in this paper.The mammogram images are clas-sified into normal,benign,and malignant types using the Adaptive Neuro-Fuzzy Inference System(ANFIS)approach in this paper.This mammogram classification process consists of a noise filtering module,spatial-frequency transformation module,feature computation module,and classification mod-ule.The Gaussian Filtering Algorithm(GFA)is used as the pixel smooth filtering method and the Ridgelet transform is used as the spatial-frequency transformation module.The statistical Ridgelet feature metrics are computed from the transformed coefficients and these values are classified by the ANFIS technique in this paper.Finally,Probability Histogram Segmentation Algo-rithm(PHSA)is proposed in this work to compute and segment the tumor pixels in the abnormal mammogram images.This proposed breast cancer detection approach is evaluated on the mammogram images in MIAS and DDSM datasets.From the extensive analysis of the proposed tumor detection methods stated in this work with other works,the proposed work significantly achieves a higher performance.The methodologies proposed in this paper can be used in breast cancer detection hospitals to assist the breast surgeon to detect and segment the cancer regions.

Hyperparameter Tuned Deep Hybrid Denoising Autoencoder Breast Cancer Classification on Digital Mammograms

Breast Cancer(BC)is considered the most commonly scrutinized can-cer in women worldwide,affecting one in eight women in a lifetime.Mammogra-phy screening becomes one such standard method that is helpful in identifying suspicious masses’malignancy of BC at an initial level.However,the prior iden-tification of masses in mammograms was still challenging for extremely dense and dense breast categories and needs an effective and automatic mechanisms for helping radiotherapists in diagnosis.Deep learning(DL)techniques were broadly utilized for medical imaging applications,particularly breast mass classi-fication.The advancements in the DL field paved the way for highly intellectual and self-reliant computer-aided diagnosis(CAD)systems since the learning cap-ability of Machine Learning(ML)techniques was constantly improving.This paper presents a new Hyperparameter Tuned Deep Hybrid Denoising Autoenco-der Breast Cancer Classification(HTDHDAE-BCC)on Digital Mammograms.The presented HTDHDAE-BCC model examines the mammogram images for the identification of BC.In the HTDHDAE-BCC model,the initial stage of image preprocessing is carried out using an average median filter.In addition,the deep convolutional neural network-based Inception v4 model is employed to generate feature vectors.The parameter tuning process uses the binary spider monkey opti-mization(BSMO)algorithm.The HTDHDAE-BCC model exploits chameleon swarm optimization(CSO)with the DHDAE model for BC classification.The experimental analysis of the HTDHDAE-BCC model is performed using the MIAS database.The experimental outcomes demonstrate the betterments of the HTDHDAE-BCC model over other recent approaches.

Micro Calcification Detection in Mammogram Images Using Contiguous Convolutional Neural Network Algorithm

The mortality rate decreases as the early detection of Breast Cancer(BC)methods are emerging very fast,and when the starting stage of BC is detected,it is curable.The early detection of the disease depends on the image processing techniques,and it is used to identify the disease easily and accurately,especially the micro calcifications are visible on mammography when they are 0.1 mm or bigger,and cancer cells are about 0.03 mm,which is crucial for identifying in the BC area.To achieve this micro calcification in the BC images,it is necessary to focus on the four main steps presented in this work.There are three significant stages of the process assigned to find the BC using a thermal image;the image processing procedures are described below.In the first stage of the process,the Gaussian filter technique is implemented to magnify the screening image.During the second stage,BC detection is separated from the pre-processed image.The Proposed Versatile K-means clustering(VKC)algorithm with segmentation is used to identify the BC detection form of the screening image.The centroids are then recalculated using proposed VKC,which takes the mean of all data points allocated to that centroid’s cluster,lowering the overall intracluster variance in comparison to the prior phase.The“means”in K-means refers to the process of averaging the data and determining a new centroid.This process eliminates unnecessary areas of interest.First,the mammogram screening image information is taken from the patient and begins with the Contiguous Convolutional Neural Network(CCNN)method.The proposed CCNN is used to classify the Micro calcification in the BC spot using the feature values is the fourth stage of the process.The assess the presence of high-definition digital infrared thermography technology and knowledge base and suggests that future diagnostic and treatment services in breast cancer imaging will be developed.The use of sophisticated CCNN techniques in thermography is being developed to attain a greater level of consistency.The implemented(CCNN)technique’s performance is examined with different classification parameters like Recall,Precision,F-measure and accuracy.Finally,the Breast Cancer stages will be classified based on the true positive and true negative values.

FAST SCREENING OUT TRUE NEGATIVE REGIONS FOR MICROCALCIFICATION DETECTION IN DIGITAL MAMMOGRAMS

A method is proposed to avoid complex computation in finding the region of interest （ROI） in a mammogram. In the method, the true negative region （TNR） definitely containing no microcalcification clusters （MCCs） is screened out, thus obtaining ROIs, The strategy consists of three steps： （1） the mammogram is partitioned into a set of non-overlapping blocks with an equal size, and for each block, five statistical features are computed, （2） negative blocks are screened out by the threshold method through rough analyses, （3） the more accurate analysis is done by the cost-sensitive support vector machine to eliminate the block definitely containing no MCCs, Experimental results on real mammograms show that 81.71% of TNRs can be screened out by the proposed method.

Automated Deep Learning Empowered Breast Cancer Diagnosis UsingBiomedical Mammogram Images

Biomedical image processing is a hot research topic which helps to majorly assist the disease diagnostic process.At the same time,breast cancer becomes the deadliest disease among women and can be detected by the use of different imaging techniques.Digital mammograms can be used for the earlier identification and diagnostic of breast cancer to minimize the death rate.But the proper identification of breast cancer has mainly relied on the mammography findings and results to increased false positives.For resolving the issues of false positives of breast cancer diagnosis,this paper presents an automated deep learning based breast cancer diagnosis(ADL-BCD)model using digital mammograms.The goal of the ADL-BCD technique is to properly detect the existence of breast lesions using digital mammograms.The proposed model involves Gaussian filter based pre-processing and Tsallis entropy based image segmentation.In addition,Deep Convolutional Neural Network based Residual Network(ResNet 34)is applied for feature extraction purposes.Specifically,a hyper parameter tuning process using chimp optimization algorithm(COA)is applied to tune the parameters involved in ResNet 34 model.The wavelet neural network(WNN)is used for the classification of digital mammograms for the detection of breast cancer.The ADL-BCD method is evaluated using a benchmark dataset and the results are analyzed under several performance measures.The simulation outcome indicated that the ADL-BCD model outperforms the state of art methods in terms of different measures.

Mammogram Images Thresholding for Breast Cancer Detection Using Different Thresholding Methods

The purpose of this study is to apply different thresholding in mammogram images, and then we will determine which technique is the best in thresholding (extraction) malignant and benign tumors from the rest breast tissues. The used technique is Otsu method, because it is one of the most effective methods for most real world views with regard to uniformity and shape measures. Also, we present all the thresholding methods that used the concept of between class variance. We found from the experimental results that all the used thresholding techniques work well in detection normal breast tissues. But in abnormal tissues (breast tumors), we found that only neighborhood valley emphasis method gave best detection of malignant tumors. Also, the results demonstrate that variance and intensity contrast technique is the best in extraction the micro calcifications which represent the first signs of breast cancer.

Enhancement of Microcalcifications Based on Fractal Techniques in Mammograms

Microcalcification clusters in mammograms are an important early sign of breast cancer. The enhancement of mieroealcifications in mammograms is one of the most important preprocessing techniques for the extraction of cluster mierocalcifications. In this paper, we present a novel method for the enhancement of microcalcifications. Firstly, the initial microcaleification edges were extracted by using kirsch edge operator, and the diseontinouse edges were linked by employing fi＇aetal teehnique, Then, the continuous closed edges of microcalcifications were filled by using seed filling algorithm. The pixel values of the filled region were replaced by the corresponding pixel values in the original image. Finally, the enhancement of microcalcifications in mammograms was achieved by adding the filled image to the original image. We evaluated the performance of our algorithm by using 50 regions of interesting （ROIs） with microcalcification clusters from DDSM database. The experiment results demonstrate that our CAD system can give better enhancement effect compared with other methods.

A Computer Aided Consultant System for Mammogram Diagnosis

A computer aided consultant system for mammogram diagnosis is proposed in this paper based on mammogram segmentation as an image mining technique, to aid radiologistis in X ray film interpretation. The general architecture of the system is introduced first, followed by a discussion of mammogram segmentation using logic filter, an analysis of the statistical data to the diagnostics with respect to different clinical information, and a brief introduction to a fuzzy decision making subsystem. Finally some experimental results are given.

Fuzzy-Rough Feature Selection for Mammogram Classification

Feature selection（FS） refers to the process of selecting those input attributes that are most predictive of a given outcome. Unlike other dimensionality reduction methods,feature selectors preserve the original meaning of the features after reduction. The benefits of FS are twofold：it considerably decreases the running time of the induction algorithm,and increases the accuracy of the resulting model. This paper analyses the FS process in mammogram classification using fuzzy logic and rough set theory. Rough set and fuzzy logic based Quickreduct algorithms are applied for the FS from the features extracted using gray level co-occurence matrix（GLCM） constructed over the mammogram region. The predictive accuracy of the features is tested using NaiveBayes,Ripper,C4.5,and ant-miner algorithms. The results show that the ant-miner produces significant result comparing with others and the number of features selected using fuzzy-rough quick reduct algorithm is minimum,too.

Assessment of a Database Management System in a Mammogram Unit of a Radiologic Department in Fann Teaching Hospital (Senegal)

Objectives: To evaluate a medical data management system of a mammogram unit in a department of Radiology. Methods: This is a qualitative and quantitative assessment study in Fann Teaching Hospital between April 2014 and June 2015 one year after its implementation. The quantitative component consisted of the audit of the database to determine the socio-demographic characteristics of patients and the results of mammograms. The qualitative component assessed users’ experience. For analysis, quantitative data were extracted and transferred to Microsoft Excel. For scale variables, we calculated the averages and extremes. For qualitative variables, we established percentages. Results: During the study period, 433 patients underwent mammograms. The average age of patients was 48 years. The completion rate maintained above 85% was below 26% in the first two months of use. As to the completeness given examinations, it was still above 83%. The results of mammogram examinations were normal in the majority of cases: 96% for the right breast and 95.2% for the left breast. All users had a favourable opinion about the database. The reasons were better work organization, comprehensiveness, accessibility and standardization of information about the patient and especially the immediate availability of statistics. For 60% of these health professionals, complaints related to the use of the software were the time-consuming of filling data. Conclusion: This study mainly describes the perception of health professionals on the computerization of radiological examinations. It offers some advantages, proposes improvements and opens avenues for reflection on the globalization of the computerization of patient records in Radiology.

Simply Fine-Tuned Deep Learning-Based Classification for Breast Cancer with Mammograms

A lump growing in the breast may be referred to as a breast mass related to the tumor.However,not all tumors are cancerous or malignant.Breast masses can cause discomfort and pain,depending on the size and texture of the breast.With an appropriate diagnosis,non-cancerous breast masses can be diagnosed earlier to prevent their cultivation from being malignant.With the development of the artificial neural network,the deep discriminative model,such as a convolutional neural network,may evaluate the breast lesion to distinguish benign and malignant cancers frommammogram breast masses images.This work accomplished breastmasses classification relative to benign and malignant cancers using a digital database for screening mammography image datasets.A residual neural network 50(ResNet50)model along with an adaptive gradient algorithm,adaptive moment estimation,and stochastic gradient descent optimizers,as well as data augmentations and fine-tuning methods,were implemented.In addition,a learning rate scheduler and 5-fold cross-validation were applied with 60 training procedures to determine the best models.The results of training accuracy,p-value,test accuracy,area under the curve,sensitivity,precision,F1-score,specificity,and kappa for adaptive gradient algorithm 25%,75%,100%,and stochastic gradient descent 100%fine-tunings indicate that the classifier is feasible for categorizing breast cancer into benign and malignant from the mammographic breast masses images.

Extracting and smoothing contours in mammograms using Fourier descriptors

Contour is an important pattern descriptor in image processing and particularly in region description, registration and length estimation. In many applications where contour is used, a good segmentation and an efficient smoothing method are needed. In X-ray images, such as mammograms, where object edge is not clearly discernible, estimating the object’s contour may yield substantial shift along the boundary due to noise or segmentation drawbacks. An appropriate smoothing is therefore required to reduce these effects. In this paper, an approach based on local adaptive threshold segmentation to extract contour and a new smoothing approach founded on Fourier descriptors are introduced. The experimental results of extraction obtained from a set of mammograms and compared with the breast regions delineated by radiologists yielded a percent overlap area of 98.7% ± 0.9% with false positive and negative rates of 0.36 ± 0.74 and 0.93 ± 0.44 respectively. The proposed method was tested on a set of images and improved the accuracy, leading to an average error of less than one pixel.

Digital Mammogram Inferencing System Using Intuitionistic Fuzzy Theory

In the medical field,the detection of breast cancer may be a mysterious task.Physicians must deduce a conclusion from a significantly vague knowledge base.A mammogram can offer early diagnosis at a low cost if the breasts'satisfactory mammogram images are analyzed.A multi-decision Intuitionistic Fuzzy Evidential Reasoning(IFER)approach is introduced in this paper to deal with imprecise mammogram classification efficiently.The proposed IFER approach combines intuitionistic trapezoidal fuzzy numbers and inclusion measures to improve representation and reasoning accuracy.The results of the proposed technique are approved through simulation.The simulation is created utilizing MATLAB software.The screening results are classified and finally grouped into three categories:normal,malignant,and benign.Simulation results show that this IFER method performs classification with accuracy almost 95%compared to the already existing algorithms.The IFER mammography provides high accuracy in providing early diagnosis,and it is a convenient diagnostic tool for physicians.

Ultrasound Alongside with Mammogram in Women with Physically Dense Breast

We report usefulness of ultrasound used as an adjunct diagnostic tool to mammogram in routine annual checkup for women breasts of certain ages and breast mass. The purpose of breast imaging is to detect areas of tissue distortion and breast cancers. A mammogram is the common diagnostic imaging modality used to find breast diseases but sometimes the mammogram might not give the doctor enough information especially in women with dense breasts. As a result, the patient may be asked to undergo ultrasound or magnetic resonance imaging as a better mean of judgment to the case. Because ultrasound is widely used, simple and safe to patients we were encouraged to emphasis on exploring its role adjunct to mammogram. A retrospective observation study was done at the diagnostic radiology department at King Abdulaziz University Hospital (KAUH) in the period from January 2012 to June 2012;we covered all women with dense breasts in mammography and ultrasound units. The study group was 40 patients. All patients were imaged with both mammography and ultrasound. The statistical measures of accuracy, sensitivity and specificity were calculated using the SPSS program. The results we obtained suggest that age and the physical density of breast potentially affect mammogram images of women with 41 years or smaller with sensitivity 66% and specificity 68%. Therefore, we recommend using ultrasound alongside the mammogram?in women with dense breast for better diagnosis of small cancers that were not identified on mammography or clinical breast examination alone.

Automatic Detection of Stellate Lesions in Digital Mammograms Using Multi-scale SIFT

This paper presents a novel automatic mammography recognition approach used to develop computer-aided diagnostic systems that require a robust method to assist the radiologist in identifying and recognizing speculations from a multitude of lines corresponding to the normal fibrous breast tissue.Following this rationale,this paper introduces a novel approach for detecting the speculated lesions in digital mammograms based on multi-scale SIFT(scale-invariant feature transform)orientations.The proposed method starts by estimating a set of key points that best represent the image mammography in a scale space.We then benefit from SIFT algorithm to locally characterize each key point by assigning a consistent orientation.Thereafter,a set of three features are extracted for each pixel in the image mammogram based on these orientations.The extracted features are fed into BDT(binary decision tree)in order to perform per pixel classification and decide whether the pixel is normal or abnormal.We evaluate the proposed system on BCDR(breast cancer digital repository)database and the experimental results show that our method is accurate with 97.95%recognition rate,while it is robust to illumination changes,rotation and scale variations.

Spatial Attention Integrated EfficientNet Architecture for Breast Cancer Classification with Explainable AI

Breast cancer is a type of cancer responsible for higher mortality rates among women.The cruelty of breast cancer always requires a promising approach for its earlier detection.In light of this,the proposed research leverages the representation ability of pretrained EfficientNet-B0 model and the classification ability of the XGBoost model for the binary classification of breast tumors.In addition,the above transfer learning model is modified in such a way that it will focus more on tumor cells in the input mammogram.Accordingly,the work proposed an EfficientNet-B0 having a Spatial Attention Layer with XGBoost(ESA-XGBNet)for binary classification of mammograms.For this,the work is trained,tested,and validated using original and augmented mammogram images of three public datasets namely CBIS-DDSM,INbreast,and MIAS databases.Maximumclassification accuracy of 97.585%(CBISDDSM),98.255%(INbreast),and 98.91%(MIAS)is obtained using the proposed ESA-XGBNet architecture as compared with the existing models.Furthermore,the decision-making of the proposed ESA-XGBNet architecture is visualized and validated using the Attention Guided GradCAM-based Explainable AI technique.

A Logic Filter for Tumor Detection on Mammograms

This paper presents a novel approach for detection of suspicious regions in digitized mammograms. The edges of the suspicious region in mammogram are enhanced using an improved logic filter. The result of further image processing gives a gray-level histogram with distinguished characteristics, which facilitates the segmentation of the suspicious masses. The experiment results based on 25 digital sample mammograms, which are definitely diagnosed, are analyzed and evaluated briefly.

融合自注意力的乳腺钼靶图像特征引导分割算法

为提高对乳腺癌钼靶图像中病灶区域的识别精度,本研究设计了一种面向乳腺肿块和钙化区域分割的特征引导注意网络。首先,该网络通过特征提取模块学习乳腺组织的语义特征;其次,利用融合自校正注意力的解码模块,增强对病灶区域边缘信息的关注度,提高边界的清晰度;最后,采用特征引导注意模块增强通道的依赖关系,进一步还原病灶区域边缘细节,提高分割精度。实验结果表明,本研究网络在扩充后的INBreast1数据库中肿块和钙化分割的平均骰子系数(mDice)分别达到了0.971和0.888,在DDSM数据库肿块分割的mDice达到了0.911,优于其他常规的分割模型,对乳腺癌的早期诊断和治疗具有重要意义。

基于Attention U-Net的乳腺X线图像微钙化检测模型的临床应用

目的:通过开发基于Attention U-Net的乳腺X线图像微钙化检测模型,实现微钙化的高效率检出,并探究不同性质钙化、不同乳腺密度对该深度学习模型微钙化检测性能的影响。方法:回顾性分析接受乳腺常规X线检查的347例患者的694幅图像。通过低年资医师独立阅片,高年资医师审核的方式,建立微钙化检出的参考标准。进行神经网络训练,建立深度学习模型。以钙化面积和数量分别计算,并采用精确率、召回率、F1分数、交并比等指标评估微钙化检测性能,分析不同性质钙化(良性vs恶性)、不同乳腺密度(a+b类vs c+d类)对深度学习模型微钙化检测性能的影响。结果:深度学习模型对微钙化检测的精确率为85.12%±18.39%(以钙化面积计算)和76.72%±19.85%(以钙化数量计算);召回率为78.18%±19.25%(以钙化面积计算)和85.12%±18.39%(以钙化数量计算);交并比为68.29%±21.39%(以钙化面积计算)和67.13%±23.84%(以钙化数量计算);F1分数为78.96%±17.70%(以钙化面积计算)和77.65%±9.37%(以钙化数量计算)。深度学习模型在不同钙化性质(良性vs恶性)中的精确率、召回率、交并比、F1分数之间差异均无统计学意义(P>0.05),在不同乳腺密度(a+b类vs c+d类)中对微钙化检测的精确率、召回率、交并比、F1分数之间差异无统计学意义(P>0.05)。结论:基于Attention U-Net的乳腺X线图像微钙化检测模型能够对乳腺微钙化进行有效的检测、有助于乳腺微钙化的定量研究,同时该模型稳定性强,钙化性质及乳腺密度对该模型的检测性能无影响。