Digital Mammogram Tumor Preprocessing Segmentation Feature Extraction and Classification

Breast cancer continues to be a significant public health problem in the world. Early detection is the key for improving breast cancer prognosis. Mammography has been one of the most reliable methods for early detection of breast carcinomas. However, it is difficult for radiologists to provide both accurate and uniform evaluation for the enormous mammograms generated in widespread screening. The main objective of this paper is to enhance, detect and classify masses in digital mammogram. We develop a performance based case-based reasoning classification algorithm for mammographic findings to provide support for the clinical decision to perform biopsy of the breast. The developed classifier will be used for training and testing the images which is cancerous and noncancerous and improve the performance of the system.

Background: Alongside traditional imaging, current development in non-invasive breast carcinoma (BC) diagnosis, such as cell-free DNA-based technologies, have demonstrated promising potential in BC diagnosis. However, achieving early detection remains a challenge. Micronuclei (MN) are extranuclear bodies containing damaged chromosome segments indicative of genomic instability. Elevated levels of MN-containing erythrocytes have been studied in genotoxicity testing and cancer diagnosis. Here, we sequenced the isolated and purified micronuclei DNA (MN-DNA) from erythrocytes in 2ml peripheral blood using a method we previously developed (WO2021/228246 A1) and identified a series of distinct features in read counts that enabled to detect early-stage BC patients. Methods: We collected a clinical study cohort of 172 BC patients and 216 non-BC controls from 1-2ml peripheral blood. MN-DNA was isolated from erythrocytes and performed whole-genome sequencing. We selected distinctive tumor-associated MN-DNA features identified by genome-wide analysis to differentiate between controls and BC for KEGG pathway enrichment and constructed a classification model based on these features. Results: Genome-wide analysis of MN-DNA revealed significant regions between controls and BC patients. Based on covered genes, we observed significant enrichment in the estrogen signaling pathway. Utilizing these distinct MN- DNA features, we constructed a classification model that achieved an AUC of 96%, with a 95% specificity and 88% sensitivity in the test cohort. The model identified early BC (stage 0-I) and advanced BC (stage II-IV) with sensitivities of 100% or 84%, respectively. Conclusions: Our findings highlight that MN-DNA, a form of cytoplasmic DNA, provides a valuable tool for the accurate detection of BC, particularly in its early stages. The MN-DNA pattern is linked to the estrogen signaling pathway, offering novel insights and mechanisms that differ from current diagnostic approaches. Research sponsor: Timing Biotech. Early Detection of Primary Cancer and Relapse Citation Format: Wenjv Mo, Xingyun Yao, Haobo Sun, Honghao Liang, Jie Jin, Xiaowen Ding, Xiaofei Gao. Utilizing breast cancer-specific micronuclei DNA features from erythrocytes for early detection of breast carcinoma [abstract]. In: Proceedings of the AACR Special Conference: Liquid Biopsy: From Discovery to Clinical Implementation; 2024 Nov 13-16; San Diego, CA. Philadelphia (PA): AACR; Clin Cancer Res 2024;30(21_Suppl):Abstract nr A044.

Purpose:Breast cancer can be cured if diagnosed early, with digital mammography which is one of the most effective imaging modalities for early detection. However mammogram images often come with low contrast, high background noises and artifacts, making diagnosis difficult. The purpose of this research is to preprocess mammogram images to improve results with a computer aided diagnosis system. The focus is on three preprocessing methods: a breast border segmentation method; a contrast enhancement method; and a pectoral muscle removal method.Methods:The proposed breast border extraction method employs a threshold based segmentation technique along with a combination of morphological operations. The contrast enhancement method presented here is divided into two phages. In phase I, a bi-level histogram modification technique is applied to enhance the image globally and in phase II a non-linear filter based on local mean and local standard deviation for each pixel is applied to the histogram modified image. The pectoral muscle removal method discussed here is implemented by applying a region growing algorithm.Results:The proposed techniques are tested with the Mini MIAS dataset. The breast border extraction method is applied to 322 images and achieved 98.7% segmentation accuracy. The contrast enhancement method is evaluated based on quantitative measures like measure of enhancement, absolute mean brightness error, combined enhancement measure and discrete entropy. The proposed contrast enhancement method when applied to 14 images with different types of masses, the quantitative measures showed an optimum level of contrast enhancement compared to other enhancement methods with preservation of local detail. Removal of the pectoral muscle from MLO mammogram images reduced the search region while identifying abnormalities like masses and calcification.Conclusions:The preprocessing steps proposed here show promising results in terms of both qualitative and quantitative analysis.

Die effektive Brustkrebsfruherkennung muss innerhalb eines umfassenden, strukturveranderten und sektorubergreifenden Konzeptes erfolgen. In diesem Konzept ist die Mammographie nur ein Teil der eingesetzten Diagnostik. Die konsequente Dokumentation der mammadiagnostischen Versorgung im ambulanten Bereich muss mit den Daten des mammographischen Screenings und der stationaren Therapie verknupft werden. Im Rahmen der Brustkrebsfruherkennung ist die gemeinsame institutions- und sektorubergreifende Dokumentation einschlieslich der Daten aus Krebsregistern die Grundvoraussetzung fur ein effektives Handeln auf der Basis einer wissenschaftlich fundierten Evaluation der eingeschlagenen Wege. Ohne Dokumentation auf Krebsregisterbasis sind die Auswirkungen der Fruherkennung nicht mess- und steuerbar. Das in Deutschland initiierte Screening deckt nur einen Teilaspekt der nationalen Brustkrebsfruherkennung ab und musste im Hinblick auf eine Effektivitatssteigerung korrigiert werden.

Circulatory RNA based non-invasive detection and diagnosis of breast cancer

Octamer 4 (OCT4) has been postulated as a cancer stem cell biomarker. This study aimed to delineate the role and the clinicopathological implications of OCT4 expression in the development, progression, and lymph nodal metastatic potential of breast carcinoma. This study was conducted on a total number of 60 breast specimens. OCT4 immunohistochemical expression was assessed in 12 benign epithelial hyperplasia cases and 48 breast carcinoma cases with their adjacent carcinoma in-situ component and their corresponding metastatic lymph nodes (MLNs). OCT4 expression in breast carcinomas was significantly higher than that in hyperplastic lesions. Moderate agreement was found between OCT4 expression in invasive carcinoma and the adjacent in-situ component. In breast carcinomas, OCT4 was significantly correlated with larger tumor size, higher tumor grade, positive lymph node metastasis, high Ki67 index, and triple-negative subtype and inversely correlated with estrogen and progesterone status. By logistic regression, positive lymph node metastasis and high Ki67 index were found to be independently associated with OCT4 expression. Moderate significant agreement was recognized between OCT4 expression in the primary breast carcinomas and their matched metastatic lymph nodes. Thus, OCT4 might serve as an important biological marker for contemplating the carcinogenesis, progression, metastatic potential, and prognosis of breast carcinoma. Moreover, it could be an emerging tool for early detection of breast carcinoma and a forthcoming candidate for predicting lymph node metastasis.

Background:Breast lumps constitute a significant proportion of surgical cases in women of both developed and developing countries. The aim of this study is to look the frequency distribution of various breast lesions on fine needle aspiration (FNA).Materials and Methods:The 902 patients who presented with palpable breast lump, irrespective of age and sex were included in the study. Frequency distribution of various breast lesions with respect to age and sex was studied. Cytology grading in breast carcinoma was correlated in 69 cases with histology grading.Results:The majority (N = 871) of patients were females with maximum (N = 566) patients between second and third decade. The 773 patients had benign breast lesions and maximum (N = 341) patients were in the second decade. Fibroadenoma was the commonest benign lesion followed by fibrocystic change and mastitis. Out of 119 malignant breast lesions, 31.93% [N = 38] were between 41-50 years of age, 28.57% [N = 34] in 51-60 years and 22.68% [N = 27] in between 31-40 years of age. Out of 119 malignant breast lumps and majority were infiltrating ductal carcinoma (N = 108). Cytology grading correlated maximum with histology grade in Grade I followed by Grade II and Grade III.Conclusion:With experienced hands, FNA is safe, cost effective and a reliable technique for preoperative evaluation of palpable breast lumps. FNA features are more informative when combined with physical and radiology findings (Triple test). Fibroadenoma, fibrocystic change and mastitis form the major bulk of benign breast lesions. Epithelioid cells when seen in inflammatory breast FNA smears, tuberculosis must be ruled out. In India, breast carcinoma arises in younger patients as compared to western countries. Grading of breast carcinomas must be done on FNA smears for selecting neoadjuvent therapy. Clinical breast examination and mammography screening in females should be encouraged in developing countries from the third decade onwards for early detection of breast carcinoma.

Deep convolutional neural networks (CNNs) have emerged as a new paradigm for Mammogram diagnosis. Contemporary CNN-based computer-aided-diagnosis systems (CADs) for breast cancer directly extract latent features from input mammogram image and ignore the importance of morphological features. In this paper, we introduce a novel end-to-end deep learning framework for mammogram image processing, which computes mass segmentation and simultaneously predicts diagnosis results. Specifically, our method is constructed in a dual-path architecture that solves the mapping in a dual-problem manner, with an additional consideration of important shape and boundary knowledge. One path, called the Locality Preserving Learner (LPL), is devoted to hierarchically extracting and exploiting intrinsic features of the input. Whereas the other path, called the Conditional Graph Learner (CGL), focuses on generating geometrical features via modeling pixel-wise image to mask correlations. By integrating the two learners, both the cancer semantics and cancer representations are well learned, and the component learning paths in return complement each other, contributing an improvement to the mass segmentation and cancer classification problem at the same time. In addition, by integrating an automatic detection set-up, the DualCoreNet achieves fully automatic breast cancer diagnosis practically. Experimental results show that in benchmark DDSM dataset, DualCoreNet has outperformed other related works in both segmentation and classification tasks, achieving 92.27% DI coefficient and 0.85 AUC score. In another benchmark INbreast dataset, DualCoreNet achieves the best mammography segmentation (93.69% DI coefficient) and competitive classification performance (0.93 AUC score).

Mammography is the most customarily used prior procedures to diagnose breast cancer. The preprocessing step is essential to improve the contrast in the mammogram image, in this paper top-hat and bottom-hat transforms are employed. This paper proposes the feature extraction method using curvelet transform in digital mammogram to distinguish the normal and abnormal breast cancer. The curvelet handles curves by utilising only a small quantity of coefficients, and it can also perform the curve discontinuities. It is a multiscale directional transform that permits sparse representation of objects with edges. The pectoral muscles and labels have been removed by the selection of Region of Interest (ROI). The proposed work uses morphological transform for preprocessing, followed by application of curvelet transform to enhance and sharpen the images finally wrapping is done by using fast Fourier transform. The features were extracted from the curvelet coefficients by applying gray level co-occurrance matrix and features were also extracted from the regional properties of the preprocessed image. The experimental result gives 98% accuracy of the classification of cancer classes using SVM classifier. From the result, it indicates that a curvelet transformation is a capable tool for analysis of digital mammogram.

Feature Selection (FS) has become one of the most active research topics in the area of data mining. It performs to remove redundant and noisy features from high-dimensional data sets. A good feature selection has several advantages for a learning algorithm such as reducing computational cost, increasing its classification accuracy and improving result comprehensibility. In the supervised FS methods various feature subsets are evaluated using an evaluation function or metric to select only those features which are related to the decision classes of the data under consideration. However, for many data mining applications, decision class labels are often unknown or incomplete, thus indicating the significance of unsupervised feature selection. However, in unsupervised learning, decision class labels are not provided. The problem is that not all features are important. Some of the features may be redundant, and others may be irrelevant and noisy. In this paper, a novel unsupervised feature selection in mammogram image, using rough set based measures, is proposed. A typical mammogram image processing system generally consists of mammogram image acquisition, preprocessing of image, segmentation, features extracted from the segmented mammogram image. The proposed method is used to select features from data set, the method is compared with existing rough set based supervised feature selection methods and classification performance of both methods are recorded and demonstrates the efficiency of the method.

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 classified 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 module. 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 Algorithm (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.

Pectoral Muscle (PM), a significant region in Medio-Lateral Oblique (MLO) view of mammogram may adversely affect anomaly detection due to its resemblance to abnormal tissues. The removal of PM region can be considered as a prerequisite step for early breast cancer detection using mammographic images. The principal component of PM boundary component is extracted using the orientation and eccentricity property of Canny edge detected components of coarse mammographic image obtained after a multiscale decomposition technique using Laplacian Pyramid (LP). The principal component of PM boundary is extended to top and left boundaries using nearest neighbor approach. The algorithm was tested on images from the Mammographic Image Analysis Society (MIAS) database as well as mammograms obtained from a representative set of Indian populace provided by Lakeshore Hospital Kochi, India. On comparison with the PM boundary assessed by radiologists, the proposed method yielded an average false positive rate of 0.28%, average false negative rate of 3.67% and low Hausdorff distance for 83 images in mammographic database. Based on the performance analysis of the proposed algorithm, it is observed that 97% of images have an average error less than 3 mm which is promising.

This paper attempts to pinpoint different techniques for Pectoral Muscle (PM) segmentation, Microcalcification (MC) detection and classification in digital mammograms. The segmentation of PM and detection of MC and its classification are mostly based on image processing and data mining techniques. The review centered on major techniques in image processing and data mining that is employed for PM segmentation, MC detection and classification in digital mammograms. Breast cancer is one of the significant causes of death among women aged above 40. Mammography is considered the most successful means for prompt and timely detection of breast cancers. One notable visual indication of the malignant growth is the appearance of Masses, Architectural Distortions, and Microcalcification Clusters (MCCs). There are some disadvantages and hurdles for mankind viewers, and it is hard for radiologists to supply both precise and steady assessment for a large number of mammograms created in extensive screening. Computer Aided Detection has been employed to help radiologists in detecting MC and MCCs. The automatic recognition of malignant MCCs could be very helpful for diagnostic purpose. In this paper, we summarize the methods of automatic detection and classification of MCs in digitized mammograms. Pectoral muscle segmentation techniques are also summarized. The techniques used for segmentation of PM, MC detection and classification in a digitized mammogram are reviewed.

In the last decade, machine learning plays a vital role in the detection of breast cancer. Mammography is a proficient tool for early stage detection of breast cancer. In this work, a simple technique for breast cancer image classification in l mammogram images is proposed. Highly discriminant local binary patterns are extracted from the wavelet normalized mammogram images. K-nearest neighbor classifier is used to categorize the abnormal cancer cell images. A mammogram database is created to evaluate the efficacy of our algorithm. From the experimental results, the performance of our algorithms is comparatively good with very less computational time.

Over the past years many researchers proposed biomedical imaging methods for computer-aided detection and classification of suspicious regions in mammograms. Mammogram interpretation is performed by radiologists by visual inspection. The large volume of mammograms to be analyzed makes such readings labour intensive and often inaccurate. For this purpose, in this paper we propose a new unsupervised method to automatically detect suspicious regions in mammogram images. The method consists mainly of two steps: preprocessing; feature extraction and selection. Preprocessing steps allow to separate background region from the breast profile region. In greater detail, gray levels mapping transform and histogram specifications are used to enhance the visual representation of mammogram details. Then, local keypoints and descriptors such as SURF have been extracted in breast profile region. The extracted keypoints are filtered by proper parameters tuning to detect suspicious regions. The results, in terms of sensitivity and confidence interval are very encouraging.