Abnormal Breast Tissue Research Articles

Multimodal Photoacoustic/Elastography Imaging for the Detection of Acute Radiation Dermatitis in Breast Radiation Therapy

Purpose: This study evaluates whether photoacoustic (PA), sound touch elastography (STE), and viscoelasticity (VE) can distinguish between normal and abnormal post-radiotherapy breast skin and compare these methods with RTOG criteria at a 20 Gy threshold. Methods: Patients meeting inclusion and exclusion criteria underwent PA, STE, and VE on the same day. Collected data included radiation dose, molecular type, RTOG, Fitzpatrick skin type, pathology, neoadjuvant chemotherapy status, TNM classification, surgical procedures, primary breast cancer location, BMI, and age. A sample of 41 patients was determined using a two-sample t-test. Statistical tools like T-tests, variance analysis, rank sum tests, and chi-square tests, along with random forest analysis and ROC curves, were used to evaluate the radiation dose effects. Results: Data from 66 patients showed significant differences in parameters such as dermis and subcutaneous tissue oxygen saturation, dermal thickness, and skin elasticity (p-values < 0.05). However, minimum values of oxygen saturation and some photoacoustic measures were not significantly different. Notably, at a 20 Gy radiation threshold, significant variations in oxygen saturation, abnormal dermal thickness, and skin STE and VE were observed, proving more accurate than RTOG grading. Conclusion: The study confirms the effectiveness of PA and elastography imaging in differentiating normal from abnormal breast tissue and assessing radiation-induced changes, highlighting the potential of these imaging techniques.

Detection of invasive ductal carcinoma by electrical impedance spectroscopy implementing gaussian relaxation-time distribution (EIS-GRTD)

Breast cancer detection and differentiation of breast tissues are critical for accurate diagnosis and treatment planning. This study addresses the challenge of distinguishing between invasive ductal carcinoma (IDC), normal glandular breast tissues (nGBT), and adipose tissue using electrical impedance spectroscopy combined with Gaussian relaxation-time distribution (EIS-GRTD). The primary objective is to investigate the relaxation-time characteristics of these tissues and their potential to differentiate between normal and abnormal breast tissues. We applied a single-point EIS-GRTD measurement to ten mastectomy specimens across a frequency range f = 4 Hz to 5 MHz. The method calculates the differential ratio of the relaxation-time distribution function Δγ between IDC and nGBT, which is denoted by ΔγIDC−nGBT, and Δγ between IDC and adipose tissues, which is denoted by ΔγIDC−adipose. As a result, the differential ratio of Δγ between IDC and nGBT ΔγIDC−nGBT is 0.36, and between IDC and adipose ΔγIDC−adipose is 0.27, which included in the α -dispersion at τpeak1=0.033±0.001s. In all specimens, the relaxation-time distribution function γ of IDC γIDC is higher, and there is no intersection with γ of nGBT γnGBT and adipose γadipose. The difference in γ suggests potential variations in relaxation properties at the molecular or structural level within each breast tissue that contribute to the overall relaxation response. The average mean percentage error δ for IDC, nGBT, and adipose tissues are 5.90%, 6.33%, and 8.07%, respectively, demonstrating the model’s accuracy and reliability. This study provides novel insights into the use of relaxation-time characteristic for differentiating breast tissue types, offering potential advancements in diagnosis methods. Future research will focus on correlating EIS-GRTD finding with pathological results from the same test sites to further validate the method’s efficacy.

Optical Emission Spectroscopy for the Real-Time Identification of Malignant Breast Tissue

Breast conserving resection with free margins is the gold standard treatment for early breast cancer recommended by guidelines worldwide. Therefore, reliable discrimination between normal and malignant tissue at the resection margins is essential. In this study, normal and abnormal tissue samples from breast cancer patients were characterized ex vivo by optical emission spectroscopy (OES) based on ionized atoms and molecules generated during electrosurgical treatment. The aim of the study was to determine spectroscopic features which are typical for healthy and neoplastic breast tissue allowing for future real-time tissue differentiation and margin assessment during breast cancer surgery. A total of 972 spectra generated by electrosurgical sparking on normal and abnormal tissue were used for support vector classifier (SVC) training. Specific spectroscopic features were selected for the classification of tissues in the included breast cancer patients. The average classification accuracy for all patients was 96.9%. Normal and abnormal breast tissue could be differentiated with a mean sensitivity of 94.8%, a specificity of 99.0%, a positive predictive value (PPV) of 99.1% and a negative predictive value (NPV) of 96.1%. For 66.6% patients all classifications reached 100%. Based on this convincing data, a future clinical application of OES-based tissue differentiation in breast cancer surgery seems to be feasible.

Genetic algorithm based detection of breast cancer using least square-support vector machine classifier

Breast tumors are a dangerous disease among women worldwide. They are the second leading cause of death among all forms of cancers in women. Their early detection is critical to increasing the survival rate of women. Mammography is a reliable screening technique in the early detection of abnormal breast tissue severity. Radiologist abnormalities in the breast tissue, radiologists employ mammography. However, detecting breast abnormalities through digital diagnostic techniques by a radiologist could be time consuming. Consequently, computerized studying of digital mammography has emerged via the development of CAD systems. Several CAD systems have been developed for breast cancer detection. However, obtaining a satisfactory performance of CAD systems is a challenging task. We propose a CAD architecture for the classification of breast tissues as either benign or malignant using an LS-SVM classifier with various kernels namely linear, quadratic, polynomial, MLP, and RBF kernels. From the experimental outputs, it is clear that GA based LS-SVM classifier with RBF kernel outputs classification accuracy of 94.59% for normal/abnormal case classification is better, when it is compared with all other kernels. It is also stated that GA based LS-SVM classifier with RBF kernel produces a better classification accuracy of 98.26% for benign/malignant case classification when it is compared with other reported works.

98. Incidental Findings of Malignant and Pre-malignant Lesions in Gender-Affirming Mastectomies: A Retrospective Review with Recommendations for Post-operative Management

INTRODUCTION: Gender-affirming mastectomy is the most common gender-affirming operation performed in the United States. Given expanding insurance coverage and access, there are increasing reports of incidental pathologic findings of breast cancer and high-risk lesions on routine surgical specimens. While these incidental findings have been described in the literature, there is little guidance regarding treatment and subsequent follow-up. In this retrospective review, we describe our incidental findings with a focus on post-operative treatment and ongoing surveillance. METHODS: A retrospective review of gender-affirming mastectomy patients from 2007 to 2022 at a single institution was performed to identify the incidence of abnormal perioperative breast tissue pathology. For patients with atypical or malignant lesions, a case series was performed to evaluate demographics, personal and family history of cancer, genetic risk factors, mammography, smoking history, medical comorbidities, reproductive health history, testosterone use, oncologic treatment course, and ongoing surveillance. RESULTS: Four hundred and sixty-six patients over a fifteen-year period underwent gender-affirming mastectomy with tissue submitted for pathologic evaluation. Incidental atypical or malignant lesions were identified in nine patients (1.9%). Individual treatment and surveillance plans varied. All patients with malignant pathology (one with invasive ductal carcinoma (IDC) and two with ductal carcinoma in situ (DCIS)) were referred to surgical oncology at our institution. The patient with IDC underwent axillary workup with imaging and active treatment with chemotherapy. Two patients with DCIS did not require axillary staging, radiation, or chemotherapy, and are undergoing active surveillance. Among the patients with atypical findings (four with atypical ductal hyperplasia (ADH), two with atypical lobular hyperplasia (ALH)), there was variation in referral and treatment plans. Surveillance recommendations ranged from monthly self-exams and annual clinical exams to no recorded surveillance recommendations. CONCLUSION: In this retrospective review, the overall incidence of high risk and malignant lesions is low. While we found there was consistent treatment offered to patients with malignant findings, there was variation in follow-up and surveillance recommendations for atypical lesions. As we perform gender-affirming mastectomy with increasing pace, it is essential to understand breast cancer risk, treatment, and long-term screening. Further studies are needed for accurately characterizing risk to inform surgical decision making, patient counseling for treatment and surveillance, and long-term screening guidelines.

Spectrum of histopathological patterns of breast lesions in a tertiary care hospital

Background: Breast lesions are areas of abnormal breast tissue. It is estimated that 20% of women develop breast lesions. Breast lesions are either benign or malignant. Breast carcinomas are the most common cause of deaths in women accounting for 10.6% deaths in India, 6.9% deaths per year globally. Aims and Objectives: To study various histopathological patterns of breast lesions in women admitted in Government General Hospital. Materials and Methods: It is a prospective study in the department of pathology, Government General Hospital, Nizamabad, from December 2020 to July 2022. Grossing was done and details noted, tissue bits were processed and sections were stained with conventional H and E Staining. Results: Out of 70 cases of breast lesions, 55 (78.5%) are benign breast lesions and 15 (21.5%) are malignant breast lesions. Fibroadenoma is the most common breast lesion accounting for 33 cases, that is, 60 %, followed by fibrocystic diseases (21.8%). Infiltrating ductal carcinoma of No Special Type is the most common malignant breast lesion accounting for 10 cases, that is, 66.6%. Among invasive breast carcinomas, the majority of cases are Grade II (69%). Conclusion: Study of histopathological patterns of breast lesions plays an important role in diagnosis, treatment, and prognosis of breast lesions. This study highlighted incidence and pathological characteristics of a wide range of breast lesions.

Modeling light propagation in a reconstructed breast model using the Monte Carlo method

Breast cancer is one of the most common types of cancer that causes the highest death among women. One of the main aspects of effectively monitoring the effectiveness of breast cancer treatment is the early detection of abnormalities in breast tissue. It is one of the most critical factors that may determine how well breast cancer therapy is carried out. On the other hand, the techniques used in breast diagnostics today have several limitations, including the hazards connected with radiation exposure, the high price, and the likelihood of creating false-positive results. Therefore, optical techniques are considered feasible strategies for the early diagnosis of breast abnormalities. The optical-based technique has been studied to develop self-diagnostic aids and high-tech devices such as optical tomography. The recent development of therapeutic applications using low-power light also creates a new direction in treating benign tumors through photobiological effects. Therefore, much more research must be conducted to understand how light interacts with breast tissue for developing the diagnosis device and treatment device. In this study, light interaction with tissue was investigated by simulating light propagation at wavelengths of 650 nm, 800 nm, and 950 nm in a breast model that was generated using an MRI data collection. Specifically, the tissue's absorption of these three wavelengths was evaluated to understand better how light interacts with breast tissue. The results of the simulation show that normal breast models and sick breast models have significantly different total quantities of energy absorbed, and these results indicate this discrepancy. The distribution map acquired correlates with the image obtained using optical transillumination imaging. The simulation results can steer further research toward developing a novel breast diagnostics methodology based on optical approaches.

Enhancement of Mammographic Images Using Histogram-Based Techniques for Their Classification Using CNN

In the world, one in eight women will develop breast cancer. Men can also develop it, but less frequently. This condition starts with uncontrolled cell division brought on by a change in the genes that regulate cell division and growth, which leads to the development of a nodule or tumour. These tumours can be either benign, which poses no health risk, or malignant, also known as cancerous, which puts patients' lives in jeopardy and has the potential to spread. The most common way to diagnose this problem is via mammograms. This kind of examination enables the detection of abnormalities in breast tissue, such as masses and microcalcifications, which are thought to be indicators of the presence of disease. This study aims to determine how histogram-based image enhancement methods affect the classification of mammograms into five groups: benign calcifications, benign masses, malignant calcifications, malignant masses, and healthy tissue, as determined by a CAD system of automatic mammography classification using convolutional neural networks. Both Contrast-limited Adaptive Histogram Equalization (CAHE) and Histogram Intensity Windowing (HIW) will be used (CLAHE). By improving the contrast between the image's background, fibrous tissue, dense tissue, and sick tissue, which includes microcalcifications and masses, the mammography histogram is modified using these procedures. In order to help neural networks, learn, the contrast has been increased to make it easier to distinguish between various types of tissue. The proportion of correctly classified images could rise with this technique. Using Deep Convolutional Neural Networks, a model was developed that allows classifying different types of lesions. The model achieved an accuracy of 62%, based on mini-MIAS data. The final goal of the project is the creation of an update algorithm that will be incorporated into the CAD system and will enhance the automatic identification and categorization of microcalcifications and masses. As a result, it would be possible to increase the possibility of early disease identification, which is important because early discovery increases the likelihood of a cure to almost 100%.

Hybrid Detection of Breast Abnormalities Based on Contrast Agents: Introducing a Proof of Concept from a Physics Perspective.

This manuscript attempts to present a proof of concept from a physics perspective of a hybrid detective system based on the utilization of contrast agents with the purpose of indicating breast tissue abnormalities. In the present concept, the photon-counting module of the detector is set up to the K-characteristic radiation emitted by the contrast agent. Τwo X-ray spectra were used: 40 kV- W/Al (1.6 mm) and 50 kV- W/Al (1.6 mm) with additional filtration of 0.3 mm Gd. Iodine (I) contrast agent was studied as a ‘‘fingerprint’’ for tissue abnormality indication. A computational Monte Carlo model, based on previously published validated tabulated data and tissue experimental measurements, was developed with the purpose of showing that the present concept has practical potential; however, with a weakness of not being accompanied by experimental validation. The study considered two types of internal tissue layers (fibrous/tumor with thickness values of 0.2–2.5 mm) within an external layer of fat tissue (4 and 8 cm). Quantitative (number of encountered K-photons) and qualitative (tumor–fibrous ratio) advantages of using X-ray spectra of a higher tube voltage (50 kV) and of counting the Κα photons were found. In addition, the quantitative and qualitative benefits were correspondingly more dominant at high (2.5 mm) and low (0.2 mm) tissue thickness values. In conclusion, by utilizing suitable contrast agents as ‘‘fingerprint’’ tissue abnormalities, the acquisition of combined morphological and functional imaging features (through the counting of K-X-rays) could enhance breast imaging in its present form and lead to advanced prognostic capabilities of breast abnormalities.

Diagnostic Reference Levels in Mammography in the Asian Context

Background: Breast cancer is the most frequent cancer among the female population globally. Therefore, early detection is helpful for effective treatments and to reduce the mortality rate. Mammography is a radiological examination done with low-energy X-rays to detect abnormalities in breast tissue. This study aims to review the literature to evaluate the techniques, protocols, and conversion factors used to determine the diagnostic reference levels (DRLs); within the Asian continent using both phantom- and patient-based data.&#x0D; Methods: Related articles were systematically reviewed via Pub Med, Google scholar, and freehand search with the aid of relevant terms. Related abstracts in English were screened, and suitable articles were selected after reviewing the full-text. Four hundred and thirty abstracts were screened for relevance, and 12 articles were selected.&#x0D; Results: The study comprises four phantom-based and eight patient-based studies. The studies varied between the types of test subjects, conversion factors, breast compression thickness, and dose calculation protocols. This obstructs continuing the DRLs with the updates and comparisons among countries. Establishments of DRLs in Asian countries are less than the rest of the world. DRLs should be measured continuously, and should be updated based on other clinical parameters of the patients.&#x0D; Conclusion: DRLs in mammography were measured from time to time in different geographical locations in Asia by following various techniques. But when compared with the other regions of the world, there is less consideration for establishing DRLs in Asia. There should be standard protocols and updated conversion factors according to the advancements of the technology to ensure radiation protection with optimal absorbed dose with appropriate image quality.

A deep learning approach for automatic detection, segmentation and classification of breast lesions from thermal images

PurposeThis study investigates implementation of deep learning (DL) approaches to breast tumor recognition based on thermal images. We propose to utilize Mask R-CNN technique on images by first assigning bounding boxes and then creating a border for each tumor volume to differentiate it from adjacent tissues and structures. In this manner, thermal images can be handled by a single DL model to successfully perform detection, classification, and segmentation of normal and abnormal breast tissues. MethodsThis study employs Mask R-CNN technique along with transfer learning models to accurately delineate breast volumes from adjacent tissues and structures. It is a novel study that uses a single DL model to carry out three steps of breast tumor diagnosis, namely detection, segmentation and classification of normal and abnormal tissues based on thermal images. ResultsTwo network architectures (ResNet-50 and ResNet-101) were trained for 60 epochs and were then evaluated based on their classification and segmentation performances. The testing process resulted in higher classification success for ResNet-50 backbone pretrained on COCO images (%97.1 accuracy). Detection and segmentation performances of this model were also higher with mAP of 0.921 and overlap score of 0.868. ConclusionsOur results indicate that the classification and segmentation performances of Mask R-CNN method on ResNet-50 architecture are better than the data reported in the literature for thermal breast image studies. It should be emphasized that our approach employs a single DL model that successfully performs detection, classification, and segmentation procedures for diagnosing normal and abnormal breast tissues.

Meta-Heuristic Algorithm-Tuned Neural Network for Breast Cancer Diagnosis Using Ultrasound Images.

Breast cancer is the most menacing cancer among all types of cancer in women around the globe. Early diagnosis is the only way to increase the treatment options which then decreases the death rate and increases the chance of survival in patients. However, it is a challenging task to differentiate abnormal breast tissues from normal tissues because of their structure and unclear boundaries. Therefore, early and accurate diagnosis and classification of breast lesions into malignant or benign lesions is an active domain of research. Over the decade, numerous artificial neural network (ANN)-based techniques were adopted in order to diagnose and classify breast cancer due to the unique characteristics of learning key features from complex data via a training process. However, these schemes have limitations like slow convergence and longer training time. To address the above mentioned issues, this paper employs a meta-heuristic algorithm for tuning the parameters of the neural network. The main novelty of this work is the computer-aided diagnosis scheme for detecting abnormalities in breast ultrasound images by integrating a wavelet neural network (WNN) and the grey wolf optimization (GWO) algorithm. Here, breast ultrasound (US) images are preprocessed with a sigmoid filter followed by interference-based despeckling and then by anisotropic diffusion. The automatic segmentation algorithm is adopted to extract the region of interest, and subsequently morphological and texture features are computed. Finally, the GWO-tuned WNN is exploited to accomplish the classification task. The classification performance of the proposed scheme is validated on 346 ultrasound images. Efficiency of the proposed methodology is evaluated by computing the confusion matrix and receiver operating characteristic (ROC) curve. Numerical analysis revealed that the proposed work can yield higher classification accuracy when compared to the prevailing methods and thereby proves its potential in effective breast tumor detection and classification. The proposed GWO-WNN method (98%) gives better accuracy than other methods like SOM-SVM (87.5), LOFA-SVM (93.62%), MBA-RF (96.85%), and BAS-BPNN (96.3%)

Deep learning model for fully automated breast cancer detection system from thermograms.

Breast cancer is one of the most common diseases among women worldwide. It is considered one of the leading causes of death among women. Therefore, early detection is necessary to save lives. Thermography imaging is an effective diagnostic technique which is used for breast cancer detection with the help of infrared technology. In this paper, we propose a fully automatic breast cancer detection system. First, U-Net network is used to automatically extract and isolate the breast area from the rest of the body which behaves as noise during the breast cancer detection model. Second, we propose a two-class deep learning model, which is trained from scratch for the classification of normal and abnormal breast tissues from thermal images. Also, it is used to extract more characteristics from the dataset that is helpful in training the network and improve the efficiency of the classification process. The proposed system is evaluated using real data (A benchmark, database (DMR-IR)) and achieved accuracy = 99.33%, sensitivity = 100% and specificity = 98.67%. The proposed system is expected to be a helpful tool for physicians in clinical use.

An automatic and efficient technique for tumor location identification and classification through breast MR images

AimThe basic objective of this paper is to develop a single structured algorithm to classify the breast tissues into normal or abnormal. Material & MethodologyFor this study, the breast MR images dataset of 448 images collected from Healthmap diagnostics centre, PGIMS, Rohtak, India. The proposed algorithm consists of several steps i.e. an integrated (Median Wiener & Median) filtering technique is used for de-noising; breast boundary region extraction via selection of nipple and mid- sternum points to make the image rotation invariant; determined the tumor region intensity by using morphological operations & hole filling; classify the normal and abnormal breast tissues by SVM using 14 texture features extracted through GLCM & 13 morphological or kinetic features; evaluated the exact location as well as area of abnormal tissues. ResultsThe proposed algorithm has been evaluated statistically as well as visually. The quality parameters achieved are accuracy, sensitivity and specificity with values 0.937, 0.956 and 0.872 respectively. The Jaccard Index coefficient achieved is 0.921, which indicates promising overlap between the predicted tumor and the manually done image by the radiologist so called ground truth image. ConclusionThis work may be taken as a second opinion by the radiologists. The evaluated results may give a basic foundation for optimization by selecting the features more precisely and also different evolutionary algorithms using multi-classifiers can be designed in future.

A Contrast Enhancement Model for X-Ray Mammograms Using Modified Local s-Curve Transformation Based on Multi-Objective Optimization

The biased X-ray field and the subtle difference in X-ray attenuation between normal and abnormal breast tissues prevent the biomedical sensors to generate mammograms with good quality. These are the common imperfections in the acquisition systems (sensor) of Mammography which contribute towards the degradation of mammogram image quality. This paper aims to develop a contrast enhancement model for mammograms so that the human or machine vision can distinguish easily among the variants of breast lesions and predict their severity. The proposed mammogram enhancement model exploits locally the proprieties of the sigmoidal function based on multi-objective Genetic Algorithm for improving the contrast of lesions along with a simultaneous increment in various Image Quality Assessment (IQA) parameters like: $\textit {EME}$ , $\textit {EC}$ , $\textit {AMBEn}$ and $\textit {FSIM}$ . Herein, the multi-objective optimization problem is transformed into a single-objective optimization to find a unique solution for mammogram enhancement. An increase in these IQA metrics is indicative of better contrast, edge content, and conservation of brightness and other diagnostics information. Simulations of the proposed enhancement model are carried out on mammograms from the mini-MIAS database for benchmarking and validation of results.

An introduction to the Cyrcadia Breast Monitor: A wearable breast health monitoring device

BackgroundThe most common breast cancer detection modalities are generally limited by radiation exposure, discomfort, high costs, inter-observer variabilities in image interpretation, and low sensitivity in detecting cancer in dense breast tissue. Therefore, there is a clear need for an affordable and effective adjunct modality that can address these limitations. The Cyrcadia Breast Monitor (CBM) is a non-invasive, non-compressive, and non-radiogenic wearable device developed as an adjunct to current modalities to assist in the detection of breast tissue abnormalities in any type of breast tissue. MethodsThe CBM records thermodynamic metabolic data from the breast skin surface over a period of time using two wearable biometric patches consisting of eight sensors each and a data recording device. The acquired multi-dimensional temperature time series data are analyzed to determine the presence of breast tissue abnormalities. The objective of this paper is to present the scientific background of CBM and also to describe the history around the design and development of the technology. ResultsThe results of using the CBM device in the initial clinical studies are also presented. Twenty four-hour long breast skin temperature circadian rhythm data was collected from 93 benign and 108 malignant female study subjects in the initial clinical studies. The predictive model developed using these datasets could differentiate benign and malignant lesions with 78% accuracy, 83.6% sensitivity and 71.5% specificity. A pilot study of 173 female study subjects is underway, in order to validate this predictive model in an independent test population. ConclusionsThe results from the initial studies indicate that the CBM may be valuable for breast health monitoring under physician supervision for confirmation of any abnormal changes, potentially prior to other methods, such as, biopsies. Studies are being conducted and planned to validate the technology and also to evaluate its ability as an adjunct breast health monitoring device for identifying abnormalities in difficult-to-diagnose dense breast tissue.

Thermal Imaging Techniques for Breast Screening - A Survey.

Breast cancer is the second leading cause of cancer death among women preceded by cervix cancer. It has been reported that at the early stage of detection there is 85% chance of getting cured, whereas only 10% chance at later stage diagnosis. The current screening modalities are expensive, they have intricate imaging measures and they are unhealthy due to radiation exposure. Therefore, a screening tool that is non-invasive, has no connection with the body, free from radiation, such as Medical Thermography is necessary. It is reported that the sensitivity and specificity of medical thermography are high largely in dense breast tissues. The clinical interpretation primarily depends on the asymmetrical analysis of these thermograms subjectively. The appearance of an asymmetric thermal image may indicate the pathological conditions. For earlier detection of breast cancer, it is essential to identify the advanced methods in image processing techniques which enhance the significance of diagnostics. In that analysis, the required breast region is unglued from the background image. The segmented image is separated into symmetrical left and right breast tissues. The statistical and histogram features extracted from both regions are used to identify the abnormal thermograms using machine learning algorithms. From literature, it is reported that the thermal images are inherently low contrast images and have low single to noise ratio. Moreover, they are amorphous in nature and no clear edges are seen. The difficulty lies in the detection of lower breast boundaries and inframammary folds. So, in general, the first attempt is made in improving the signal to noise ratio and contrast of the image which helps to extract the true regions of breast tissues. Then, asymmetry analysis of the normal and abnormal breast tissues is performed using different techniques. This work demonstrates the review of a few image processing methods or the development which are elaborated in the detection of breast cancer from thermal images.

Pseudocolor enhancement of mammogram texture abnormalities

We present a novel method for enhancing texture irregularities, both lesions and microcalcifications, in digital X-ray mammograms. It can be implemented in computer-aided diagnostic systems to help improve radiologists’ diagnosis precision. The method provides three different outputs aimed at enhancing three different sizes of mammogram abnormalities. Our approach uses a two-dimensional adaptive causal autoregressive texture model to represent local texture characteristics. Based on these, we enhance suspicious breast tissue abnormalities, such as microcalcifications and masses, to make signs of developing cancer better visually discernible. We extract over 200 local textural features from different frequency bands, which are then combined into a single multichannel image using the Karhunen–Loeve transform. We propose an extension to existing contrast measures for the evaluation of contrast around regions of interest. Our method was extensively tested on the INbreast database and compared both visually and numerically with three state-of-the-art enhancement methods, with favorable results.

Proposal of index to estimate breast similarities in thermograms using fuzzy C means and anisotropic diffusion filter based fuzzy C means clustering

Clustering analysis of medical infrared images is found to be very effective for extracting local information, which helps to identify the abnormalities associated with region of interest. Breast tissue extraction in breast thermograms is challenging due to its low signal-to-noise ratio, low contrast and absence of clear edges. Accurate detection of asymmetrical thermal patterns in breast cancer screening is essential for better diagnosis. Therefore, in this work, the limitations are addressed using Anisotropic Diffusion filter based Fuzzy-C-Means clustering (ADFCM). This technique is employed for precise segmentation of hottest regions in breast tissues. Initially, the ground truth masks are multiplied with raw images for the extraction of non-breast tissues. The corresponding left and right breasts are delineated. Anisotropic diffusion filtering is performed on delineated breast tissues and simple thresholding edge map output is integrated with FCM clustering. For comparison, conventional FCM clustering is employed to segment hottest region. The grey level values of hottest regions are extracted by multiplying the clustered output with input image. The mean area of hottest region gray level values is computed for normal and abnormal images. The performance comparison of FCM and ADFCM has been carried out qualitatively and quantitatively. The results show that ADFCM performs very well compared to FCM method. Further, it is observed that asymmetry analysis of breast tissues using ADFCM based clustering is found to be statistically highly significant (ρ < 0.0001) thereby efficiently discriminate normal and abnormal breast tissues. Hottest Area Index (HAI) is found to be very effective for identifying normal and abnormal conditions in breast cancer diagnosis. Therefore, the proposed pipeline of work could enhance the diagnostic relevance of thermal images and assist in the early diagnosis of breast abnormalities.

Multi-scale CNN based on region proposals for efficient breast abnormality recognition

Mammographic pattern recognition is one of the most essential tasks in breast cancer diagnosis, and has been studied for several years now to make it suitable and faster. In this paper, we developed a novel deep Convolutional Neural Network (CNN) approach to discriminate normal from abnormal breast tissues using Gaussian pyramid representation for multi-scale analysis (Pyramid-CNN). In order to improve image processing time, we extracted representative region proposals from each mammogram using determinant of the Hessian operator. To improve performance of our model and avoid overfitting, data augmentation techniques based on geometric transformation and sub-histogram equalization were applied on all regions to increase the variance of significant mammographic samples. We evaluated our methodology on the publicly available mammography dataset such as Breast Cancer Digital Repository (BCDR) database. In comparison with the current state-of-the-art methods, the experiments show that our proposed system provides efficient results, achieving the average accuracy of 96.84%, sensitivity of 92.12%, specificity of 98.02%, precision of 92.15%, F1-score of 92.12%, and area under the receiver operating characteristic curve (AUC) of 96.76%. Hence, the study demonstrates that our proposed approach has the potential to significantly improve the conventional recognition and classification strategies for use in advanced clinical application and practice or in general, biomedical imaging field.