Narrow Results

The classification of tumoral masses and normal breast tissue is targeted. A mass detection algorithm which does not refer explicitly to shape, border, size, contrast or texture of mammographic suspicious regions is evaluated. In the present approach, classification features are embodied by the image representation used to encode suspicious regions. Classification is performed by means of a support vector machine (SVM) classifier. To investigate whether improvements can be achieved with respect to a previously proposed overcomplete wavelet image representation, a pixel and a discrete wavelet image representations are developed and tested. Evaluation is performed by extracting 6000 suspicious regions from the digital database for screening mammography (DDSM) collected by the University of South Florida (USF). More specifically, 1000 regions representing biopsy-proven tumoral masses (either benign or malignant) and 5000 regions representing normal breast tissue are extracted. Results demonstrate very high performance levels. The area A_z under the receiver operating characteristic (ROC) curve reaches values of 0.973 ± 0.002, 0.948 ± 0.004 and 0.956 ± 0.003 for the pixel, discrete wavelet and overcomplete wavelet image representations, respectively. In particular, the improvement in the A_z value with the pixel image representation is statistically significant compared to that obtained with the discrete wavelet and overcomplete wavelet image representations (two-tailed p-value < 0.0001). Additionally, 90% true positive fraction (TPF) values are achieved with false positive fraction (FPF) values of 6%, 11% and 7%, respectively.

Computer support for early detection of breast cancer requires a proper mimicking of the way radiologists compare mammographic images; by comparing bilateral (images of the left and right breasts) and temporal images. In this paper, one method for bilateral registration and intensity normalization and two methods for difference analysis are described. The bilateral registration is based on anatomical features and assumptions of how the female breast is deformed under compression. The first method for differential analysis is based on the absolute difference between the registered images while the second method is based on statistical differences between properties of corresponding neighborhoods. The methods are tested on images from the MIAS database (on 100 images with 59 abnormalities distributed over four types) and evaluated by FROC-analysis. The performances of the two methods are similar but the statistical method gives better performance at a lower false positive rate and is better in particular for detecting asymmetrical developments.

Breast cancer has been reported to be the first deadly disease that affects women worldwide. This type of cancer has been reported to be the second leading cause of death in women worldwide. Medical reports have also reported that every woman is exposed to having breast cancer with an average probability of about 12%. It has also been reported to be the most common cancer that affects women. Fatality could be due to the cancer detection delay; in other words, early detection of the tumor can increase the survival rate of patients. Routine techniques of imaging modalities for cancer screening such as Mammography, Computated Tomography (CT) scan, Magnetic Resonance Imaging (MRI) and ultrasound are impractical tools for many reasons such as the irreproducible nature, the high error rate in cases of thick breasts, the pain and the annoyance they cause. Consequently, there is a need for more convincing strategies with high accuracy rates in breast cancer detection. Therefore, among the large variety of medical breast scanning techniques, thermography has attracted attention in applications related to detection and diagnosis. It is capable of providing helpful and useful information about the physiological variations and accordingly, it can detect tumors even in early stages. In addition, it is a very safe scanning tool, so as many needed tests can be held in proper time and manner. Thermography relies on the fact that human body temperature generally is a natural norm for the diagnosis of diseases. Thermography in medical applications applies infrared body examination tool which is fast, noninvasive, noncontact, pain free, radiation free and flexible to monitor the temperature of the human body. The fundamental principle of thermography relies on physiology such as the distribution of temperature on the skin surface. Infrared thermography scanning for breasts is an imaging technique which essentially searches for temperature change in human body. Temperature variance could be considered as a good indicator of tumor occurrence in the scanned area. Tumor mainly causes a noteworthy increase in blood vessel circulation and metabolic activity, so it causes higher radiations emitted from the human body around the regions of tumor. The paper surveys the literature work conducted in the field of breast cancer detection from thermogram scans. The survey is followed by a discussion of the strengths and weaknesses of thermography-based tumor detection. A new research idea and some considerations are then suggested based on that discussion to achieve better results in this critical area.

Mammographic screening programmes generate large numbers of highly variable, complex images, most of which are unequivocally normal. When present, abnormalities may be small or subtle. Two processes critical to the success of screening programmes are the perception of potential abnormalities and the subsequent analy-sis of each detected lesion to determine its clinical significance. The consequences of errors are costly, and in many screening centres, films are read by two radiologists in an attempt to reduce errors. The prime objective of our research is to improve the accuracy of the detection and analysis of breast lesions by providing radiologists with computer-aided digital image analysis tools. In this paper we focus on the detection and analysis of mammographic microcalcifications.

We describe a philosophy of research aimed at generating useful computer-based aids for radiologists. Firstly, it is necessary to accurately identify specific tasks which are difficult for the human observer. Having correctly identified a problem, appropriate computer vision methods must be developed and their performance evaluated. It is then important to determine effective ways of using such methods to aid radiologists, and it is essential to prove that the effect on radiologists’ performance is entirely beneficial.

We present results of experiments to determine factors affecting radiologists’ perception of microcalcifications, and to investigate the effects of attention-cueing on detection performance. Our results show that radiologists’ performance can be significantly improved with the use of prompts generated from automatically-detected microcalcification clusters.

We describe a new method for the delineation of mammographic abnormalities based on the analysis of multiple high quality X-ray projections of excised lesions. Biopsy specimens are secured inside a rigid tetrahedron, the edges of which provide a reference frame to which the locations of features can be related. A three-dimensional representation of an abnormality can be formed and rotated to resemble its appearance in the original mammogram.

A framework for computer-aided analysis of mammograms is described. General computer vision algorithms are combined with application specific procedures in a hierarchical fashion. The system is under development and is currently limited to detection of a few types of suspicious areas.

The image features are extracted by using feature extraction methods where wavelet techniques are utilized. A low-pass pyramid representation of the image is convolved with a number of quadrature filters. The filter outputs are combined according to simple local Fourier domain models into parameters describing the local neighbourhood with respect to the model. This produces estimates for each pixel describing local size, orientation, Fourier phase, and shape with confidence measures associated to each parameter.

Tentative object descriptions are then extracted from the pixel-based features by application-specific procedures with knowledge of relevant structures in mammograms. The orientation, relative brightness and shape of the object are obtained by selection of the pixel feature estimates which best describe the object.

The list of object descriptions is examined by procedures, where each procedure corresponds to a specific type of suspicious area, e.g. clusters of microcalcifications.

In this paper, multi-resolution analysis of two edge-texture based descriptors, Discriminative Robust Local Binary Pattern (DRlbp) and Discriminative Robust Local Ternary Pattern (DRltp), are proposed for the determination of mammographic masses as benign or malignant. As an extension of Local Binary Pattern (LBP) and Local Ternary Pattern (LTP), DRlbp and LTP-based features overcome the drawbacks of these features preserving the edge information along with texture. With the hypothesis that multi-resolution analysis of these features for different regions related to mammaographic masses with wavelet transform will capture more discriminating patterns and thus can help in characterizing masses. In order to evaluate the efficiency of the proposed approach, several experiments are carried out using the mini-MIAS database where a 5-fold cross validation technique is incorporated with Support Vector Machine (SVM) on the optimal set of features obtained via stepwise logistic regression method. An area under the receiver operating characteristic (ROC) curve ($A_z$ value) of 0.96 is achieved with DRlbp attributes as the best performance. The superiority of the proposed scheme is established by comparing the obtained results with recently developed other competing schemes.

Discrimination between malignant tumors and benign masses in mammograms can be difficult because of the diversity of tumor shape. In order to facilitate this discrimination, we propose a method of shape analysis based only on characterization of boundaries of tumors via parabolic modeling. The method relies on the observation that most benign masses possess smooth or macro-lobulated contours, while most malignant tumors have complex contours with spiculations, concavities, and micro-lobulations. The contours used in this work were drawn on digitized mammograms by an expert radiologist (JELD). Classification of 54 tumors as benign or malignant was achieved with an accuracy of 76 percent.

CSIRO Develops Statistical Tools for Cancer Diagnosis.

Childhood Stress Linked to Early Puberty in Girls.

Sonography More Accurate than Mammography.

New Genes Related to Parkinson’s Disease Discovered.

China to Use Microbial Genome Technology for Oil Extraction.

Curcumin May Prevent Alcohol-Related Liver Disease.

Reliance Life Starts Trials on Bio-Engineered Skin Products.

Japanese Scientists Use Mouth Cells for Treatment of Damaged Eyes.

New Anti-Mad Cow Disease Enzyme Discovered.

Breakthrough Study in Congenital Birth Abnormalities.

Faster Healing with New Drug.

Singapore's Jurong Technologies to Unveil Revolutionary Breast Cancer Screening System in Asia.

Ultrasound-Mediated Gene Therapy - Philips and Glygenix Team Up.

Vietnam Healthcare - The Next Growth Frontier?

New Ultrasound System Could Lower Number of Biopsies.

SINGAPORE – New Material Kills E. Coli Bacteria in 30 Seconds.

SINGAPORE – Real-Time Imaging of Embryo Development Could Pave the Way for More Effective Human Reproduction Therapies.

UNITED STATES – Innovative Device Allows 3D Imaging of the Breast with Less Radiation.

UNITED STATES – BIOTRONIK Launches Smallest MR Conditional Pacing Lead Available in the United States.

UNITED STATES – Computer App Whets Children's Appetites for Eco-Friendly Meals.

AUSTRALIA & CHINA – Uscom Agreement with CIIC Shanghai.

JAPAN – Chugai Announces License Agreement for Recycling Antibody SA237.

KOREA – ORIG3N(TM) and Kangstem Biotech Announce Collaboration to Study Neural Stem Cells.

Due to the successful union between computational technologies and basic laws of physics and biological sciences, many biomedical imaging systems now find significant presence in clinical settings, aiding physicians in diagnosing most forms of human illness with more confidence. In the case of breast imaging, apart from the basic diagnosis, these imaging systems also help in locating the abnormal tissues for biopsy, identifying the exact margins of the lesion for good lumpectomy results, staging and restaging the cancer, detecting locations of metastases, and planning and following up treatment protocols. It is well known that early detection of cancer is the only way to increase the survival rate of the patient. Without such imaging systems, it would be hard and almost impossible for the physicians to determine the nature and extent of the disease by merely simple physical examinations and biopsies. This article presents a description of most of these invaluable breast-imaging systems. Moreover, a comparison of these modalities and a review of a few of the developments these devices have come across over the years are also given.

This study optimized spatial resolution of mammography imaging quality using a CIRS-016A commercial line gauge and the Taguchi methodology. The line gauge with a precise line pair from 5lp/mm to 20lp/mm was placed on top of triangular PMMA plates to simulate the female breast undergoing mammography. Five factors: target/filter, kVp, mAs, PMMA plate thickness, and compression force, were organized into 18 groups according to the Taguchi L${}_{18}$ orthogonal array. Tactically, the 18 various combinations of factors could provide similar confidence levels, as those following the full factorial combination in reality. Seven experienced radiology experts judged the 18 imaging qualities based on contrast, sharpness, and spatial resolution. Then the signal-to-noise ratio was calculated according to the “the larger, the better” ranking order. The optimal preset of mammography was verified from the unique fish bone plot and the follow-up analysis of variance (ANOVA) test. The optimal combination of factors was as follows: Rh/Ag as target/filter, 32kVp, 36mAs, a 45mm thick PMMA plate, and a 13daN compression force in routine diagnosis. The concurrent resolution of 6lp/mm or about a 0.09mm minimum detectable difference (MDD) was superior to 5lp/mm of the conventional preset or combinations of factors of either highest Avg or lowest std. Compared to other studies with various facilities, this was the finest resolution among the routine X-ray, cardiac X-ray or computed tomography (CT), and computed tomography angiography (CTA).

Pre-operative X-ray mammography and intraoperative X-ray specimen radiography are routinely used to identify breast cancer pathology. Recent advances in optical coherence tomography (OCT) have enabled its use for the intraoperative assessment of surgical margins during breast cancer surgery. While each modality offers distinct contrast of normal and pathological features, there is an essential need to correlate image-based features between the two modalities to take advantage of the diagnostic capabilities of each technique. We compare OCT to X-ray images of resected human breast tissue and correlate different tissue features between modalities for future use in real-time intraoperative OCT imaging. X-ray imaging (specimen radiography) is currently used during surgical breast cancer procedures to verify tumor margins, but cannot image tissue in situ. OCT has the potential to solve this problem by providing intraoperative imaging of the resected specimen as well as the in situ tumor cavity. OCT and micro-CT (X-ray) images are automatically segmented using different computational approaches, and quantitatively compared to determine the ability of these algorithms to automatically differentiate regions of adipose tissue from tumor. Furthermore, two-dimensional (2D) and three-dimensional (3D) results are compared. These correlations, combined with real-time intraoperative OCT, have the potential to identify possible regions of tumor within breast tissue which correlate to tumor regions identified previously on X-ray imaging (mammography or specimen radiography).

Computer aided detection and Diagnosis systems are becoming very useful and helpful in supporting physicians for early detection and control of some diseases such as neoplastic pathologies. In this paper, a computer aided system for breast cancer diagnosis in mammographic images is presented. In particular, the method looks for microcalcification cluster occurrence and makes the diagnosis of the detected abnormality. The procedure first detects microcalcifications having a cluster pattern and then classifies the abnormalities as benign or malignant clusters. The method formulates the differentiation between malignant and benign microcalcification clusters as a supervised learning problem implementing an artificial neural network classifier. As input to the classifier, the procedure uses image features automatically extracted from the detected clusters. The seven features used are related both to the distribution of microcalcifications within cluster and to the uniformity of their shape. The performance of the implemented system is evaluated taking into account the accuracy of classifying clusters. The obtained results make this method able to operate as a "second opinion" helping radiologists during the routine clinical practice. Moreover, the implemented method has a general validity and can be used to detect and to classify microcalcification clusters independently from the acquisition equipment adopted during the mammographic screening.

Mammogram registration is an important preprocessing technique, which helps in finding asymmetrical regions in left and right breast. However, correct nipple position is the crucial key point of mammogram registration since it is the only consistent and stable landmark upon a mammogram. To locate the nipple coordinates accurately in mammogram images, this work improves previous algorithms such as maximum height of the breast border (MHBB) and proposes a novel method consisting of local spatial-maximum mean intensity (LSMMI), local maximum zero-crossing (LMZC) based on the second-order derivative, and a combined approach dependent on LSMMI and LMZC. The proposed method is tested on 413 mammogram images from MIAS and DDSM databases. Consequently, the mean Euclidean distance (MED) between the ground truth identified by the radiologist and the detected nipple position is 0.64 cm, within 1 cm of the gold standard, for estimating the proposed method. The experimental results hence indicate that our proposed method can detect the nipple positions more accurately than other previous methods. Furthermore, the proposed select visible-nipple mammograms (SVNM) algorithm with the ability of generalization has achieved a 99% selection rate for automatic clustering of nipples in a mammography database, besides automatically detecting the breast border and nipple positions in mammograms.

A framework for computer-aided analysis of mammograms is described. General computer vision algorithms are combined with application specific procedures in a hierarchical fashion. The system is under development and is currently limited to detection of a few types of suspicious areas.

The image features are extracted by using feature extraction methods where wavelet techniques are utilized. A low-pass pyramid representation of the image is convolved with a number of quadrature filters. The filter outputs are combined according to simple local Fourier domain models into parameters describing the local neighbourhood with respect to the model. This produces estimates for each pixel describing local size, orientation, Fourier phase, and shape with confidence measures associated to each parameter.

Tentative object descriptions are then extracted from the pixel-based features by application-specific procedures with knowledge of relevant structures in mammograms. The orientation, relative brightness and shape of the object are obtained by selection of the pixel feature estimates which best describe the object.

The list of object descriptions is examined by procedures, where each procedure corresponds to a specific type of suspicious area, e.g. clusters of microcalcifications.