Catálogo de publicaciones - libros

Radiographic signs indicating the presence of a malignancy are a result of the morphology and composition of the lesion. Assessment of the size, distribution, extent and location of disease are crucial in guiding patient management. Often mammographic estimates of size and extent are underestimated. Radiologic/pathologic correlation between features is often by indirect classification methods rather than a direct, whole-volume, one-to-one spatial correlation between radiologic and pathologic images. As an initial step toward understanding how tumour morphology and composition yields a mammographic sign, we have begun work on correlating whole-mount histology sections to cone-beam computed tomography (CBCT) images of the same specimen. Preliminary results for a lumpectomy sample containing a 3.5 cm invasive ductal carcinoma qualitatively show a remarkable correspondence between CBCT slices and histology sections. Ultimately, the 3D CBCT data could be used to predict mammographic features, which could then be correlated precisely to the anatomy of the tumour.

In recent years a large number of digital mammograms have been generated in hospitals and breast screening centers. To assist diagnosis through indexing those mammogram databases, we proposed a content-based image retrieval framework along with a novel feature extraction technique for describing the degree of calcification phenomenon revealed in the mammograms and six relevance feedback learning algorithms, which fall in the category of query point movement , for improving system performance. The results show that the proposed system can reach a precision rate of 0.716 after five rounds of relevance feedback have been performed.

We evaluated three combinations of filters (Mo/Mo, Mo/ Rh, and W /Rh) in direct a-Se FFDM system to optimize radiation dose clinically. We measured CNR (Contrast to Noise Ratio) as physical characteristics changing radiation dose and phantom thickness in clinical range. In 20, 30, 40, and 50mm PMMA phantoms, Mo / Mo showed the best performances. On the other hand, in 60 and 70 mm, W/Rh 30kV showed best performance. In addition, in 40 and 50mm PMMA phantoms, W/Rh 30kV showed the second best performance. In direct a-Se FFDM system, W/Rh was valuable in minimizing radiation dose.

Classifier plays an important role in a system detecting abnormal shadows from mammograms. In this paper, we propose the novel classification system that cascades four weak classifiers and a classifier ensemble to improve both computational cost and classification accuracy. The first several weak classifiers eliminate a large number of false positives in a short time which are easy to distinguish from abnormal regions, and the final classifier ensemble focuses on the remaining candidate regions difficult to classify, which results in high accuracy. We also show the experimental results using 2,564 mammograms.

In this paper, co-occurrence matrix based texture features are used to classify masses as benign or malignant. As (digitized) mammograms have high depth resolution (4096 gray levels in this study) and the size of a co-occurrence matrix depends on Q , the number of gray levels used for image intensity (depth) quantization, computation using co-occurrence matrices derived from mammograms can be expensive. Re-quantization using a lower value of Q is routinely performed but the effect of such procedure has not been sufficiently investigated. This paper investigates the effect of re-quantization using different Q . Four feature pools are formed with features measured on co-occurrence matrices with Q ∈{400}, Q ∈{100}, Q ∈{50} and Q ∈{400, 100, 50}. Classification results are obtained from each pool separately with the use of a genetic algorithm and the Fisher’s linear discriminant classifier. For Q ∈{400, 100, 50}, the best feature subsets selected by the genetic algorithm and of size k =6,7,8 have a leave-one-out area under the receiver operating characteristic (ROC) curve of 0.92, 0.93 and 0.94, respectively. Pairwise comparisons of the area index show that the differences in classification results for Q ∈{400, 100, 50} and Q ∈{50} are significant ( p <0.06) for all k while that for Q ∈{400, 100, 50} and Q ∈{400} or Q ∈{100} are not significant.

A novel approach to the detection of masses and clustered microcalcification is presented. Lesion detection is considered as a two-class pattern recognition problem. In order to get an effective and stable representation, the detection scheme codifies the image by using a ranklet transform. The vectors of ranklet coefficients obtained are classified by means of an SVM classifier. Our approach has two main advantages. First it does not need any feature selected by the trainer. Second, it is quite stable, with respect to the image histogram. That allows us to tune the detection parameters in one database and use the trained CAD on other databases without needing any adjustment. In this paper, training is accomplished on images coming from different databases (both digitized and digital). Test results are calculated on images coming from a few FFDM Giotto Image MD clinical units. The sensitivity of our CAD system is about 85% with a false-positive rate of 0.5 marks per image.

Breast cancer is one of the main leading causes to women mortality in the world especially in the western countries. Since the causes are still unknown, breast cancer cannot be prevented completely even till now. Microcalcification clusters are primary indicators of malignant types of breast cancer, the detection is important to prevent and treat the disease. The microcalcifications appear in the small clusters of a few pixels with relatively high intensity and closed contours compared with their neighboring pixels. However, it is a challenge to detect all the microcalcifications since they appear as spots which are slightly brighter than their backgrounds. This paper presents an approach for detecting microcalcifications in digital mammograms employing a dual-threshold method. These microcalcifications can be located by our new method which is developed from LoG edge detection method. Two thresholds are proposed in our method based on two additional criterions. Experimental results show that the proposed method can locate the microcalcifications exactly in mammogram as well as restrain the contours produced by the noises.

A link between increased breast density, as visualised in mammograms, and increased risk of developing breast cancer has been established. Recently, a number of objective, quantitative methods for measuring breast density have been described. One such method requires a calibration object to be imaged alongside the breast. However, it is important that this should not interfere with the routine imaging process. In this paper, we investigate the amount of space in mammographic images which is not currently occupied by the breast or existing patient labels and markers, and which would therefore be available for imaging an additional calibration device. We do this with a view to estimating the likelihood of failure of the method, and also to determining whether, without detriment to the imaging process, a device could be permanently fixed to the breast support platform. We also examine the impact of markers attached to the compression plate on the visibility of breast tissue. The results show that our existing calibration device may be used successfully without interfering with the routine imaging process, although permanently fixing such a device may present problems in a small minority of cases, and we demonstrate that the number of cases which would fail can be reduced by using a smaller stepwedge.

An experimental method of determining the optimal beam qualities and doses for digital mammography systems is described, and applied to a CR system. The mean glandular dose (MGD) and contrast-to-noise ratio (CNR) were measured using phantoms. For each thickness of phantom a range of kV and target/filter combinations were tested. Optimal beam quality was defined as that giving a target CNR for the lowest MGD. The target CNR was that necessary to achieve at least the minimum standard of image quality defined in European Guidance. An inverse relationship between CNR and threshold contrast was confirmed over a range of thicknesses of PMMA and different beam qualities and doses. Optimisation indicated that relatively high energy beam qualities (e.g. 31 kV Rh/Rh) should be used with a greater detector dose to compensate for the lower contrast when compared to using lower energy X-rays. The results also indicate that current AEC designs that aim for a fixed detector dose are not optimal.

The clustered microcalcifications and mass are the important findings in interpreting breast cancer, architectural distortion on mammograms as well. We have developed the detection algorithm for distorted area based on concentration of mammary gland. The purpose of this study is to suggest the improvement of extraction method of mammary gland in order to achieve higher sensitivity. The mean curvature, and the combination of shape index and curvedness were performed for extracting of mammary gland in our previous methods. In our new method, the dynamic-range compression was added as the pre-processing before extracting mammary gland by mean curvature. The detection rate at initial pick-up stage was improved by this improvement. It was concluded that our detection method would be effective.