W. C. A. Pereira

Analysis of Co-Occurrence Texture Statistics as a Function of Gray-Level Quantization for Classifying Breast Ultrasound

In this paper, we investigated the behavior of 22 co-occurrence statistics combined to six gray-scale quantization levels to classify breast lesions on ultrasound (BUS) images. The database of 436 BUS images used in this investigation was formed by 217 carcinoma and 219 benign lesions images. The region delimited by a minimum bounding rectangle around the lesion was employed to calculate the gray-level co-occurrence matrix (GLCM). Next, 22 co-occurrence statistics were computed regarding six quantization levels (8, 16, 32, 64, 128, and 256), four orientations (0° , 45° , 90° , and 135° ), and ten distances (1, 2,...,10 pixels). Also, to reduce feature space dimensionality, texture descriptors of the same distance were averaged over all orientations, which is a common practice in the literature. Thereafter, the feature space was ranked using mutual information technique with minimal-redundancy-maximal-relevance (mRMR) criterion. Fisher linear discriminant analysis (FLDA) was applied to assess the discrimination power of texture features, by adding the first m-ranked features to the classification procedure iteratively until all of them were considered. The area under ROC curve (AUC) was used as figure of merit to measure the performance of the classifier. It was observed that averaging texture descriptors of a same distance impacts negatively the classification performance, since the best AUC of 0.81 was achieved with 32 gray levels and 109 features. On the other hand, regarding the single texture features (i.e., without averaging procedure), the quantization level does not impact the discrimination power, since AUC=0.87 was obtained for the six quantization levels. Moreover, the number of features was reduced (between 17 and 24 features). The texture descriptors that contributed notably to distinguish breast lesions were contrast and correlation computed from GLCMs with orientation of 90° and distance more than five pixels.

Complexity curve and grey level co‐occurrence matrix in the texture evaluation of breast tumor on ultrasound images

This work aims at investigating texture parameters in distinguishing malign and benign breast tumors on ultrasound images. A rectangular region of interest (ROI) containing the tumor and its neighboring was defined for each image. Five parameters were extracted from the complexity curve (CC) of the ROI. Another five parameters were calculated from the grey-level co-occurrence matrix (GLCM) also for the ROI. The same was carried out for internal tumor region, hence, totaling 20 parameters. The linear discriminant analysis was applied to sets of up to five parameters and then the performances were assessed. The most relevant individual parameters were the contrast (con) (from the GLCM over the ROI) and the maximum value (mvi) from the CC just for the tumor internal region). When they were taken together, a correct classification slightly over 80% of the breast tumors was achieved. The highest performance (accuracy=84.2%, sensitivity=87.0%, and specificity=78.8%) was obtained with mvi, con, the standard deviation of the pixel pairs and the entropy, both for GLCM, and the internal region contrast also from GLCM. Parameters extracted from the internal region generally performed better and were more significant than those from the ROI. Moreover, parameters calculated only from CC or GLCM resulted in no statistically significant performance difference. These findings suggest that the texture parameters can be useful to help radiologist in distinguishing between benign or malign breast tumors on ultrasound images.

Computer simulation and discrete-event models in the analysis of a mammography clinic patient flow

OBJECTIVE This work develops a discrete-event computer simulation model for the analysis of a mammography clinic performance. MATERIAL AND METHODS Two mammography clinic computer simulation models were developed, based on an existing public sector clinic of the Brazilian Cancer Institute, located in Rio de Janeiro city, Brazil. Two clinics in a total of seven configurations (number of equipment units and working personnel) were studied. Models tried to simulate changes in patient arrival rates, number of equipment units, available personnel (technicians and physicians), equipment maintenance scheduling schemes and exam repeat rates. Model parameters were obtained by direct measurements and literature reviews. A commercially-available simulation software was used for model building. RESULTS The best patient scheduling (patient arrival rate) for the studied configurations had an average of 29 min for Clinic 1 (consisting of one mammography equipment, one to three technicians and one physician) and 21 min for Clinic 2 (two mammography equipment units, one to four technicians and one physician). The exam repeat rates and equipment maintenance scheduling simulations indicated that a large impact over patient waiting time would appear in the smaller capacity configurations. CONCLUSIONS Discrete-event simulation was a useful tool for defining optimal operating conditions for the studied clinics, indicating the most adequate capacity configurations and equipment maintenance schedules.

Performance of ultrasound echo decomposition using singular spectrum analysis.

Diagnostic ultrasonography has its well-established role in medicine. Nevertheless, the quantitative characterisation of biological tissues by ultrasound (US) is still a main topic of research. Several parameters have been explored with this purpose, (e.g. attenuation, backscatter coefficient, US speed). More recently, mean scatterer space (MSS) has been proposed as a characterisation parameter. The objective of this work was to investigate the potential of the singular spectrum analysis (SSA) to estimate MSS. This method proposes the reconstruction of the periodic part of the original US signal from where the MSS of the medium can be estimated. SSA is applied to simulated and real backscattered echoes from a phantom and a bovine liver sample. Consistent results were obtained from both Monte-Carlo simulation and real data. They were compared with literature. Presently, precision, accuracy and sensibility of SSA are being investigated.

Intraobserver interpretation of breast ultrasonography following the BI-RADS classification.

PURPOSE To use the BI-RADS ultrasound classification in an intraobserver retrospective study of the interpretation of breast images. MATERIALS AND METHODS The study used 40 breast ultrasound images recorded in orthogonal planes, obtained from patients with an indication for surgery. Eight professionals experienced in breast imaging analysis retrospectively reviewed these lesions, in three rounds of image interpretation (with a 3-6 months interval between rounds). Observers had no access to information from medical records or histopathological results, and, without their knowledge, in each new round were assigned the same images previously interpreted by them. Fleiss-modified Kappa measures were the study main concordance index. Besides the BI-RADS, a scale grouping its categories 2-3 and 4-5 was also used. The statistical analysis concerned the intraobserver agreement. RESULTS Kappa values ranged from 0.37 to 0.75 (original categories) and from 0.73 to 0.87 (grouped categories). Overall, out of the 8 observers, 7 presented moderate to substantial concordance (Kappa values 0.51 to 0.74). CONCLUSION The BI-RADS is a reporting tool that provides a standardized terminology for US exams. In this study, moderate to substantial concordance in Kappa values was found, in agreement with other studies of the literature.

Computational Evaluation of the Compositional Factors in Fracture Healing Affecting Ultrasound Axial Transmission Measurements

This work aimed at computationally evaluating the compositional factors in fracture healing affecting ultrasound axial transmission (UAT), using four numerical daily-changing healing models, representing more realistic clinical conditions. Using two-dimensional (2-D) simulations, a 1-MHz source and a receiver were positioned parallel to the bone surface to detect the first arriving signal (FAS). The time-of-flight of the FAS (TOF(FAS)) was found to be sensitive only to superficial modifications in the propagation path. It was also shown that callus mature bone better explained alone the variation in TOF(FAS) (R(2) >or= 0.70, p < 0.001). Better TOF(FAS) predictions are obtained when using the callus composition inside cortical fracture gap (R(2) = 0.98, p < 0.01). Callus composition could not well explain the changes in energy attenuation. These results suggest that UAT may be an important clinical tool for fracture healing assessment, identifying callus degree of mineralization and possible consolidation delays and nonunions.

A Soft-Computing Methodology for Noninvasive Time-Spatial Temperature Estimation

The safe and effective application of thermal therapies is restricted due to lack of reliable noninvasive temperature estimators. In this paper, the temporal echo-shifts of backscattered ultrasound signals, collected from a gel-based phantom, were tracked and assigned with the past temperature values as radial basis functions neural networks input information. The phantom was heated using a piston-like therapeutic ultrasound transducer. The neural models were assigned to estimate the temperature at different intensities and points arranged across the therapeutic transducer radial line (60 mm apart from the transducer face). Model inputs, as well as the number of neurons were selected using the multiobjective genetic algorithm (MOGA). The best attained models present, in average, a maximum absolute error less than 0.5 C, which is pointed as the borderline between a reliable and an unreliable estimator in hyperthermia/diathermia. In order to test the spatial generalization capacity, the best models were tested using spatial points not yet assessed, and some of them presented a maximum absolute error inferior to 0.5 C, being ldquoelectedrdquo as the best models. It should be also stressed that these best models present implementational low-complexity, as desired for real-time applications.

Experimental and simulation results on the effect of cortical bone mineralization in ultrasound axial transmission measurements: a model for fracture healing ultrasound monitoring.

Ultrasound axial transmission (UAT), a technique using propagation of ultrasound waves along the cortex of cortical bones, has been proposed as a diagnostic technique for the evaluation of fracture healing. Quantitative ultrasound parameters have been reported to be sensitive to callus changes during the regeneration process. The aim of this work was to identify the specific effect of cortical bone mineralization on UAT measurements by means of numerical simulations and experiments using a reverse fracture healing approach. A cortical bovine femur sample was used, in which a 3mm fracture gap was drilled. A 3mm thick cortical bone slice, extracted from another location in the bone sample, was submitted to a progressive demineralization process with EDTA during 12 days. UAT measurements and simulations using a 1MHz probe were performed with the demineralized slice placed into the fracture gap to mimic different stages of mineralization during the healing process. The calcium loss of the slice due to the EDTA treatment was recorded everyday, and its temporal evolution could be modeled by an exponential law. A 50MHz scanning acoustic microscopy was also used to assess the mineralization degree of the bone slice at the end of the intervention. These data were used in the numerical simulations to derive a model of the time evolution of bone slice mechanical properties. From both the experiments and the simulations, a significant and progressive increase in the time of flight (TOF; p<0.001) of the propagating waves measured by UAT was observed during the beginning of the demineralization process (first 4 days). Although the simulated TOF values were slightly larger than the experimental ones, they both exhibited a similar time-dependence, validating the simulation approach. Our results suggest that TOF measured in axial transmission is affected by local changes of speed of sound induced by changes in local mineralization. TOF may be an appropriate indicator to monitor callus maturation.

Assessing the performance of morphological parameters in distinguishing breast tumors on ultrasound images

This work aims at investigating seven morphological parameters in distinguishing malignant and benign breast tumors on ultrasound images. Linear discriminant analysis was applied to sets of up to five parameters and then the performances were assessed using the area Az (+/- standard error) under the ROC curve, accuracy (Ac), sensitivity (Se), specificity (Sp), positive predictive value and negative predictive value. The most relevant individual parameters were the normalized residual value (nrv) and overlap ratio (RS), both calculated from the convex polygon technique, and the circularity (C). When nrv and C were taken together with roughness (R), calculated from normalized radial length (NRL), a performance slightly over 83% in distinguishing malignant and benign breast tumors was achieved.

Classification of breast tumours on ultrasound images using morphometric parameters

This work aims to assess the potentiality of morphometric parameters in separating breast tumour, on ultrasonic images, as malign or benign. Parameters were calculated over normalised radial length and convex polygons from 152 segmented tumour images. Linear discriminant analysis was applied and parameters performance assessed (accuracy, sensitivity and specificity). The best parameter performances for individual parameters were the normalised residual mean square value and the circularity. Taking these last two and the roughness the best separation performance was obtained: specificity (90.4%) and sensitivity (88.0%). These three parameters were also applied to a multilayer perceptron network using GA-backpropagation hybrid training. The initial results pointed out that hybrid GA-backpropagation training was capable to produce similar high performance both to training (accuracy = 90.3%, sensitivity = 90.0% and specificity = 90.9%) and test (accuracy, sensitivity and specificity equal to 90.0%) procedures. Besides, the performances obtained with two training sets of distinct sizes (30% and 50% of all samples) were slightly different.