RESUMO
This work presents the usefulness of texture features in the classification of breast lesions in 5,518 images of regions of interest, which were obtained from the Digital Database for Screening Mammography that included microcalcifications, masses, and normal cases. Sixteen texture features were used, i.e., 13 were based on the spatial gray-level dependence matrix and 3 on the wavelet transform. The nonparametric K-NN classifier was used in the classification stage. The results obtained from receiver operating characteristic analysis indicated that the texture features can be used for separating normal regions and lesions with masses and microcalcifications, yielding the area under the curve (AUC) values of 0.957 and 0.859, respectively. However, the texture features were not very effective for distinguishing between malignant and benign lesions because the AUC was 0.617 for masses and 0.607 for microcalcifications. The study showed that the texture features can be used for the detection of suspicious regions in mammograms.
Assuntos
Neoplasias da Mama/diagnóstico por imagem , Mamografia , Interpretação de Imagem Radiográfica Assistida por Computador , Neoplasias da Mama/patologia , Calibragem , Diagnóstico Diferencial , Feminino , Humanos , Imagens de Fantasmas , Curva ROC , Análise de Regressão , Estatísticas não ParamétricasRESUMO
The myoelectric signal can be used to control many rehabilitation systems, for instance, prostheses and artificial neuromuscular electrical stimulation toward restoring movement to spinal cord injured subjects. These mobile systems are usually used in different environments and thus are being exposed to different noise levels with characteristics not completely known. In this article, three main techniques for noise reduction were evaluated: wavelet transform (WT), adaptive digital filters, and nonadaptive digital filters. The WT was used to reconstruct the signal with the components without noise information. Adaptive filters were designed using least mean square (LMS) and recursive least square (RLS) algorithms. Finite-impulse response (FIR) and infinite-impulse response (IIR) nonadaptive filters were used for comparison to both the adaptive filters and the signal reconstruction through the WT.