Despeckling optical coherence tomograms from the human retina is a fundamental step to a better diagnosis or as a preprocessing stage for retinal layer segmentation. Both of these applications are particularly important in monitoring the progression of retinal disorders. In this study we propose a new formulation for a well-known nonlinear complex diffusion filter. A regularization factor is now made to be dependent on data, and the process itself is now an adaptive one. Experimental results making use of synthetic data show the good performance of the proposed formulation by achieving better quantitative results and increasing computation speed.

Ocular fundus imaging plays a key role in monitoring the health status of the human eye. Currently, a large number of imaging modalities allow the assessment and/or quantification of ocular changes from a healthy status. This review focuses on the main digital fundus imaging modality, color fundus photography, with a brief overview of complementary techniques, such as fluorescein angiography. While focusing on two-dimensional color fundus photography, the authors address the evolution from nondigital to digital imaging and its impact on diagnosis. They also compare several studies performed along the transitional path of this technology. Retinal image processing and analysis, automated disease detection and identification of the stage of diabetic retinopathy (DR) are addressed as well. The authors emphasize the problems of image segmentation, focusing on the major landmark structures of the ocular fundus: the vascular network, optic disk and the fovea. Several proposed approaches for the automatic detection of signs of disease onset and progression, such as microaneurysms, are surveyed. A thorough comparison is conducted among different studies with regard to the number of eyes/subjects, imaging modality, fundus camera used, field of view and image resolution to identify the large variation in characteristics from one study to another. Similarly, the main features of the proposed classifications and algorithms for the automatic detection of DR are compared, thereby addressing computer-aided diagnosis and computer-aided detection for use in screening programs.

We present a new method for solving the time-harmonic inverse scattering problem for sound-soft or perfectly conducting cracks in two dimensions. Our approach extends a method that was recently suggested by one of us for inverse obstacle scattering. It can be viewed as a hybrid between a regularized Newton iteration method applied to a nonlinear operator equation involving the operator that, for a fixed incident wave, maps the crack onto the far-field pattern of the scattered wave and a decomposition method due to Kirsch and Kress. As an important feature, in contrast to the traditional Newton iterations for solving inverse scattering problems, our method does not require a forward solver for each iteration step. The theoretical background of the method is based on the minimization of a cost function containing an additional penalty term to deal with reconstructing the full crack. Numerical examples illustrate the feasibility of the method and its stability with respect to noisy data. We expect that the method can also be extended to sound-hard cracks.

Correia, P. (2015). Recurrence quantification analysis and support vector machines for golf handicap and low back pain EMG classification. Journal of Electromyography & Kinesiology, 25, AbstractThe quantification of non-linear characteristics of electromyography (EMG) must contain information allowing to discriminate neuromuscular strategies during dynamic skills. In golf, both handicap (Hc) and low back pain (LBP) are main factors associated with the occurrence of injuries. The aim of this study was to analyze the accuracy of support vector machines SVM on EMG-based classification to discriminate Hc (low and high handicap) and LBP prevalence (with and without LPB) in the main phases of golf swing. For this purpose recurrence quantification analysis (RQA) features of the trunk and the lower limb muscles were used to feed a SVM classifier. Recurrence rate (RR) and the ratio between determinism (DET) and RR showed a high discriminant weight. The Hc classifications accuracy for the swing, backswing (BS), and downswing (DS) were 94.4±2.7%, 97.1±2.3%, and 95.3±2.6%, respectively. For LBP, the accuracy was 96.9±3.8% in the swing, and 99.7%±0.4% in BS. External oblique (EO), biceps femoris (BF), semitendinosus (ST) and rectus femoris (RF) showed high accuracy depending on the laterality within the phase. RQA features and SVM showed a high capacity in discriminating muscles within swing phases by Hc and by LBP. Low back pain golfers showed less neuromuscular coordination strategies than asymptomatic.