Discriminative Optimization: Theory and Applications to Computer Vision.

Many computer vision problems are formulated as the optimization of a cost function. This approach faces two main challenges: designing a cost function with a local optimum at an acceptable solution, and developing an efficient numerical method to search for this optimum. While designing such functions is feasible in the noiseless case, the stability and location of local optima are mostly unknown under noise, occlusion, or missing data. In practice, this can result in undesirable local optima or not having a local optimum in the expected place. On the other hand, numerical optimization algorithms in high-dimensional spaces are typically local and often rely on expensive first or second order information to guide the search. To overcome these limitations, we propose Discriminative Optimization (DO), a method that learns search directions from data without the need of a cost function. DO explicitly learns a sequence of updates in the search space that leads to stationary points that correspond to the desired solutions. We provide a formal analysis of DO and illustrate its benefits in the problem of 3D registration, camera pose estimation, and image denoising. We show that DO outperformed or matched state-of-the-art algorithms in terms of accuracy, robustness, and computational efficiency.

Matrix Completion for Weakly-Supervised Multi-Label Image Classification.

In the last few years, image classification has become an incredibly active research topic, with widespread applications. Most methods for visual recognition are fully supervised, as they make use of bounding boxes or pixelwise segmentations to locate objects of interest. However, this type of manual labeling is time consuming, error prone and it has been shown that manual segmentations are not necessarily the optimal spatial enclosure for object classifiers. This paper proposes a weakly-supervised system for multi-label image classification. In this setting, training images are annotated with a set of keywords describing their contents, but the visual concepts are not explicitly segmented in the images. We formulate the weakly-supervised image classification as a low-rank matrix completion problem. Compared to previous work, our proposed framework has three advantages: (1) Unlike existing solutions based on multiple-instance learning methods, our model is convex. We propose two alternative algorithms for matrix completion specifically tailored to visual data, and prove their convergence. (2) Unlike existing discriminative methods, our algorithm is robust to labeling errors, background noise and partial occlusions. (3) Our method can potentially be used for semantic segmentation. Experimental validation on several data sets shows that our method outperforms state-of-the-art classification algorithms, while effectively capturing each class appearance.

[Musculoskeletal tuberculosis - a clinical case report]. / Tuberculose musculoesquelética - A propósito de um caso clínico.

Tuberculosis of the chest wall constitutes 1% to 5% of all cases of musculoskeletal TB. Abscesses of the chest wall are rare tuberculous locations. Because of the resurgence of the tuberculosis associated to AIDS, that diagnosis must be considered more frequently. The authors present a case of osseous tuberculosis with an abscess rib in a patient with HIV. The combination of indolent onset of symptoms and compatible radiographic findings, strongly suggests the diagnosis. However, the confirmation with positive culture or histopathologic are essential for definitive diagnosis.

[Multi-drug resistant tuberculosis and the red queen - diagnosis speed is crucial]. / A tuberculose multirresistente e a rainha vermelha - A rapidez do diagnóstico é decisiva.

The multi-drug resistant Tuberculosis (MDRTB) is a huge menace to Tuberculosis control. The early detection of MDRTB is essential to best appropriate measures. The detection methods for drug resistance based in evaluation of the genetic determinants (genotypic methods), instead of phenotypic methods, allows for faster results, the possibility of direct application in clinical samples and simultaneous identification of Mycobacterium tuberculosis complex. The inpatients data analysis in the "Serviço de Pneumologia 2 do Hospital Pulido Valente", showed a high prevalence of MDRTB (10.3%). In 34.1% of the MDRTB patients the multi-drug resistance was not been identified, with a mortality ratio in this cases of 31% versus 18.4% in the subset of patients with resistance previously identified. Moreover the mortality ratio was worst in MDRTB/AIDS patients with 50% versus 15%, respectively. Targeting for rapid drug resistance detection, in hospitalized patients at "Serviço de Pneumologia 2 do Hospital Pulido Valente", the test INNO-LIPA Rif.TB, to identify the rifampicin resistance as a marker of multi-drug resistance, was evaluated. The test was performed in 113 samples and had a high ratio of sensitivity (91.6%), specificity (98%), positive predictive value (84, 6%) and negative predictive value (99%). Time to obtain the results was 7.6 days for the genotypic test versus 23.4 days to the phenotypic test (BACTEC MGIT 960). The INNO-LIPA Rif.TB test is, now, performed in every patient with smear-positive Tuberculosis with no previous knows resistance profile, with good outcome. Rev