Semi-supervised learning techniques are gaining popularity due to their capability of building models that are effective, even when scarce amounts of labeled data are available. In this paper, we ...present a framework and specific tasks for self-supervised pretraining of multichannel models, such as the fusion of multispectral and synthetic aperture radar images. We show that the proposed self-supervised approach is highly effective at learning features that correlate with the labels for land cover classification. This is enabled by an explicit design of pretraining tasks which promotes bridging the gaps between sensing modalities and exploiting the spectral characteristics of the input. In a semi-supervised setting, when limited labels are available, using the proposed self-supervised pretraining, followed by supervised finetuning for land cover classification with SAR and multispectral data, outperforms conventional approaches such as purely supervised learning, initialization from training on ImageNet and other recent self-supervised approaches.
Hyperbolic neural networks are emerging as an effective technique to better capture hierarchical representations of many data types, from text to images and, recently, point clouds. In this paper, we ...extend our earlier work, that showed how to use regularizers in the hyperbolic space to improve performance of point cloud classification models, to the problem of part segmentation. This requires careful modeling of the hierarchical relationships between parts and whole point cloud to properly control the hyperbolic geometry of the feature space produced by the neural network. We show how the proposed method improves the performance of commonly used neural network architectures, reaching state-of-the-art performance on the part segmentation task.
Inverse problems consist in reconstructing signals from incomplete sets of measurements and their performance is highly dependent on the quality of the prior knowledge encoded via regularization. ...While traditional approaches focus on obtaining a unique solution, an emerging trend considers exploring multiple feasibile solutions. In this paper, we propose a method to generate multiple reconstructions that fit both the measurements and a data-driven prior learned by a generative adversarial network. In particular, we show that, starting from an initial solution, it is possible to find directions in the latent space of the generative model that are null to the forward operator, and thus keep consistency with the measurements, while inducing significant perceptual change. Our exploration approach allows to generate multiple solutions to the inverse problem an order of magnitude faster than existing approaches; we show results on image super-resolution and inpainting problems.
The Tracking Ultraviolet Setup (TUS) was the first orbital detector aimed to check the possibility of recording ultra-high energy cosmic rays (UHECRs) at E≳100 EeV by measuring the fluorescence ...signal of extensive air showers in the atmosphere. TUS was an experiment funded by the Russian Space Agency ROSCOSMOS, and it operated as a part of the scientific payload of the Lomonosov satellite since April 2016 till late 2017. During its mission, TUS registered almost 80,000 events in its main operation mode, with a few of them being sufficiently interesting to be more deeply scrutinized as they showed light profile and duration similar to UHECR events, even though much brighter. At the same time, the data acquired by TUS in different acquisition modes have been used to search for more exotic matter such us strangelets and nuclearites, and to measure occurrence, time profile and signal amplitude of different classes of transient luminous events among other scientific objectives, showing the interdisciplinary capability of a space-based observatory for UHECRs. In this paper, we report a selection of studies and results obtained with the TUS telescope which will be presented and placed in the contest of the present and future missions dedicated to the observation of UHECRs from space such as Mini-EUSO, K-EUSO and POEMMA.
In this article, we present cutting-edge machine learning-based techniques for the detection and reconstruction of meteors and space debris in the Mini-EUSO experiment, a detector installed on board ...of the International Space Station, and pointing toward the Earth. We base our approach on a recent technique, the STACKing method plus Convolutional Neural Network (STACK-CNN), originally developed as an online trigger in an orbiting remediation system to detect space debris. Our proposed method, the refined-STACKing method plus convolutional neural network (R-Stack-CNN), makes the STACKing method plus convolutional neural network (STACK-CNN) more robust, thanks to a random forest that learns the temporal development of these events in the camera. We prove the flexibility of our method by showing that it is sensitive to any space object that moves linearly in the field of view. First, we search small space debris, never observed by Mini-EUSO. Due to the limiting statistics, also in this case, no debris were found. However, since meteors produce signals similar to space debris but they are much more frequent, the R-Stack-CNN is adapted to identify such events while avoiding the numerous false positives of the Stack-CNN. Results from real data show that the R-Stack-CNN is able to find more meteors than a classical thresholding method and a new method of two neural networks. We also show that the method is also able to accurately reconstruct speed and direction of meteors with simulated data.
Cerebellar infarction: analysis of 151 patients Rosi, Jr, Jefferson; de Oliveira, Paulo Geraldo Dorsa; Montanaro, Antônio Carlos ...
Arquivos de neuro-psiquiatria
64, Številka:
2B
Journal Article
Recenzirano
Odprti dostop
This report presents the treatment of 151 patients with cerebellar infarction, 98 men (65%) and 53 women (35%), mean age 62.4 years old. Occlusive hydrocephalus was diagnosed in 7.9% of the patients ...associated with an extensive cerebellar infarction and in all 11 surgical patients (7.2%). Four patients underwent an external ventricular drainage with 3 deaths (75%) and 7 underwent a decompressive suboccipital craniectomy with 2 deaths (28.5%). Mortality of the clinical group was 15 patients (10.7%). Vertigo, vomiting, Romberg sign and dysmetria were the signs and symptoms of cerebellar involvement that were most frequently observed. Cerebellar infarction from embolism after cardiovascular surgery occurred in 57 patients (37.7%). Cerebellar infarction, as an isolated fact, occurred in 59 patients (39%) and cerebellar plus infarction in other regions occurred in 92 patients (61%). Magnetic resonance image was the best diagnostic form for cerebellar lesions, however computerized tomography could show cerebellar infarction in 68 patients (78%).
Point clouds of 3D objects exhibit an inherent compositional nature where simple parts can be assembled into progressively more complex shapes to form whole objects. Explicitly capturing such ...part-whole hierarchy is a long-sought objective in order to build effective models, but its tree-like nature has made the task elusive. In this paper, we propose to embed the features of a point cloud classifier into the hyperbolic space and explicitly regularize the space to account for the part-whole hierarchy. The hyperbolic space is the only space that can successfully embed the tree-like nature of the hierarchy. This leads to substantial improvements in the performance of state-of-art supervised models for point cloud classification.
Inverse problems consist in reconstructing signals from incomplete sets of measurements and their performance is highly dependent on the quality of the prior knowledge encoded via regularization. ...While traditional approaches focus on obtaining a unique solution, an emerging trend considers exploring multiple feasibile solutions. In this paper, we propose a method to generate multiple reconstructions that fit both the measurements and a data-driven prior learned by a generative adversarial network. In particular, we show that, starting from an initial solution, it is possible to find directions in the latent space of the generative model that are null to the forward operator, and thus keep consistency with the measurements, while inducing significant perceptual change. Our exploration approach allows to generate multiple solutions to the inverse problem an order of magnitude faster than existing approaches; we show results on image super-resolution and inpainting problems.
Resumo
O infarto maligno da artéria cerebral média é definido como a ocorrência de edema cerebral intenso, circunjacente à área de um infarto extenso. O edema pode causar deterioração da consciência, ...aumentar a pressão intracraniana, provocar desvio das estruturas da linha média e, finalmente, herniação cerebral e morte. Indivíduos que desenvolvem acidente vascular cerebral isquêmico maligno representam de 1% a 10% dos casos de isquemia cerebral supratentorial. A história natural dessa doença segue um curso previsível na maior parte dos casos, chegando a apresentar uma mortalidade de até 80% quando tratados clinicamente. Os sobreviventes são incapacitados e afligidos por graves seqüelas neurológicas, tornando-se dependentes de cuidados e acamados. A craniectomia descompressiva tem evidenciado resultados animadores, com redução na mortalidade para níveis que variam de 16% a 42% e uma melhor qualidade de vida aos sobreviventes. A presente revisão da literatura tem como principal objetivo caracterizar, de forma prática, o acidente vascular cerebral maligno – epidemiologia, etiologia, apresentação clínica, história natural da doença, medidas terapêuticas e prognóstico – bem como buscar embasamento científico à indicação de hemicraniectomia descompressiva.