RESUMEN

Neurodegenerative diseases, tauopathies, constitute a serious global health problem. The etiology of these diseases is unclear and an increase in their incidence has been projected in the next 30 years. Therefore, the study of the molecular mechanisms that might stop these neurodegenerative processes is very relevant. Classification of neurodegenerative diseases using Machine and Deep Learning algorithms has been widely studied for medical imaging such as Magnetic Resonance Imaging. However, post-mortem immunofluorescence imaging studies of the brains of patients have not yet been used for this purpose. These studies may represent a valuable tool for monitoring aberrant chemical changes or pathological post-translational modifications of the Tau polypeptide. We propose a Convolutional Neural Network pipeline for the classification of Tau pathology of Alzheimer's disease and Progressive Supranuclear Palsy by analyzing post-mortem immunofluorescence images with different Tau biomarkers performed with models generated with the architecture ResNet-IFT using Transfer Learning. These models' outputs were interpreted with interpretability algorithms such as Guided Grad-CAM and Occlusion Analysis. To determine the best classifier, four different architectures were tested. We demonstrated that our design was able to classify diseases with an accuracy of 98.41% on average whilst providing an interpretation concerning the proper classification involving different structural patterns in the immunoreactivity of the Tau protein in NFTs present in the brains of patients with Progressive Supranuclear Palsy and Alzheimer's disease.

RESUMEN

Motor imagery is a complex mental task that represents muscular movement without the execution of muscular action, involving cognitive processes of motor planning and sensorimotor proprioception of the body. Since the mental task has similar behavior to that of the motor execution process, it can be used to create rehabilitation routines for patients with some motor skill impairment. However, due to the nature of this mental task, its execution is complicated. Hence, the classification of these signals in scenarios such as brain-computer interface systems tends to have a poor performance. In this work, we study in depth different forms of data representation of motor imagery EEG signals for distinct CNN-based models as well as novel EEG data representations including spectrograms and multidimensional raw data. With the aid of transfer learning, we achieve results up to 93% accuracy, exceeding the current state of the art. However, although these results are strong, they entail the use of high computational resources to generate the samples, since they are based on spectrograms. Thus, we searched further for alternative forms of EEG representations, based on 1D, 2D, and 3D variations of the raw data, leading to promising results for motor imagery classification that still exceed the state of the art. Hence, in this work, we focus on exploring alternative methods to process and improve the classification of motor imagery features with few preprocessing techniques.

Asunto(s)

Interfaces Cerebro-Computador , Imaginación , Algoritmos , Electroencefalografía/métodos , Humanos , Aprendizaje Automático , Movimiento , Redes Neurales de la Computación

RESUMEN

Efforts have been made to diagnose and predict the course of different neurodegenerative diseases through various imaging techniques. Particularly tauopathies, where the tau polypeptide is a key participant in molecular pathogenesis, have significantly increased their morbidity and mortality in the human population over the years. However, the standard approach to exploring the phenomenon of neurodegeneration in tauopathies has not been directed at understanding the molecular mechanism that causes the aberrant polymeric and fibrillar behavior of the tau protein, which forms neurofibrillary tangles that replace neuronal populations in the hippocampal and cortical regions. The main objective of this work is to implement a novel quantification protocol for different biomarkers based on pathological post-translational modifications undergone by tau in the brains of patients with tauopathies. The quantification protocol consists of an adaptation of the U-Net neural network architecture. We used the resulting segmentation masks for the quantification of combined fluorescent signals of the different molecular changes tau underwent in neurofibrillary tangles. The quantification considers the neurofibrillary tangles as an individual study structure separated from the rest of the quadrant present in the images. This allows us to detect unconventional interaction signals between the different biomarkers. Our algorithm provides information that will be fundamental to understanding the pathogenesis of dementias with another computational analysis approach in subsequent studies.

RESUMEN

We investigated biomass, size-structure, composition, depth distributions and spatial variability of the phytoplankton community in the Costa Rica Dome (CRD) in June-July 2010. Euphotic zone profiles were sampled daily during Lagrangian experiments in and out of the dome region, and the community was analyzed using a combination of digital epifluorescence microscopy, flow cytometry and HPLC pigments. The mean depth-integrated biomass of phytoplankton ranged 2-fold, from 1089 to 1858 mg C m-2 (mean ± SE = 1378 ± 112 mg C m-2), among 4 water parcels tracked for 4 days. Corresponding mean (±SE) integrated values for total chlorophyll a (Chl a) and the ratio of autotrophic carbon to Chl a were 24.1 ± 1.5 mg Chl a m-2 and 57.5 ± 3.4, respectively. Absolute and relative contributions of picophytoplankton (∼60%), Synechococcus (>33%) and Prochlorococcus (17%) to phytoplankton community biomass were highest in the central dome region, while >20 µm phytoplankton accounted for ≤10%, and diatoms <2%, of biomass in all areas. Nonetheless, autotrophic flagellates, dominated by dinoflagellates, exceeded biomass contributions of Synechococcus at all locations. Order-of-magnitude discrepancies in the relative contributions of diatoms (overestimated) and dinoflagellates (underestimated) based on diagnostic pigments relative to microscopy highlight potential significant biases associated with making community inferences from pigments.

RESUMEN

During summer 2010, we investigated phytoplankton production and growth rates at 19 stations in the eastern tropical Pacific, where winds and strong opposing currents generate the Costa Rica Dome (CRD), an open-ocean upwelling feature. Primary production (14C-incorporation) and group-specific growth and net growth rates (two-treatment seawater dilution method) were estimated from samples incubated in situ at eight depths. Our cruise coincided with a mild El Niño event, and only weak upwelling was observed in the CRD. Nevertheless, the highest phytoplankton abundances were found near the dome center. However, mixed-layer growth rates were lowest in the dome center (∼0.5-0.9 day-1), but higher on the edge of the dome (∼0.9-1.0 day-1) and in adjacent coastal waters (0.9-1.3 day-1). We found good agreement between independent methods to estimate growth rates. Mixed-layer growth rates of Prochlorococcus and Synechococcus were largely balanced by mortality, whereas eukaryotic phytoplankton showed positive net growth (∼0.5-0.6 day-1), that is, growth available to support larger (mesozooplankton) consumer biomass. These are the first group-specific phytoplankton rate estimates in this region, and they demonstrate that integrated primary production is high, exceeding 1 g C m-2 day-1 on average, even during a period of reduced upwelling.

RESUMEN

We investigated phytoplankton production rates and grazing fates in the Costa Rica Dome (CRD) during summer 2010 based on dilution depth profiles analyzed by flow cytometry and pigments and mesozooplankton grazing assessed by gut fluorescence. Three community production estimates, from 14C uptake (1025 ± 113 mg C m-2 day-1) and from dilution experiments analyzed for total Chla (990 ± 106 mg C m-2 day-1) and flow cytometry populations (862 ± 71 mg C m-2 day-1), exceeded regional ship-based values by 2-3-fold. Picophytoplankton accounted for 56% of community biomass and 39% of production. Production profiles extended deeper for Prochlorococcus (PRO) and picoeukaryotes than for Synechococcus (SYN) and larger eukaryotes, but 93% of total production occurred above 40 m. Microzooplankton consumed all PRO and SYN growth and two-third of total production. Positive net growth of larger eukaryotes in the upper 40 m was balanced by independently measured consumption by mesozooplankton. Among larger eukaryotes, diatoms contributed ∼3% to production. On the basis of this analysis, the CRD region is characterized by high production and grazing turnover, comparable with or higher than estimates for the eastern equatorial Pacific. The region nonetheless displays characteristics atypical of high productivity, such as picophytoplankton dominance and suppressed diatom roles.

RESUMEN

Improving fingerprint matching algorithms is an active and important research area in fingerprint recognition. Algorithms based on minutia triplets, an important matcher family, present some drawbacks that impact their accuracy, such as dependency to the order of minutiae in the feature, insensitivity to the reflection of minutiae triplets, and insensitivity to the directions of the minutiae relative to the sides of the triangle. To alleviate these drawbacks, we introduce in this paper a novel fingerprint matching algorithm, named M3gl. This algorithm contains three components: a new feature representation containing clockwise-arranged minutiae without a central minutia, a new similarity measure that shifts the triplets to find the best minutiae correspondence, and a global matching procedure that selects the alignment by maximizing the amount of global matching minutiae. To make M3gl faster, it includes some optimizations to discard non-matching minutia triplets without comparing the whole representation. In comparison with six verification algorithms, M3gl achieves the highest accuracy in the lowest matching time, using FVC2002 and FVC2004 databases.