Robótica móvel: sistema de condução autónoma

Dissertação para obtenção do grau de Mestre em Engenharia Electrotécnica Ramo Automação e Electrónica Industrial

Value of a lower limb immobilization device for SPECT/CT image fusion optimization

The foot and the ankle are small structures commonly affected by disorders, and their complex anatomy represent significant diagnostic challenges. SPECT/CT Image fusion can provide missing anatomical and bone structure information to functional imaging, which is particularly useful to increase diagnosis certainty of bone pathology. However, due to SPECT acquisition duration, patient’s involuntary movements may lead to misalignment between SPECT and CT images. Patient motion can be reduced using a dedicated patient support. We aimed at designing an ankle and foot immobilizing device and measuring its efficacy at improving image fusion. Methods: We enrolled 20 patients undergoing distal lower-limb SPECT/CT of the ankle and the foot with and without a foot holder. The misalignment between SPECT and CT images was computed by manually measuring 14 fiducial markers chosen among anatomical landmarks also visible on bone scintigraphy. Analysis of variance was performed for statistical analysis. Results: The obtained absolute average difference without and with support was 5.1±5.2 mm (mean±SD) and 3.1±2.7 mm, respectively, which is significant (p<0.001). Conclusion: The introduction of the foot holder significantly decreases misalignment between SPECT and CT images, which may have clinical influence in the precise localization of foot and ankle pathology.

Solid-state sensory properties of Calix-Poly(Phenylene Ethynylene)s toward nitroaromatic explosives

This study is primarily focused in establishing the solid-state sensory abilities of several luminescent polymeric calix[4]arene-based materials toward selected nitroaromatic compounds (NACs), creating the foundations for their future application as high performance materials for detection of high explosives. The phenylene ethynylene-type polymers possessing bis-calix[4]arene scaffolds in their core were designed to take advantage of the known recognition abilities of calixarene compounds toward neutral guests, particularly in solid-state, therefore providing enhanced sensitivity and selectivity in the sensing of a given analyte. It was found that all the calix[4]arene-poly(para-phenylene ethynylene)s here reported displayed high sensitivities toward the detection of nitrobenzene, 2,4-dinitrotoluene and 2,4,6-trinitrotoluene (TNT). Particularly effective and significant was the response of the films (25-60 nm of thickness) upon exposure to TNT vapor (10 ppb): over 50% of fluorescence quenching was achieved in only 10 s. In contrast, a model polymer lacking the calixarene units showed only reduced quenching activity for the same set of analytes, clearly highlighting the relevance of the macrocyclics in promoting the signaling of the transduction event. The films exhibited high photostability (less than 0.5% loss of fluorescence intensity up to 15 min of continuous irradiation) and the fluorescence quenching sensitivity could be fully recovered after exposure of the quenched films to saturated vapors of hydrazine (the initial fluorescence intensities were usually recovered within 2-5 min of exposure to hydrazine).

Robot visual localization through local feature fusion: an evaluation of multiple classifiers combination approaches

In the last decade, local image features have been widely used in robot visual localization. In order to assess image similarity, a strategy exploiting these features compares raw descriptors extracted from the current image with those in the models of places. This paper addresses the ensuing step in this process, where a combining function must be used to aggregate results and assign each place a score. Casting the problem in the multiple classifier systems framework, in this paper we compare several candidate combiners with respect to their performance in the visual localization task. For this evaluation, we selected the most popular methods in the class of non-trained combiners, namely the sum rule and product rule. A deeper insight into the potential of these combiners is provided through a discriminativity analysis involving the algebraic rules and two extensions of these methods: the threshold, as well as the weighted modifications. In addition, a voting method, previously used in robot visual localization, is assessed. Furthermore, we address the process of constructing a model of the environment by describing how the model granularity impacts upon performance. All combiners are tested on a visual localization task, carried out on a public dataset. It is experimentally demonstrated that the sum rule extensions globally achieve the best performance, confirming the general agreement on the robustness of this rule in other classification problems. The voting method, whilst competitive with the product rule in its standard form, is shown to be outperformed by its modified versions.