No description available.

No description available.

The ability to walk and negotiate stairs is an important predictor of independent ambulation. The superposition of mobility impairments to the effects of natural aging in persons with stroke render the completion of many daily activities unsafe, thus limiting individuals’ independence within their communities. Currently however, no means exist for the monitoring of mobility levels during daily living in survivors after the completion of rehabilitation programs. The application of inertial sensors for stroke survivors could provide a basis for the study of gait outside of traditional laboratory settings. The main objective of this thesis was to evaluate the performance of inertial sensors in measuring gait of hemiparetic stroke survivors through the completion of three studies. The first study explored the use of inertial measurement units (IMUs) for the measurement of lower limb joint kinematics during stair ascent and descent in both stroke survivors and healthy older adults. Results suggested that IMUs were suitable for the measurement of lower limb range of motion in both healthy and post-stroke subjects during stair ambulation. The second study evaluated the measurement of step length and spatial symmetry during overground walking using IMUs. A systematic error resulting in the underestimation of step lengths calculated using IMUs compared with those measured using video analysis was found, however results suggested that IMUs were suitable for the assessment of spatial symmetry between affected and less-affected limbs in stroke survivors. The final study evaluated the automatic classification of gait activities using inertial sensor data. Findings revealed that the use of a classifier composed of frequency-features extracted from IMU accelerometer and gyroscope data from both the affected and less-affected limbs most accurately identified gait activities from post stroke gait data. This thesis provides a first attempt at applying IMUs to the study of gait post-stroke. Future work may extend the findings of these studies to provide a better understanding to rehabilitation professionals of the demands of everyday life for stroke survivors. / Thesis (Master, Mechanical and Materials Engineering) -- Queen's University, 2013-08-29 12:42:05.505

The Anostraca are a group of crustaceans belonging to the class Branchiopoda. They are predominantly restricted to temporary, inland waterbodies, which in southern Africa, include rockpools, animal wallows, and large "pans". The anostracan fauna of southern Africa was last examined in detail in 1929 by Barnard, and recent collecting in a number of areas indicated the need to update earlier work. A total of four genera, each belonging to a separate family, and forty-six species, some of which had not previously been described, are presented in this thesis. The monogeneric family Streptocephalidae is well represented in Africa, and the fauna of Africa south and north of the Zambezi and Kunene Rivers, as well as Madagascar is reviewed. The Streptocephalus species are characterised by having an S-shaped antennal process, terminating in a cheliform "hand" region. The species were divided into ten groups based on similarities in antennal process morphology. The descriptions of five new species have been published, and an update on distribution and specimen data for a number of species are presented. An additional, as yet unpublished new species from Zimbabwe is presented. The genus Artemia, although well researched in other parts of the world, has been largely ignored in southern Africa. Bisexual populations occur along the Cape coast, and a set of specimens from Namibia, which includes only females, may indicate that A. parthenogenetica is also present on the subcontinent. The frontal knob morphology of the Cape specimens is similar to that of A. tunisiana from northern Africa and the Mediterranean. The family Branchipodidae is characterised by the fusion of the basal joints of the male antennae to form a so-called "clypeus". This family is represented in southern Africa by a diverse fauna of the genus Branchipodopsis. Eleven previously described species, as well as five new species are presented. An attempt was made to divide the sixteen species into species groups, as was done for the streptocephalids, but this proved difficult. The taxonomically useful characters in this genus are largely restricted to the basal processes. An additional character, the presence of expansions of the posterior thoracic segments in the females of some species could be useful. Much intraspecific variation in clypeus morphology was evident, particularly in two of the widespread species. Branchipodopsis species commonly occur in small pools which fill a number of times during the wet season and this has probably led to the development of localised adaptations, and intraspecific variation, or even species endemic to very restricted areas. Three sets of specimens which belong to the family Branchipodidae, but to no known genus, were recently collected from north-eastern Natal, the eastern Cape and the Namib desert in Namibia. These specimens share a single, large process positioned medianly on the fused basal joints, as opposed to a pair of processes in this position. However, the morphology of the penes, and the position and form of other antennal processes could indicate that the specimens are not congeneric. Unfortunately, more material is necessary before the descriptions of two of these possible new genera can be published. Three species of the genus Branchinella, of the family Thamnocephalidae, occur in southern Africa. These species have a well developed frontal process, and fully retractible penes. Bran chin ella spinosa was recorded from the Makgadikgadi Pan in northern Botswana, and was probably introduced to this area from north Africa, or Europe, where it has a wide distribution. The other two southern African species have been collected from few localities, and in small numbers, and this may be connected to cyst hatching processes. The southern African anostracan fauna exhibits distinct distribution patterns and these . appear to be influenced by climate, with rainfall having the dominant influence in the western half of the subcontinent, and temperature influencing distribution in the eastern half. The high altitude Escarpment forms a barrier to the movement of species between the coastal margin and the central plateau region. The formation of the Escarpment could have resulted in vicariance speciation in some anostracans, but in most cases, allopatric speciation appears to have occurred. The type of habitat also influences species distribution in a number of ways. The great anostracan species richness in southern Africa is probably related to the climatic heterogeneity of the subcontinent, as well as the possible origin of two genera in Gondwanaland, and the absence of a Pleistocene glaciation in Africa. Large parts of southern Africa have not been sampled, and the exact status and distribution of many species is uncertain. / Thesis (Ph.D.)-University of Natal, Pietermaritzburg, 1994.

No description available.

Two molecular phylogenetic studies were undertaken to examine relationships within Myricaceae. Analyses of DNA sequences of the plastid rbcL gene, trnL-Fregion and nuclear ITS region showed the family to be monophyletic. In all analyses Canacomyrica, a monotypic genus endemic to New Caledonian that bears several distinctive features such as staminodes in the female flowers, fell into a well- supported clade sister to the rest of Myricaceae. Phylogenetic analyses of ITS and trnL-F sequence data, representing all genera and subgeneric groups, were undertaken using maximum parsimony and Bayesian methods. The following relationships were strongly supported: (Canacomyrica (Comptonia (Myrica, Morella))). The clade containing all species formerly considered to comprise Myrica s.l. was split into two strongly supported clades corresponding to Myrica s.s. and Morella; this finding strengthens the argument for recognition of these as separate genera. Within Morella, two clades corresponded to previously recognized subgenera. Molecular dating analyses were performed using Penalized Likelihood. Close correlations between lineage-specific diversification and major orogenic or climatic events were inferred. This study suggests that much of the diversity in Morella arose during the Neogene and seed-dispersal by birds has been a significant factor in determining the modem distribution. A study of the conservation status of Canacomyrica was conducted using field observations and data from herbarium specimens. This species was found to occur in just eleven fragmented localities: six outside protected areas and three threatened by mining or bush fires. IUCN Red List status of Endangered was recommended. The morphology and ecology of Canacomyrica was studied to enhance knowledge of this poorly known species and provide comparative data for use in a study of the morphology of the entire family. A new classification scheme with keys was presented including, for the first time, Canacomyrica. New combinations in Morella were recommended.

Examining surface shape appearance by touching and observing a lesion from different points of view is a part of the clinical process for skin lesion diagnosis. Motivated by this, we hypothesise that surface shape embodies important information that serves to represent lesion identity and status. A new sensor, Dense Stereo Imaging System (DSIS) allows us to capture 1:1 aligned 3D surface data and 2D colour images simultaneously. This thesis investigates whether the extra surface shape appearance information, represented by features derived from the captured 3D data benefits skin lesion analysis, particularly on the tasks of segmentation and classification. In order to validate the contribution of 3D data to lesion identification, we compare the segmentations resulting from various combinations of images cues (e.g., colour, depth and texture) embedded in a region-based level set segmentation method. The experiments indicate that depth is complementary to colour. Adding the 3D information reduces the error rate from 7:8% to 6:6%. For the purpose of evaluating the segmentation results, we propose a novel ground truth estimation approach that incorporates a prior pattern analysis of a set of manual segmentations. The experiments on both synthetic and real data show that this method performs favourably compared to the state of the art approach STAPLE [1] on ground truth estimation. Finally, we explore the usefulness of 3D information to non-melanoma lesion diagnosis by tests on both human and computer based classifications of five lesion types. The results provide evidence for the benefit of the additional 3D information, i.e., adding the 3D-based features gives a significantly improved classification rate of 80:7% compared to only using colour features (75:3%). The three main contributions of the thesis are improved methods for lesion segmentation, non-melanoma lesion classification and lesion boundary ground-truth estimation.

This thesis presents a way to generate a Digital Terrain Model (dtm) from a Digital Surface Model (dsm) and multi spectral images (including the Near Infrared (nir) color band). An Artificial Neural Network (ann) is used to pre-classify the dsm and multi spectral images. This in turn is used to filter the dsm to a dtm. The use of an ann as a classifier provided good results. Additionally, the addition of the nir color band resulted in an improvement of the accuracy of the classifier. Using the classifier, a dtm was easily extracted without removing natural edges or height variations in the forests and cities. These challenges are handled with great satisfaction as compared to earlier methods.

Three autonomous cropping and feature extraction algorithms are examined that can be used for classification of low probability of intercept radar modulations using time-frequency (T-F) images. The first approach, Erosion Dilation Adaptive Binarization (EDAB), uses erosion and a new adaptive threshold binarization algorithm embedded within a recursive dilation process to determine the modulation energy centroid (radar's carrier frequency) and properly place a fixed-width cropping window. The second approach, Marginal Frequency Adaptive Binarization (MFAB), uses the marginal frequency distribution and the adaptive threshold binarization algorithm to determine the start and stop frequencies of the modulation energy to locate and adapt the size of the cropping window. The third approach, Fast Image Filtering, uses the fast Fourier transform and a Gaussian lowpass filter to isolate the modulation energy. The modulation is then cropped from the original T-F image and the adaptive binarization algorithm is used again to compute a binary feature vector for input into a classification network. The binary feature vector allows the image detail to be preserved without overwhelming the classification network that follows. A multi-layer perceptron and a radial basis function network are used for classification and the results are compared. Classification results for nine simulated radar modulations are shown to demonstrate the three feature-extraction approaches and quantify the performance of the algorithms. It is shown that the best results are obtained using the Choi-Williams distribution followed by the MFAB algorithm and a multi-layer perceptron. This setup produced an overall percent correct classification (Pcc) of 87.2% for testing with noise variation and 77.8% for testing with modulation variation. In an operational context, the ability to process and classify LPI signals autonomously allows the operator in the field to receive real-time results.

Master of Science / Department of Computing and Information Sciences / Doina Caragea / Machine learning is the subfield of Artificial intelligence that gives computers the ability to learn without being explicitly programmed, as it was defined by Arthur Samuel - the American pioneer in the field of computer gaming and artificial intelligence who was born in Emporia, Kansas. Supervised Machine Learning is focused on building predictive models given labeled training data. Data may come from a variety of sources, for instance, social media networks. In our research, we use Twitter data, specifically, user-generated tweets about disasters such as floods, hurricanes, terrorist attacks, etc., to build classifiers that could help disaster management teams identify useful information. A supervised classifier trained on data (training data) from a particular domain (i.e. disaster) is expected to give accurate predictions on unseen data (testing data) from the same domain, assuming that the training and test data have similar characteristics. Labeled data is not easily available for a current target disaster. However, labeled data from a prior source disaster is presumably available, and can be used to learn a supervised classifier for the target disaster. Unfortunately, the source disaster data and the target disaster data may not share the same characteristics, and the classifier learned from the source may not perform well on the target. Domain adaptation techniques, which use unlabeled target data in addition to labeled source data, can be used to address this problem. We study single-source and multi-source domain adaptation techniques, using Nave Bayes classifier. Experimental results on Twitter datasets corresponding to six disasters show that domain adaptation techniques improve the overall performance as compared to basic supervised learning classifiers. Domain adaptation is crucial for many machine learning applications, as it enables the use of unlabeled data in domains where labeled data is not available.