Orientador: Camila Coelho Greco / Banca: Fabrizio Caputo / Banca: Dalton Muller Pessoa Filho / Resumo: O principal objetivo deste estudo foi analisar e comparar as respostas metabólicas e cardiorrespiratórias durante o exercício realizado na MLSS e PC em indivíduos com diferentes níveis de treinamento aeróbio no ciclismo. Participaram do estudo 7 ciclistas (C) bem treinados, especializados em provas de estrada e 9 sujeitos não treinados (NT), sem experiência prévia de treinamento no ciclismo. Os voluntários realizaram em dias diferentes os seguintes testes, em um cicloergômetro: 1) teste incremental até a exaustão para a determinação do limiar anaeróbio (Lan), consumo máximo de oxigênio ( O2max) e da intensidade correspondente ao O2max (I O2max); 2) 2 a 4 testes de carga constante de 30 minutos em diferentes intensidades para a determinação da máxima fase estável de lactato sanguíneo (MLSS); 3) 3 testes de carga constante a 95, 100 e 110% I O2max até a exaustão voluntária para a determinação da potência crítica (PC), e; 4) um teste de carga constante até a exaustão na PC. A MLSS foi considerada como a maior intensidade de exercício onde a concentração de lactato não aumentou mais do que 1 mM entre o 10o e o 30o minuto de exercício. Os valores individuais de potência (95, 100 e 110% I O2max) e seu respectivo tempo máximo de exercício (tlim) foram ajustados a partir do modelo hiperbólico de 2 parâmetros para determinação da PC. A PC para ambos os grupos C (318 ± 29W) e NT (200 ± 33W) foi maior significantemente do que a MLSS para os C (288 ± 35W) e NT (169 ± 34W). A MLSS e a PC foram significantemente maiores no grupo C. Em valores relativos ao O2max a MLSS foi maior no grupo C (83 ± 7%) do que no grupo NT (79 ± 6), porém a PC foi similar entre os grupos (91 ± 5% e 90 ± 5%, respectivamente). Da mesma forma, o consumo de oxigênio na PC ( O2PC) foi significantemente maior do que na MLSS ( O2MLSS) nos grupos NT... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: The objective of this study was to analyze and to compare metabolic and cardiorrespiratory responses durin the exercise performed at MLSS and CP in subjects with different aerobic trainning levels in cycling. Participated of this study, 7 well trained cyclists (C), specialized in road events and 9 non-trained subjects (NT), without previous training experience in cycling. Ths subjects performed the following protocols in different days in cyclergometer: 1) incremental test until exhaustion to determine anaerobic threshold (AT), maximal oxygen uptake ( O2max) and the intensity at O2max (I O2max); 2) 2 to 4 constant workload tests in different intensities to determine maximal lactate steadystate (MLSS); 3) 3 constant workload tests at 95, 100 e 110% I O2max until exhaustion to determine critical power (CP), and; 4) constant workload test until exhaustion at CP. MLSS was considered the highest exercise intensity at which the blood lactate concentration did not increase for more than 1 mM between 10th and 30th minute of the exercise. The individual values of power (95, 100 e 110% I O2max) and the respective times (tlim) were adjusted using the hyperbolic model with parameteres to determine CP. CP for C (318 ± 29W) and NT (200 ± 33W) was significantly higher than MLSS in C (288 ± 35W) and NT (169 ± 34W). MLSS and PC were significantly higher in C group. In values relative to O2max, the MLSS was significantly higher in C (83 ± 7%) than NT (79 ± 6), however CP was similar between groups (91 ± 5% and 90 ± 5%, respectively). In the same way, the oxygen uptake at CP ( O2PC) was significantly higher than at MLSS ( O2MLSS) for NT (2627 + 519 e 2323 ± 460 mL.min-1, 11%) e C (3607 ± 505 e 3953 ± 466 mL.min-1, 8%). The slow component at CP in C (375 ± 164 ml.min-1) was similar to NT (412 ± 175 ml.min-1). At this condition, the O2max was not attained (C - 93 ± 5%; NT - 96 ± 7%)... (Complete abstract click electronic access below) / Doutor

Made available in DSpace on 2014-06-11T19:30:53Z (GMT). No. of bitstreams: 0 Previous issue date: 2011-03-15Bitstream added on 2014-06-13T19:00:55Z : No. of bitstreams: 1 carita_rac_me_rcla.pdf: 596317 bytes, checksum: ae963e507d62f687b35122425b6c0d5b (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / O principal objetivo deste estudo foi analisar e comparar as respostas metabólicas e cardiorrespiratórias durante o exercício realizado na MLSS e PC em indivíduos com diferentes níveis de treinamento aeróbio no ciclismo. Participaram do estudo 7 ciclistas (C) bem treinados, especializados em provas de estrada e 9 sujeitos não treinados (NT), sem experiência prévia de treinamento no ciclismo. Os voluntários realizaram em dias diferentes os seguintes testes, em um cicloergômetro: 1) teste incremental até a exaustão para a determinação do limiar anaeróbio (Lan), consumo máximo de oxigênio ( O2max) e da intensidade correspondente ao O2max (I O2max); 2) 2 a 4 testes de carga constante de 30 minutos em diferentes intensidades para a determinação da máxima fase estável de lactato sanguíneo (MLSS); 3) 3 testes de carga constante a 95, 100 e 110% I O2max até a exaustão voluntária para a determinação da potência crítica (PC), e; 4) um teste de carga constante até a exaustão na PC. A MLSS foi considerada como a maior intensidade de exercício onde a concentração de lactato não aumentou mais do que 1 mM entre o 10o e o 30o minuto de exercício. Os valores individuais de potência (95, 100 e 110% I O2max) e seu respectivo tempo máximo de exercício (tlim) foram ajustados a partir do modelo hiperbólico de 2 parâmetros para determinação da PC. A PC para ambos os grupos C (318 ± 29W) e NT (200 ± 33W) foi maior significantemente do que a MLSS para os C (288 ± 35W) e NT (169 ± 34W). A MLSS e a PC foram significantemente maiores no grupo C. Em valores relativos ao O2max a MLSS foi maior no grupo C (83 ± 7%) do que no grupo NT (79 ± 6), porém a PC foi similar entre os grupos (91 ± 5% e 90 ± 5%, respectivamente). Da mesma forma, o consumo de oxigênio na PC ( O2PC) foi significantemente maior do que na MLSS ( O2MLSS) nos grupos NT... / The objective of this study was to analyze and to compare metabolic and cardiorrespiratory responses durin the exercise performed at MLSS and CP in subjects with different aerobic trainning levels in cycling. Participated of this study, 7 well trained cyclists (C), specialized in road events and 9 non-trained subjects (NT), without previous training experience in cycling. Ths subjects performed the following protocols in different days in cyclergometer: 1) incremental test until exhaustion to determine anaerobic threshold (AT), maximal oxygen uptake ( O2max) and the intensity at O2max (I O2max); 2) 2 to 4 constant workload tests in different intensities to determine maximal lactate steadystate (MLSS); 3) 3 constant workload tests at 95, 100 e 110% I O2max until exhaustion to determine critical power (CP), and; 4) constant workload test until exhaustion at CP. MLSS was considered the highest exercise intensity at which the blood lactate concentration did not increase for more than 1 mM between 10th and 30th minute of the exercise. The individual values of power (95, 100 e 110% I O2max) and the respective times (tlim) were adjusted using the hyperbolic model with parameteres to determine CP. CP for C (318 ± 29W) and NT (200 ± 33W) was significantly higher than MLSS in C (288 ± 35W) and NT (169 ± 34W). MLSS and PC were significantly higher in C group. In values relative to O2max, the MLSS was significantly higher in C (83 ± 7%) than NT (79 ± 6), however CP was similar between groups (91 ± 5% and 90 ± 5%, respectively). In the same way, the oxygen uptake at CP ( O2PC) was significantly higher than at MLSS ( O2MLSS) for NT (2627 + 519 e 2323 ± 460 mL.min-1, 11%) e C (3607 ± 505 e 3953 ± 466 mL.min-1, 8%). The slow component at CP in C (375 ± 164 ml.min-1) was similar to NT (412 ± 175 ml.min-1). At this condition, the O2max was not attained (C – 93 ± 5%; NT – 96 ± 7%)... (Complete abstract click electronic access below)

Anomaly detection is the process of detecting samples in a dataset that are atypical or abnormal. Anomaly detection can for example be of great use in an industrial setting, where faults in the manufactured products need to be detected at an early stage. In this setting, the available image data might be from different non-natural domains, such as the depth domain. However, the amount of data available is often limited in these domains. This thesis aims to investigate if a convolutional neural network (CNN) can be trained to perform anomaly detection well on limited datasets in non-natural image domains. The attempted approach is to train the CNN as an autoencoder, in which the CNN is the encoder network. The encoder is then extracted and used as a feature extractor for the anomaly detection task, which is performed using Semantic Pyramid Anomaly Detection (SPADE). The results are then evaluated and analyzed. Two autoencoder models were used in this approach. As the encoder network, one of the models uses a MobileNetV3-Small network that had been pretrained on ImageNet, while the other uses a more basic network, which is a few layers deep and initialized with random weights. Both these networks were trained as regular convolutional autoencoders, as well as variational autoencoders. The results were compared to a MobileNetV3-Small network that had been pretrained on ImageNet, but had not been trained as an autoencoder. The models were tested on six different datasets, all of which contained images from the depth and intensity domains. Three of these datasets additionally contained images from the scatter domain, and for these datasets, the combination of all three domains was tested as well. The main focus was however on the performance in the depth domain. The results show that there is generally an improvement when training the more complex autoencoder on the depth domain. Furthermore, the basic network generally obtains an equivalent result to the more complex network, suggesting that complexity is not necessarily an advantage for this approach. Looking at the different domains, there is no apparent pattern to which domain yields the best performance. This rather seems to depend on the dataset. Lastly, it was found that training the networks as variational autoencoders did generally not improve the performance in the depth domain compared to the regular autoencoders. In summary, an improved anomaly detection was obtained in the depth domain, but for optimal anomaly detection with regard to domain and network, one must look at the individual datasets. / <p>Examensarbetet är utfört vid Institutionen för teknik och naturvetenskap (ITN) vid Tekniska fakulteten, Linköpings universitet</p>