Design strategies for an artificial neural network based algorithm for automatic incident detection on major arterial streets

Traffic incidents are non-recurring events that can cause a temporary reduction in roadway capacity. They have been recognized as a major contributor to traffic congestion on our nation’s highway systems. To alleviate their impacts on capacity, automatic incident detection (AID) has been applied as an incident management strategy to reduce the total incident duration. AID relies on an algorithm to identify the occurrence of incidents by analyzing real-time traffic data collected from surveillance detectors. Significant research has been performed to develop AID algorithms for incident detection on freeways; however, similar research on major arterial streets remains largely at the initial stage of development and testing. This dissertation research aims to identify design strategies for the deployment of an Artificial Neural Network (ANN) based AID algorithm for major arterial streets. A section of the US-1 corridor in Miami-Dade County, Florida was coded in the CORSIM microscopic simulation model to generate data for both model calibration and validation. To better capture the relationship between the traffic data and the corresponding incident status, Discrete Wavelet Transform (DWT) and data normalization were applied to the simulated data. Multiple ANN models were then developed for different detector configurations, historical data usage, and the selection of traffic flow parameters. To assess the performance of different design alternatives, the model outputs were compared based on both detection rate (DR) and false alarm rate (FAR). The results show that the best models were able to achieve a high DR of between 90% and 95%, a mean time to detect (MTTD) of 55-85 seconds, and a FAR below 4%. The results also show that a detector configuration including only the mid-block and upstream detectors performs almost as well as one that also includes a downstream detector. In addition, DWT was found to be able to improve model performance, and the use of historical data from previous time cycles improved the detection rate. Speed was found to have the most significant impact on the detection rate, while volume was found to contribute the least. The results from this research provide useful insights on the design of AID for arterial street applications.

Synthesis of azulenylsilane nitrones as diagnostic tools for superoxide detection

The superoxide radical is considered to play important roles in physiological processes as well as in the genesis of diverse cytotoxic conditions such as cancer, various cardiovascular disorders and neurodegenerative diseases such as amyotrophic lateral sclerosis (ALS), Parkinson’s disease (PD) and Alzheimer’s disease (AD). The detection and quantification of superoxide within cells is of critical importance to understand biological roles of superoxide and to develop preventive strategies against free radical-mediated diseases. Cyclic nitrone spin traps such as DMPO, EMPO, DEPMPO, BMPO and their derivatives have been widely used in conjunction with ESR spectroscopy to detect cellular superoxide with some success. However, the formation of unstable superoxide adducts from the reaction of cyclic nitrones with superoxide is a stumbling block in detecting superoxide by using electron spin resonance (ESR). A chemiluminescent probe, lucigenin, and fluorogenic probes, hydroethidium and MitoSox, are the other frequently used methods in detecting superoxide. However, luceginen undergoes redox-cycling producing superoxide by itself, and hydroethidium and MitoSox react with other oxidants apart from superoxide forming red fluorescent products contributing to artefacts in these assays. Hence, both methods were deemed to be inappropriate for superoxide detection. In this study, an effective approach, a selective mechanism-based colorimetric detection of superoxide anion has been developed by using silylated azulenyl nitrones spin traps. Since a nitrone moiety and an adjacent silyl group react readily with radicals and oxygen anions respectively, such nitrones can trap superoxide efficiently because superoxide is both a radical and an oxygen anion. Moreover, the synthesized nitrone is designed to be triggered solely by superoxide and not by other commonly observed oxygen radicals such as hydroxyl radical, alkoxyl radicals and peroxyl radical. In vitro studies have shown that these synthesized silylated azylenyl nitrones and the mitochondrial-targeted guanylhydrazone analog can trap superoxide efficiently yielding UV-vis identifiable and even potentially fluorescence-detectable orange products. Therefore, the chromotropic detection of superoxide using these nitrones can be a promising method in contrast to other available methods.

Infants' detection of synchrony between sounds and pauses in the movement of an object

The present study investigated the development of sensitivity to temporal synchrony between sounds of impact and pauses in the movement of an object by infants of 2 1/2, 4 and 6 months of age. Ninety infants were tested across four experiments with side-by-side videos of a red and white square and a blue and yellow triangle along with a centralized soundtrack which was synchronized with only one of the films. This preference phase was then followed by a search phase, where the two films were accompanied by intermittent bursts of the soundtrack from each object. Twomonth- olds showed no evidence of matching films and soundtracks on the basis of synchrony, however 4-month-olds looked more on the second block of trials to the object which paused when the sound occurred and directed more first looks during the preference phase to the matching object. Six-month-olds demonstrated significantly more first looks to the mismatched object during the search phase only. These results suggest that infants relate impact sounds with synchronous pauses in continuous motion by the age of four months.

Design Strategies for an Artificial Neural Network Based Algorithm for Automatic Incident Detection on Major Arterial Streets

Traffic incidents are non-recurring events that can cause a temporary reduction in roadway capacity. They have been recognized as a major contributor to traffic congestion on our national highway systems. To alleviate their impacts on capacity, automatic incident detection (AID) has been applied as an incident management strategy to reduce the total incident duration. AID relies on an algorithm to identify the occurrence of incidents by analyzing real-time traffic data collected from surveillance detectors. Significant research has been performed to develop AID algorithms for incident detection on freeways; however, similar research on major arterial streets remains largely at the initial stage of development and testing. This dissertation research aims to identify design strategies for the deployment of an Artificial Neural Network (ANN) based AID algorithm for major arterial streets. A section of the US-1 corridor in Miami-Dade County, Florida was coded in the CORSIM microscopic simulation model to generate data for both model calibration and validation. To better capture the relationship between the traffic data and the corresponding incident status, Discrete Wavelet Transform (DWT) and data normalization were applied to the simulated data. Multiple ANN models were then developed for different detector configurations, historical data usage, and the selection of traffic flow parameters. To assess the performance of different design alternatives, the model outputs were compared based on both detection rate (DR) and false alarm rate (FAR). The results show that the best models were able to achieve a high DR of between 90% and 95%, a mean time to detect (MTTD) of 55-85 seconds, and a FAR below 4%. The results also show that a detector configuration including only the mid-block and upstream detectors performs almost as well as one that also includes a downstream detector. In addition, DWT was found to be able to improve model performance, and the use of historical data from previous time cycles improved the detection rate. Speed was found to have the most significant impact on the detection rate, while volume was found to contribute the least. The results from this research provide useful insights on the design of AID for arterial street applications.