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.

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.

Development of prediction models for freeway incident durations using data mining techniques

The nation's freeway systems are becoming increasingly congested. A major contribution to traffic congestion on freeways is due to traffic incidents. Traffic incidents are non-recurring events such as accidents or stranded vehicles that cause a temporary roadway capacity reduction, and they can account for as much as 60 percent of all traffic congestion on freeways. One major freeway incident management strategy involves diverting traffic to avoid incident locations by relaying timely information through Intelligent Transportation Systems (ITS) devices such as dynamic message signs or real-time traveler information systems. The decision to divert traffic depends foremost on the expected duration of an incident, which is difficult to predict. In addition, the duration of an incident is affected by many contributing factors. Determining and understanding these factors can help the process of identifying and developing better strategies to reduce incident durations and alleviate traffic congestion. A number of research studies have attempted to develop models to predict incident durations, yet with limited success. ^ This dissertation research attempts to improve on this previous effort by applying data mining techniques to a comprehensive incident database maintained by the District 4 ITS Office of the Florida Department of Transportation (FDOT). Two categories of incident duration prediction models were developed: "offline" models designed for use in the performance evaluation of incident management programs, and "online" models for real-time prediction of incident duration to aid in the decision making of traffic diversion in the event of an ongoing incident. Multiple data mining analysis techniques were applied and evaluated in the research. The multiple linear regression analysis and decision tree based method were applied to develop the offline models, and the rule-based method and a tree algorithm called M5P were used to develop the online models. ^ The results show that the models in general can achieve high prediction accuracy within acceptable time intervals of the actual durations. The research also identifies some new contributing factors that have not been examined in past studies. As part of the research effort, software code was developed to implement the models in the existing software system of District 4 FDOT for actual applications. ^

Predicting delay reductions from freeway traffic diversion for incident management

Traffic incidents are a major source of traffic congestion on freeways. Freeway traffic diversion using pre-planned alternate routes has been used as a strategy to reduce traffic delays due to major traffic incidents. However, it is not always beneficial to divert traffic when an incident occurs. Route diversion may adversely impact traffic on the alternate routes and may not result in an overall benefit. This dissertation research attempts to apply Artificial Neural Network (ANN) and Support Vector Regression (SVR) techniques to predict the percent of delay reduction from route diversion to help determine whether traffic should be diverted under given conditions. The DYNASMART-P mesoscopic traffic simulation model was applied to generate simulated data that were used to develop the ANN and SVR models. A sample network that comes with the DYNASMART-P package was used as the base simulation network. A combination of different levels of incident duration, capacity lost, percent of drivers diverted, VMS (variable message sign) messaging duration, and network congestion was simulated to represent different incident scenarios. The resulting percent of delay reduction, average speed, and queue length from each scenario were extracted from the simulation output. The ANN and SVR models were then calibrated for percent of delay reduction as a function of all of the simulated input and output variables. The results show that both the calibrated ANN and SVR models, when applied to the same location used to generate the calibration data, were able to predict delay reduction with a relatively high accuracy in terms of mean square error (MSE) and regression correlation. It was also found that the performance of the ANN model was superior to that of the SVR model. Likewise, when the models were applied to a new location, only the ANN model could produce comparatively good delay reduction predictions under high network congestion level.

Predicting delay reductions from freeway traffic diversion for incident management

Traffic incidents are a major source of traffic congestion on freeways. Freeway traffic diversion using pre-planned alternate routes has been used as a strategy to reduce traffic delays due to major traffic incidents. However, it is not always beneficial to divert traffic when an incident occurs. Route diversion may adversely impact traffic on the alternate routes and may not result in an overall benefit. This dissertation research attempts to apply Artificial Neural Network (ANN) and Support Vector Regression (SVR) techniques to predict the percent of delay reduction from route diversion to help determine whether traffic should be diverted under given conditions. The DYNASMART-P mesoscopic traffic simulation model was applied to generate simulated data that were used to develop the ANN and SVR models. A sample network that comes with the DYNASMART-P package was used as the base simulation network. A combination of different levels of incident duration, capacity lost, percent of drivers diverted, VMS (variable message sign) messaging duration, and network congestion was simulated to represent different incident scenarios. The resulting percent of delay reduction, average speed, and queue length from each scenario were extracted from the simulation output. The ANN and SVR models were then calibrated for percent of delay reduction as a function of all of the simulated input and output variables. The results show that both the calibrated ANN and SVR models, when applied to the same location used to generate the calibration data, were able to predict delay reduction with a relatively high accuracy in terms of mean square error (MSE) and regression correlation. It was also found that the performance of the ANN model was superior to that of the SVR model. Likewise, when the models were applied to a new location, only the ANN model could produce comparatively good delay reduction predictions under high network congestion level.

Evaluation of an innovative critical power model in intermittent vertical jump

The aim of this study was to test if the critical power model can be used to determine the critical rest interval (CRI) between vertical jumps. Ten males performed intermittent countermovement jumps on a force platform with different resting periods (4.1 +/- 0.3 s, 5.0 +/- 0.4 s, 5.9 +/- 0.6 s). Jump trials were interrupted when participants could no longer maintain 95% of their maximal jump height. After interruption, number of jumps, total exercise duration and total external work were computed. Time to exhaustion (s) and total external work (J) were used to solve the equation Work = a + b . time. The CRI (corresponding to the shortest resting interval that allowed jump height to be maintained for a long time without fatigue) was determined dividing the average external work needed to jump at a fixed height (J) by b parameter (J/s). in the final session, participants jumped at their calculated CRI. A high coefficient of determination (0.995 +/- 0.007) and the CRI (7.5 +/- 1.6 s) were obtained. In addition, the longer the resting period, the greater the number of jumps (44 13, 71 28, 105 30, 169 53 jumps; p<0.0001), time to exhaustion (179 +/- 50, 351 +/- 120, 610 +/- 141, 1,282 +/- 417 s; p<0.0001) and total external work (28.0 +/- 8.3, 45.0 +/- 16.6, 67.6 +/- 17.8, 111.9 +/- 34.6 kJ; p<0.0001). Therefore, the critical power model may be an alternative approach to determine the CRI during intermittent vertical jumps.

Improving wheat simulation capabilities in Australia from a cropping systems perspective: water and nitrogen effects on spring wheat in a semi-arid environment

Systems approaches can help to evaluate and improve the agronomic and economic viability of nitrogen application in the frequently water-limited environments. This requires a sound understanding of crop physiological processes and well tested simulation models. Thus, this experiment on spring wheat aimed to better quantify water x nitrogen effects on wheat by deriving some key crop physiological parameters that have proven useful in simulating crop growth. For spring wheat grown in Northern Australia under four levels of nitrogen (0 to 360 kg N ha(-1)) and either entirely on stored soil moisture or under full irrigation, kernel yields ranged from 343 to 719 g m(-2). Yield increases were strongly associated with increases in kernel number (9150-19950 kernels m(-2)), indicating the sensitivity of this parameter to water and N availability. Total water extraction under a rain shelter was 240 mm with a maximum extraction depth of 1.5 m. A substantial amount of mineral nitrogen available deep in the profile (below 0.9 m) was taken up by the crop. This was the source of nitrogen uptake observed after anthesis. Under dry conditions this late uptake accounted for approximately 50% of total nitrogen uptake and resulted in high (>2%) kernel nitrogen percentages even when no nitrogen was applied,Anthesis LAI values under sub-optimal water supply were reduced by 63% and under sub-optimal nitrogen supply by 50%. Radiation use efficiency (RUE) based on total incident short-wave radiation was 1.34 g MJ(-1) and did not differ among treatments. The conservative nature of RUE was the result of the crop reducing leaf area rather than leaf nitrogen content (which would have affected photosynthetic activity) under these moderate levels of nitrogen limitation. The transpiration efficiency coefficient was also conservative and averaged 4.7 Pa in the dry treatments. Kernel nitrogen percentage varied from 2.08 to 2.42%. The study provides a data set and a basis to consider ways to improve simulation capabilities of water and nitrogen effects on spring wheat. (C) 1997 Elsevier Science B.V.

Analysis of exploratory eye movements in patients with schizophrenia during visual scanning of projective tests' figures

OBJECTIVE: Compare pattern of exploratory eye movements during visual scanning of the Rorschach and TAT test cards in people with schizophrenia and controls. METHOD: 10 participants with schizophrenia and 10 controls matched by age, schooling and intellectual level participated in the study. Severity of symptoms was evaluated with the Positive and Negative Syndrome Scale. Test cards were divided into three groups: TAT cards with scenes content, TAT cards with interaction content (TAT-faces), and Rorschach cards with abstract images. Eye movements were analyzed for: total number, duration and location of fixation; and length of saccadic movements. RESULTS: Different pattern of eye movement was found, with schizophrenia participants showing lower number of fixations but longer fixation duration in Rorschach cards and TAT-faces. The biggest difference was observed in Rorschach, followed by TAT-faces and TAT-scene cards. CONCLUSIONS: Results suggest alteration in visual exploration mechanisms possibly related to integration of abstract visual information.

Changes in Running Mechanics and Spring-Mass Behaviour during a 5-km Time Trial

Research into the biomechanical manifestation of fatigue during exhaustive runs is increasingly popular but additional understanding of the adaptation of the spring-mass behaviour during the course of strenuous, self-paced exercises continues to be a challenge in order to develop optimized training and injury prevention programs. This study investigated continuous changes in running mechanics and spring-mass behaviour during a 5-km run. 12 competitive triathletes performed a 5-km running time trial (mean performance: 17 min 30 s) on a 200 m indoor track. Vertical and anterior-posterior ground reaction forces were measured every 200 m by a 5-m long force platform system, and used to determine spring-mass model characteristics. After a fast start, running velocity progressively decreased (- 11.6%; P<0.001) in the middle part of the race before an end spurt in the final 400-600 m. Stride length (- 7.4%; P<0.001) and frequency (- 4.1%; P=0.001) decreased over the 25 laps, while contact time (+ 8.9%; P<0.001) and total stride duration (+ 4.1%; P<0.001) progressively lengthened. Peak vertical forces (- 2.0%; P<0.01) and leg compression (- 4.3%; P<0.05), but not centre of mass vertical displacement (+ 3.2%; P>0.05), decreased with time. As a result, vertical stiffness decreased (- 6.0%; P<0.001) during the run, whereas leg stiffness changes were not significant (+ 1.3%; P>0.05). Spring-mass behaviour progressively changes during a 5-km time trial towards deteriorated vertical stiffness, which alters impact and force production characteristics.

How does a hilly urban environment influence daily physical activity in obese individuals?

Background: Increases in physical activity (PA) are promoted by walking in an outdoor environment. Along with walking speed, slope is a major determinant of exercise intensity, and energy expenditure. The hypothesis was that in free-living conditions, a hilly environment diminishes PA to a greater extent in obese (OB) when compared with control (CO) individuals. Methods: To assess PA types and patterns, 28 CO (22 ± 2 kg/m2) and 14 OB (33 ± 4 kg/m2) individuals wore during an entire day 2 accelerometers and 1 GPS device, around respectively their waist, ankle and shoulder. They performed their usual PA and were asked to walk an additional 60 min per day. Results: The duration of inactivity and activity with OB individuals tended to be, respectively, higher and lower than that of CO individuals (P = .06). Both groups spent less time walking uphill/downhill than on the level (20%, 19%, vs. 61% of total walking duration, respectively, P < .001). However OB individuals spent less time walking uphill/downhill per day than CO (25 ± 15 and 38 ± 15 min/d, respectively, P < 0.05) and covered a shorter distance per day (3.8 km vs 5.2 km, P < 0.01). Conclusions: BMI and outdoor topography should also be considered when prescribing extra walking in free-living conditions.

Respiratory responses during affective picture viewing

Previous research has demonstrated covariation of physiological responding with judgments of valence and arousal. However, until now links between these affective dimensions and respiratory measures have not been extensively investigated. In this study, eight picture series of different affective valence and arousal level were shown to 30 subjects, while respiration, skin conductance level (SCL), heart rate (HR) and affective judgments were measured. With increasing pleasantness, inspiratory time lengthened, mean inspiratory flow decreased and thoracic breathing increased. With increasing arousal, inspiratory time and total breath duration shortened and mean inspiratory flow, minute ventilation, thoracic breathing and electrodermal activity increased. These findings confirm the importance of arousal in respiratory responding, but also indicate a modulatory role of affective valence.We propose that the arousal effects reflect energy mobilization in preparation to act, and thatthe valence effects might be a manifestation of an attention bias toward negative stimuli. [Authors]

Psychophysiological activation during preparation, performance, and recovery in high- and low-anxious music students

The present study provides a comprehensive view of (a) the time dynamics of the psychophysiological responding in performing music students (n = 66) before, during, and after a private and a public performance and (b) the moderating effect of music performance anxiety (MPA). Heart rate (HR), minute ventilation (VE), and all affective and somatic self-report variables increased in the public session compared to the private session. Furthermore, the activation of all variables was stronger during the performances than before or after. Differences between phases were larger in the public than in the private session for HR, VE, total breath duration, anxiety, and trembling. Furthermore, while higher MPA scores were associated with higher scores and with larger changes between sessions and phases for self-reports, this association was less coherent for physiological variables. Finally, self-reported intra-individual performance improvements or deteriorations were not associated with MPA. This study makes a novel contribution by showing how the presence of an audience influences low- and high-anxious musicians' psychophysiological responding before, during and after performing. Overall, the findings are more consistent with models of anxiety that emphasize the importance of cognitive rather than physiological factors in MPA.

Repeated sprinting on natural grass impairs vertical stiffness but does not alter plantar loading in soccer players.

This study aimed to determine changes in spring-mass model (SMM) characteristics, plantar pressures, and muscle activity induced by the repetition of sprints in soccer-specific conditions; i.e., on natural grass with soccer shoes. Thirteen soccer players performed 6 × 20 m sprints interspersed with 20 s of passive recovery. Plantar pressure distribution was recorded via an insole pressure recorder device divided into nine areas for analysis. Stride temporal parameters allowed to estimate SMM characteristics. Surface electromyographic activity was monitored for vastus lateralis, rectus femoris, and biceps femoris muscles. Sprint time, contact time, and total stride duration lengthened from the first to the last repetition (+6.7, +12.9, and +9.3%; all P < 0.05), while flight time, swing time, and stride length remained constant. Stride frequency decrease across repetitions approached significance (-6.8%; P = 0.07). No main effect of the sprint number or any significant interaction between sprint number and foot region was found for maximal force, mean force, peak pressure and mean pressure (all P > 0.05). Center of mass vertical displacement increased (P < 0.01) with time, together with unchanged (both P > 0.05) peak vertical force and leg compression. Vertical stiffness decreased (-15.9%; P < 0.05) across trials, whereas leg stiffness changes were not significant (-5.9%; P > 0.05). Changes in root mean square activity of the three tested muscles over sprint repetitions were not significant. Although repeated sprinting on natural grass with players wearing soccer boots impairs their leg-spring behavior (vertical stiffness), there is no substantial concomitant alterations in muscle activation levels or plantar pressure patterns.

Management and prognosis of status epilepticus according to hospital setting: a prospective study.

BACKGROUND: The treatment of status epilepticus (SE) is based on relatively little evidence although several guidelines have been published. A recent study reported a worse SE prognosis in a large urban setting as compared to a peripheral hospital, postulating better management in the latter. The aim of this study was to analyse SE episodes occurring in different settings and address possible explanatory variables regarding outcome, including treatment quality. METHODS: Over six months we prospectively recorded consecutive adults with SE (fit lasting five or more minutes) at the Centre Hospitalier Universitaire Vaudois (CHUV) and in six peripheral hospitals (PH) in the same region. Demographical, historical and clinical variables were collected, including SE severity estimation (STESS score) and adherence to Swiss SE treatment guidelines. Outcome at discharge was categorised as "good" (return to baseline), or "poor" (persistent neurological sequelae or death). RESULTS: Of 54 patients (CHUV: 36; PH 18), 33% had a poor outcome. Whilst age, SE severity, percentage of SE episodes lasting less than 30 minutes and total SE duration were similar, fewer patients had a good outcome at the CHUV (61% vs 83%; OR 3.57; 95% CI 0.8-22.1). Mortality was 14% at the CHUV and 5% at the PH. Most treatments were in agreement with national guidelines, although less often in PH (78% vs 97%, P = 0.04). CONCLUSION: Although not statistically significant, we observed a slightly worse SE prognosis in a large academic centre as compared to smaller hospitals. Since SE severity was similar in the two settings but adherence to national treatment guidelines was higher in the academic centre, further investigation on the prognostic role of SE treatment and outcome determinants is required.