A deep reinforcement learning-based intelligent intervention framework for real-time proactive road safety management.

We propose a variable speed limit (VSL) system for improving the safety of urban expressways in real time. The system has two main functions: monitoring traffic data and then using the data to assess crash risk through a real-time crash prediction model (RTCPM). When the risk is high, the system triggers VSL control to restore traffic conditions to normal. The study addresses several weaknesses in existing VSL-based real-time safety interventions. Existing models are not widely applicable due to varying detector spacing among different freeways, and even within a study area. Therefore, with the existing detector spacing as an input, a cell transmission model (CTM) is used to simulate traffic states for the desired cell size. A dynamic Bayesian network (DBN) is used for modeling in the RTCPM. The proposed CTM model is then modified to allow VSL control. Whereas existing studies selected various VSL strategies from a predefined list, we employ a deep Q-network, which is a reinforcement learning-based machine learning algorithm, for the VSL control. Two busy segments of the Tokyo Metropolitan Expressway were used as the study area. After several iterations, our proposed real-time system reduced the crash risk by 19%.

Real-time crash prediction models: State-of-the-art, design pathways and ubiquitous requirements.

Proactive traffic safety management systems can monitor traffic conditions in real-time, identify the formation of unsafe traffic dynamics, and implement suitable interventions to bring unsafe conditions back to normal traffic situations. Recent advancements in artificial intelligence, sensor fusion and algorithms have brought about the introduction of a proactive safety management system closer to reality. The basic prerequisite for developing such a system is to have a reliable crash prediction model that takes real-time traffic data as input and evaluates their association with crash risk. Since the early 21st century, several studies have focused on developing such models. Although the idea has considerably matured over time, the endeavours have been quite discrete and fragmented at best because the fundamental aspects of the overall modelling approach substantially vary. Therefore, a number of transitional challenges have to be identified and subsequently addressed before a ubiquitous proactive safety management system can be formulated, designed and implemented in real-world scenarios. This manuscript conducts a comprehensive review of existing real-time crash prediction models with the aim of illustrating the state-of-the-art and systematically synthesizing the thoughts presented in existing studies in order to facilitate its translation from an idea into a ready to use technology. Towards that journey, it conducts a systematic review by applying various text mining methods and topic modelling. Based on the findings, this paper ascertains the development pathways followed in various studies, formulates the ubiquitous design requirements of such models from existing studies and knowledge of similar systems. Finally, this study evaluates the universality and design compatibility of existing models. This paper is, therefore, expected to serve as a one stop knowledge source for facilitating a faster transition from the idea of real-time crash prediction models to a real-world operational proactive traffic safety management system.

Data mining in road crash analysis: the context of developing countries.

The recent advancements in the field of data mining have made vast progress in extracting new information and hidden patterns from large datasets which are often overlooked by the traditional statistical approaches. These methods focus on searching for new and interesting hypothesis which were previously unobserved. Road safety researchers working with the crash data from developed world have seen encouraging success in obtaining new insight into crash mechanism through data mining. An attempt was made in this study to apply these advance methods and evaluate their performance in manifesting crash causes for Bangladesh. The study applies hierarchical clustering to identify hazardous clusters, random forest to find important variables explaining each of these clusters, and classification and regression trees to unveil their respective crash mechanisms for the road crash data of Bangladesh. The results identified several new interesting relationships and acknowledged issues related to quality of data.

Understanding crash mechanism on urban expressways using high-resolution traffic data.

Urban expressways play a vital role in the modern mega cities by serving peak hour traffic alongside reducing travel time for moderate to long distance intra-city trips. Thus, ensuring safety on these roads holds high priority. Little knowledge has been acquired till date regarding crash mechanism on these roads. This study uses high-resolution traffic data collected from the detectors to identify factors influencing crash. It also identifies traffic patterns associated with different types of crashes and explains crash phenomena thereby. Unlike most of the previous studies on conventional expressways, the research separately investigates the basic freeway segments (BFS) and the ramp areas. The study employs random multinomial logit, a random forest of logit models, to rank the variables; expectation maximization clustering algorithm to identify crash prone traffic patterns and classification and regression trees to explain crash phenomena. As accentuated by the study outcome, crash mechanism is not generic throughout the expressway and it varies from the BFS to the ramp vicinities. The level of congestion and speed difference between upstream and downstream traffic best explains crashes and their types for the BFS, whereas, the ramp flow has the highest influence in determining the types of crashes within the ramp vicinities. The paper also discusses about the applicability of different countermeasures, such as, variable speed limits, temporary restriction on lane changing, posting warnings, etc., to attenuate different patterns of hazardous traffic conditions. The study outcome can be utilized in designing location and traffic condition specific proactive road safety management systems for urban expressways.

PET/CT for staging and follow-up of pediatric nasopharyngeal carcinoma.

PURPOSE: While FDG PET/CT for the evaluation of nasopharyngeal carcinoma (NPC) in adult patients has documented advantages and disadvantages compared with conventional imaging, to our knowledge, no studies of FDG PET/CT for the evaluation of NPC in pediatric patients have been performed. In this investigation, we studied the utility of FDG PET/CT in children with NPC. METHODS: The study group comprised 18 children with biopsy-proven NPC who underwent FDG PET/CT and MRI (total 38 pairs of images). All baseline and follow-up FDG PET/CT and MRI studies were independently reviewed for restaging of disease. RESULTS: The concordance between FDG PET/CT and MRI in T, N, and overall staging was 29%, 64%, and 43%, respectively. Compared with MRI, FDG PET/CT yielded lower T and overall staging and showed less cervical and retropharyngeal lymphadenopathy. The concordance between follow-up FDG PET/CT and MRI was 79% overall and 100% 9 months after therapy. In patients who achieved complete remission, FDG PET/CT showed disease clearance 3-6 months earlier than MRI. There were no false-positive or false-negative FDG PET/CT scans during follow-up. CONCLUSION: FDG PET/CT may underestimate tumor extent and regional lymphadenopathy compared with MRI at the time of diagnosis, but it helps to detect metastases and clarify ambiguous findings. FDG PET/CT is sensitive and specific for follow-up and enables earlier determination of disease remission. FDG PET/CT is a valuable imaging modality for the evaluation and monitoring of NPC in pediatric patients.

A Bayesian network based framework for real-time crash prediction on the basic freeway segments of urban expressways.

The concept of measuring the crash risk for a very short time window in near future is gaining more practicality due to the recent advancements in the fields of information systems and traffic sensor technology. Although some real-time crash prediction models have already been proposed, they are still primitive in nature and require substantial improvements to be implemented in real-life. This manuscript investigates the major shortcomings of the existing models and offers solutions to overcome them with an improved framework and modeling method. It employs random multinomial logit model to identify the most important predictors as well as the most suitable detector locations to acquire data to build such a model. Afterwards, it applies Bayesian belief net (BBN) to build the real-time crash prediction model. The model has been constructed using high resolution detector data collected from Shibuya 3 and Shinjuku 4 expressways under the jurisdiction of Tokyo Metropolitan Expressway Company Limited, Japan. It has been specifically built for the basic freeway segments and it predicts the chance of formation of a hazardous traffic condition within the next 4-9 min for a particular 250 meter long road section. The performance evaluation results reflect that at an average threshold value the model is able to successful classify 66% of the future crashes with a false alarm rate less than 20%.

Decreased thyroid uptake of Tc-99m pertechnetate in patients with advanced-stage Sjögren syndrome: evaluation using salivary gland scintigraphy.

PURPOSE: The authors assessed the uptake of Tc-99m pertechnetate in the thyroid using salivary gland scintigraphy in patients with Sjögren syndrome and in healthy controls. MATERIALS AND METHODS: Salivary gland scintigraphy and a labial biopsy were performed in 73 patients with Sjögren syndrome. Based on the labial biopsy findings, 32 patients with a histopathologic grade of 1 or 2 were regarded as having early-stage Sjögren syndrome and 41 patients with a grade of 3 or 4 were regarded as having an advanced stage. After the administration of 370 MBq (10 mCi) Tc-99m pertechnetate, dynamic salivary gland scintigraphy was performed for 50 minutes. Lemon juice was used to stimulate the salivary glands, and the thyroid gland was included in the imaging area. Scintigraphy was also performed in an age- and sex-matched control group of 25 healthy persons. The thyroid uptake ratio was calculated for the scintigraphic images and compared among the three groups: healthy controls, patients with early-stage Sjögren syndrome, and those with advanced-stage Sjögren syndrome. RESULTS: When compared with the control group, the thyroid uptake ratio of the early-stage Sjögren syndrome group was not significantly different, whereas that of the advanced-stage group was significantly lower. CONCLUSIONS: Thyroid uptake of Tc-99m pertechnetate was less in patients with advanced-stage Sjögren syndrome than in patients with early-stage Sjögren syndrome or in healthy controls. Measuring the thyroid uptake of Tc-99m pertechnetate using salivary gland scintigraphy is an easy and useful method for assessing thyroid disorders in Sjögren syndrome and thus should be performed routinely.

Maintained Salivary Function after Brachytherapy in Patients with Head and Neck Carcinomas - Evaluation Using Quantitative Salivary Gland Scintigraphy.

The purpose of this study was to determine whether or not salivary gland dysfunction occurs within the first three months after brachytherapy in patients with head and neck carcinoma. Of the 20 patients with head and neck squamous cell carcinoma included in this study, 11 were treated with brachytherapy and the remaining 9 patients received external irradiation. All the patients underwent a salivary gland scintigraphy before and after radiotherapy. The scintigraphic parameters of each major salivary gland were then compared before and after the radiotherapy. In the brachytherapy group, none of the scintigraphic functional parameters showed a significant change before and after the radiotherapy. In contrast, all of the parameters with the exception of the uptake ratio (UR) of the submandibular glands significantly decreased after external irradiation. This observation was to be expected owing to the different irradiation doses administered by the two techniques. The scintigraphic technique used to evaluate salivary gland function should be used in future intensity-modulated radiation therapy salivary-gland-sparing studies in order to evaluate both the acute and chronic effects of irradiation in head and neck cancer patients.

Maintained salivary function after brachytherapy in patients with head and neck carcinomas--evaluation using quantitative salivary gland scintigraphy.

The purpose of this study was to determine whether or not salivary gland dysfunction occurs within the first three months after brachytherapy in patients with head and neck carcinoma. Of the 20 patients with head and neck squamous cell carcinoma included in this study, 11 were treated with brachytherapy and the remaining 9 patients received external irradiation. All the patients underwent a salivary gland scintigraphy before and after radiotherapy. The scintigraphic parameters of each major salivary gland were then compared before and after the radiotherapy. In the brachytherapy group, none of the scintigraphic functional parameters showed a significant change before and after the radiotherapy. In contrast, all of the parameters with the exception of the uptake ratio (UR) of the submandibular glands significantly decreased after external irradiation. This observation was to be expected owing to the different irradiation doses administered by the two techniques. The scintigraphic technique used to evaluate salivary gland function should be used in future intensity-modulated radiation therapy salivary-gland-sparing studies in order to evaluate both the acute and chronic effects of irradiation in head and neck cancer patients.