Dalarna University's logo and link to the university's website

BACKGROUND: Dementia is a general term for several progressive neurodegenerative disorders including Alzheimer's disease. Timely and accurate detection is crucial for early intervention. Advancements in artificial intelligence present significant potential for using machine learning to aid in early detection.

OBJECTIVE: Summarize the state-of-the-art machine learning-based approaches for dementia prediction, focusing on non-invasive methods, as the burden on the patients is lower. Specifically, the analysis of gait and speech performance can offer insights into cognitive health through clinically cost-effective screening methods.

METHODS: A systematic literature review was conducted following the PRISMA protocol (Preferred Reporting Items for Systematic Reviews and Meta-Analyses). The search was performed on three electronic databases (Scopus, Web of Science, and PubMed) to identify the relevant studies published between 2017 to 2022. A total of 40 papers were selected for review.

RESULTS: The most common machine learning methods employed were support vector machine followed by deep learning. Studies suggested the use of multimodal approaches as they can provide comprehensive and better prediction performance. Deep learning application in gait studies is still in the early stages as few studies have applied it. Moreover, including features of whole body movement contribute to better classification accuracy. Regarding speech studies, the combination of different parameters (acoustic, linguistic, cognitive testing) produced better results.

CONCLUSIONS: The review highlights the potential of machine learning, particularly non-invasive approaches, in the early prediction of dementia. The comparable prediction accuracies of manual and automatic speech analysis indicate an imminent fully automated approach for dementia detection.

Parkinson's disease (PD) is a neurodegenerative disorder that affects millions of people worldwide, causing various motor and non-motor symptoms. Early diagnosis of PD is crucial for timely intervention and management. Gait analysis provides insights into the motor impairments with PD, aiding in early detection. In this study, different deep learning models such as CNN, LSTM, and CNN-LSTM with varying neural network depths were explored to classify PD using gait data acquired through sensor technology. The study then compared the results of deep learning models with machine learning algorithms (Random Forest (RF) and Decision Trees (DT)). The dataset used in this study consists of 93 persons with PD and 73 healthy controls (HC) collected through sensor technology. The findings reveal that the RF algorithm achieved the highest accuracy of 96%, followed by the CNN-LSTM model of 95.49 %. © 2024 IEEE.

This study addresses the critical safety issue of declining retroreflectivity values of road traffic signs, which can lead to unsafe driving conditions, especially at night. The paper aims to predict the retroreflectivity coefficient values of these signs and to classify their status as acceptable or rejected (in need of replacement) using machine learning models. Moreover, logistic regression and survival analysis are used to predict the median lifespans of road traffic signs across various geographical locations, focusing on signs in Croatia and Sweden as case studies. The results indicate high accuracy in the predictive models, with classification accuracy at 94% and an R2 value of 94% for regression analysis. A significant finding is that a considerable number of signs maintain acceptable retroreflectivity levels within their warranty period, suggesting the feasibility of extending maintenance checks and warranty periods to 15 years which is longer than the current standard of 10 years. Additionally, the study reveals notable variations in the median lifespans of signs based on color and location. Blue signs in Croatia and Sweden exhibit the longest median lifespans (28 to 35 years), whereas white signs in Sweden and red signs in Croatia show the shortest (16 and 10 years, respectively). The high accuracy of logistic regression models (72–90%) for lifespan prediction confirms the effectiveness of this approach. These findings provide valuable insights for road authorities regarding the maintenance and management of road traffic signs, enhancing road safety standards. © 2024 Tongji University and Tongji University Press

The color of road signs is a critical aspect of road safety, as it helps drivers quickly and accurately identify and respond to these signs. Properly colored road signs improve visibility during the day and make it easier for drivers to make informed decisions while driving. In order to ensure the safety and efficiency of road traffic, it is essential to maintain the appropriate color level of road signs. The objective of this study was to analyze the color status and daylight chromaticity of in-use road signs using supervised machine learning models, and to explore the correlation between road sign's age and daylight chromaticity. Three algorithms were employed: Random Forest (RF), Support Vector Machine (SVM), and Artificial Neural Network (ANN). The data used in this study was collected from road signs that were in-use on roads in Sweden. The study employed classification models to assess the color status (accepted or rejected) of the road signs based on minimum acceptable color levels according to standards, and regression models to predict the daylight chromaticity values. The correlation between road sign's age and daylight chromaticity was explored through regression analysis. Daylight chromaticity describes the color quality of road signs in daylight, that is expressed in terms of X and Y chromaticity coordinates. The study revealed a linear relationship between the road sign's age and daylight chromaticity for blue, green, red, and white sheeting, but not for yellow. The lifespan of red signs was estimated to be around 12 years, much shorter than the estimated lifespans of yellow, green, blue, and white sheeting, which are 35, 42, 45, and 75 years, respectively. The supervised machine learning models successfully assessed the color status of the road signs and predicted the daylight chromaticity values using the three algorithms. The results of this study showed that the ANN classification and ANN regression models achieved high accuracy of 81% and R2 of 97%, respectively. The RF and SVM models also performed well, with accuracy values of 74% and 79% and R2 ranging from 59% to 92%. The findings demonstrate the potential of machine learning to effectively predict the status and daylight chromaticity of road signs and their impact on road safety in the Swedish context. © 2023 International Association of Traffic and Safety Sciences

AbstractRoad signs play a vital role in providing drivers with crucial information for safe driving in both day and nighttime. The color of road signs enhances visibility during daylight hours, while retroreflectivity is essential for improving visibility during nighttime conditions. Road authorities, responsible for maintaining road signs, primarily consider the levels of retroreflectivity when deciding to replace them, ensuring optimal visibility for drivers. This study focuses on examining the degradation of road signs based on retroreflectivity and color to ensure safe driving through adequate visibility in both day and nighttime conditions. The study underscores the significance of regulating the deterioration of road sign colors to enhance visibility and legibility, while minimizing maintenance and replacement costs. The primary objective of this paper is to predict the age (service life) of road signs by considering both retroreflectivity and color status and using logistic regression. The results indicate that the age of road signs can be influenced by either retroreflectivity or color. For instance, approximately 50% of red road signs are projected to lose their color after 16 years, while their retroreflectivity remains acceptable. Similarly, around 50% of yellow and white road signs experience retroreflectivity degradation after 20 and 16 years, respectively, while their color remains acceptable. Finally, blue road signs demonstrate acceptable retroreflectivity and color levels even after 35 years.

The road traffic signs in Sweden have no inventory system and it is unknown when a sign has reached the end of its service life and needs to be replaced. As a result, the road authorities do not have a systematic maintenance program for road traffic signs, and many signs which are not in compliance with the minimum retroreflectivity performance requirements are still found on the roads. Therefore, it is very important to find an inexpensive, safe, easy, and highly accurate method to judge the retroreflectivity performance of road signs. This will enable maintenance staff to determine the retroreflectivity of road signs without requiring measuring instruments for retroreflectivity or colors performance. As a first step toward the above goal, this paper aims to identify factors affecting the retroreflectivity of road signs. Two different datasets were used, namely, the VTI dataset from Sweden and NMF dataset from Denmark. After testing different models, two logarithmic regression models were found to be the best-fitting models, with R2 values of 0.50 and 0.95 for the VTI and NMF datasets, respectively. The first model identified the age, direction, GPS positions, color, and class of road signs as significant predictors, while the second model used age, color, and the class of road signs. © 2022 by the authors. Licensee MDPI, Basel, Switzerland.

Parking lots are places of high air pollution as numerous vehicles cruise to find vacant parking places. Open parking lots receive high vehicle traffic, and when limited empty spaces are available it leads to problems, such as congestion, pollution, and driver frustration. Due to lack of return on investment, open parking lots are little studied, and there is a research gap in understanding the magnitude of CO2 emissions and cruising observed at open parking lots. Thus, this paper aims to estimate CO2 emissions and cruising distances observed at an open parking lot. A thermal camera was utilized to collect videos during peak and non-peak hours. The resulting videos were utilized to collect cruising trajectories of drivers searching for empty parking spaces. These trajectories were analyzed to identify optimal and non-optimal cruising, time to park, and walking distances of drivers. A new CO2 model was proposed to estimate emissions in smaller geographical regions, such as open parking lots. The majority of drivers tend to choose parking spaces near a shopping center, and they prefer to cruise non-optimal distances to find an empty parking space near the shopping center. The observed mean non-optimal cruising distance is 2.7 times higher than the mean optimal cruising distance. Excess CO2 emissions and non-optimal cruising were mainly observed during visitor peak hours when there were limited or no empty parking spaces. During visitor peak hours, several vehicles could not find an empty parking space in the region of interest, which also leads to excess cruising.

There is no data collected and saved about road signs in Sweden and the status for these signs is unknown. Furthermore, the status of the sign colors, the quality of the sign, the type of the retroreflection material, and age of the road signs are unknown. Therefore, the status of a road sign (approved or not), which depends on these parameters, is unknown.The aim of this study is to predict the status of the road signs mounted on the Swedish roads by using supervised machine learning. This study investigates the effect of using principal component analysis (PCA) and data scaling on the accuracy of the prediction. The data were prepared before using then scaled using two methods which are the normalization and the standardization.The three algorithms that tested in this study are Random Forest, Artificial Neural Network (ANN), and Support Vector Machines (SVM). They are invoked to predict the status of the road signs. The algorithms exhibited overall high predicting accuracy (98%), high precision (98%), high recall (98%), and high F1 scores (98%).Random forest showed the best performance with 4 PC components on the normalized data with a highest accuracy of 98%.Using PCA showed different impacts on the performance of different techniques. In the case of ANN, invoking PCA improves the accuracy, while for SVM the accuracy decreases when PCA is used. On other hand, PCA has no effect on the accuracy of the random forest model when scaling is invoked.The effect of the data scaling using normalization and standardization is also investigated in this study, and it is noticed that scaling of the data increases the accuracy of the prediction for all the three models (ANN, SVM and Random Forest). Furthermore, better accuracy is achieved when the standardization is invoked compared with normalization.

Road traffic signs define a visual language that can be interpreted by drivers.They represent the current traffic situation on the road, show danger and difficulties aroundthe drivers, give them warnings, and help them with their navigation by providing usefulinformation that makes driving safe and convenient. The main part of the road traffic sign isthe retroreflective material which reflects the light from the vehicle headlights to the driver.Driving during night-time is a challenge, and the rertoreflective material on the sign boardhelps the drivers to perceive and interpret the information on the road traffic sign properly.The aim of this paper is to study the factors affecting the performance of driving during nighttimeand the role the retroreflective material that plays in this regard. The vehicle headlights,ambient conditions, and the type of retroreflection material affect the light reflected from theroad traffic signs. It is also found that the retroreflectivity depends on vehicle factors such asheadlights colour and angle of illumination. Other factors such as environmental factors andsign factors can also affect the retroreflectivity.