Abstract Patient representation learning based on electronic health records (EHR) is a critical task for disease prediction. This task aims to effectively extract useful information on dynamic features. Although various existing works have achieved remarkable progress, the model performance can be further improved by fully extracting the trends, variations, and the correlation between the trends and variations in dynamic features. In addition, sparse visit records limit the performance of deep learning models. To address these issues, we propose the multi-perspective patient representation Extractor (MPRE) for disease prediction. Specifically, we propose frequency transformation module (FTM) to extract the trend and variation information of dynamic features in the time–frequency domain, which can enhance the feature representation. In the 2D multi-extraction network (2D MEN), we form the 2D temporal tensor based on trend and variation. Then, the correlations between trend and variation are captured by the proposed dilated operation. Moreover, we propose the first-order difference attention mechanism (FODAM) to calculate the contributions of differences in adjacent variations to the disease diagnosis adaptively. To evaluate the performance of MPRE and baseline methods, we conduct extensive experiments on two real-world public datasets. The experiment results show that MPRE outperforms state-of-the-art baseline methods in terms of AUROC and AUPRC.

Monitoring electric vehicles' battery status and forecasting their state of health is still an open challenge. To determine how and why a battery degrades over time, we have extensively monitored a Nissan Leaf's battery pack for more than one year. Collecting more than 4.5 million samples via a custom monitoring connected device to investigate how different driving behaviors affect battery aging. In addition, the best driving behaviors based on the battery's optimal temperature are revealed, including speed, acceleration and brake pedal pressure, and horsepower.

Patient representation learning based on electronic health records (EHR) is a critical task for disease prediction. This task aims to effectively extract useful information on dynamic features. Although various existing works have achieved remarkable progress, the model performance can be further improved by fully extracting the trends, variations, and the correlation between the trends and variations in dynamic features. In addition, sparse visit records limit the performance of deep learning models. To address these issues, we propose the Multi-perspective Patient Representation Extractor (MPRE) for disease prediction. Specifically, we propose Frequency Transformation Module (FTM) to extract the trend and variation information of dynamic features in the time-frequency domain, which can enhance the feature representation. In the 2D Multi-Extraction Network (2D MEN), we form the 2D temporal tensor based on trend and variation. Then, the correlations between trend and variation are captured by the proposed dilated operation. Moreover, we propose the First-Order Difference Attention Mechanism (FODAM) to calculate the contributions of differences in adjacent variations to the disease diagnosis adaptively. To evaluate the performance of MPRE and baseline methods, we conduct extensive experiments on two real-world public datasets. The experiment results show that MPRE outperforms state-of-the-art baseline methods in terms of AUROC and AUPRC.

Type 2 diabetes is the most common chronic disease for the elderly people. This disease is difficult to be cured and causes continued medical expenses. The early and personalized risk assessment of type 2 diabetes is necessary. So far, various type 2 diabetes risk prediction methods have been proposed. However, these methods have three major issues: 1) not fully considering the importance of personal information and rating information of healthcare system, 2) not adopting the long-term temporal information, and 3) not comprehensively capturing the correlation between the diabetes risk factor categories. To address these issues, the personalized risk assessment framework for elderly people with type 2 diabetes is needed. However, it is very challenging due to two reasons, namely imbalanced label distribution and high-dimensional features. In this paper, we propose diabetes mellitus network framework (DMNet) for type 2 diabetes risk assessment of elderly people. Specifically, we propose tandem long short-term memory to extract the long-term temporal information of different diabetes risk categories. In addition, the tandem mechanism is used to capture the correlation between the diabetes risk factor categories. To balance the label distribution, we adopt the method of synthetic minority over-sampling technique with Tomek links. To form the better feature representations, we utilize entity embedding to solve the problem of high-dimensional features. To evaluate the performance of our proposed method, we conduct the experiments on a real-world dataset called Research on Early Life and Aging Trends and Effects. The experiment results show that DMNet outperforms the baseline methods in terms of six evaluation metrics (i.e., accuracy of 0.94, balanced accuracy of 0.94, precision of 0.95, F1-score of 0.95, recall of 0.95 and AUC of 0.94).

Electric vehicles (EV) feature detailed monitoring and control over the CAN bus. Some of this data is made available to users on the On-Board Diagnostic version II (OBDII) bus thus providing an opportunity for large scale high-frequency data collection. This paper introduces a connected monitoring system for OBDII equipped vehicles. The system comprises a low cost hardware design and monitoring algorithms designed to optimize the number of variables collected and their collection frequency. The algorithm aims at collecting a high quantity of Battery Management System (BMS) data in electric vehicles together with power-usage data to enable short and long term estimation for battery state of health (SOH) and state of charge (SOC). The proposed system has been implemented and tested on a Nissan Leaf and lead to the acquisition of 1.7 million records over 120 hours of driving.

Autonomous vehicles have received great attention in the last years, promising to impact a market worth billions. Nevertheless, the dream of fully autonomous cars has been delayed with current self-driving systems relying on complex processes coupled with supervised learning techniques. The deep reinforcement learning approach gives us newer possibilities to solve complex control tasks like the ones required by autonomous vehicles. It let the agent learn by interacting with the environment and from its mistakes. Unfortunately, RL is mainly applied in simulated environments, and transferring learning from simulations to the real world is a hard problem. In this paper, we use LIDAR data as input of a Deep Q-Network on a realistic 1/10 scale car prototype capable of performing training in real-time. The robot-driver learns how to run in race tracks by exploiting the experience gained through a mechanism of rewards that allow the agent to learn without human supervision. We provide a comparison of neural networks to find the best one for LIDAR data processing, two approaches to address the sim2real problem, and a detail of the performances of DQN in time-lap tasks for racing robots.

This paper revisits the opportunities of using WiFi offloading for Vehicle to Internet (V2I) communication, and how this has changed over the last decade. With the rollouts of provider-managed WiFi networks that are more structured and operate under authenticated regimes, WiFi offloading, or use of available (roadside) WiFi networks for V2I data communication, has different opportunities and challenges. To study the current landscape,we develop a system (X-Fi), which efficiently selects, associates to, authenticates with, and performs WiFi offloading for V2I communication with these networks, and a tool (X-Perf), which illustrates opportunities of WiFi offloading available today in these networks, with measurements and experiments across four metro areas across three continents over 22 months. Our results indicate the feasibility of achieving 1 GB/hour application goodput, an order of magnitude higher than the number provided by open WiFi networks in the past, which can take a significant load away from alternative communication paths for V2I systems. Moreover, we provide several implications on transport protocols and WiFi deployments to shed light on the use of such WiFi networks for V2I communication.

COVID-19 is a highly contagious virus. Blood test is one of effective methods for COVID-19 diagnosis. However, the issues of blood test are time-consuming and lack of medical staff. In this paper, four deep learning hybrid models are proposed to address these issues (i.e., CNN+GRU, CNN+Bi-RNN, CNN+Bi-LSTM, CNN+Bi-GRU). In addition, two best models, CNN and CNN+LSTM, from Turabieh et al. and Alakus et al., are implemented, respectively. Blood test data from Hospital Israelita Albert Einstein is used to train and test six models. The proposed best model, CNN+Bi-GRU, is accuracy of 0.9415, precision of 0.9417, recall of 0.9417, F1-score of 0.9417, AUC of 0.91, which outperforms the best models from Turabieh et al. and Alakus et al. Furthermore, the proposed model can help patients to get blood test results faster than traditional manual tests without errors caused by fatigue. The authors can envisage a wide deployment of proposed model in hospitals to alleviate the testing pressure from medical workers, especially in developing and underdeveloped countries.

Abstract Credit risk assessment is at the core of modern economies. Traditionally, it is measured by statistical methods and manual auditing. Recent advances in financial artificial intelligence stemmed from a new wave of machine learning (ML)-driven credit risk models that gained tremendous attention from both industry and academia. In this paper, we systematically review a series of major research contributions (76 papers) over the past eight years using statistical, machine learning and deep learning techniques to address the problems of credit risk. Specifically, we propose a novel classification methodology for ML-driven credit risk algorithms and their performance ranking using public datasets. We further discuss the challenges including data imbalance, dataset inconsistency, model transparency, and inadequate utilization of deep learning models. The results of our review show that: 1) most deep learning models outperform classic machine learning and statistical algorithms in credit risk estimation, and 2) ensemble methods provide higher accuracy compared with single models. Finally, we present summary tables in terms of datasets and proposed models.

COVID-19 pandemic has led to serious economic and life losses. Face Masks serve as first infection barrier when used in public spaces. In this paper, we propose a new near-realtime method to automatically recognize face mask wearing that combines human posture recognition with convolutional neural network (CNN). We use the power of human posture recognition to perform background filtering and spatial reduction in the original images. The outcome is then used by a trained CNN model to identify if the subject is wearing a mask. We exploit Openpose to identify the skeleton of human body and locate the facial region thus spatially reducing the area to be processed by the CNN framework. We then adopt supervised learning approach to detect if a face mask is present. The CNN is trained using images, cropped to the supposed face mask covered region. This approach led to a substantial reduction in neural network complexity yet improving the recognition accuracy. The system has been evaluated in a multitude of scenarios using images taken in public places at different time of day and with different angles. Overall, our system achieves a recognition accuracy of 95.8% and 94.6% in daytime and nighttime respectively.

Humans generally spend more than 90% of their time in indoor environments. Such value can reach 100% due to the restrictions and limitations we are experiencing because of the current COVID-19 pandemic. Indeed, monitoring the indoor air quality (IAQ) becomes strategic to prevent and limit risks and adverse effects on building occupants’ health, comfort, and well-being. To reduce one of the variables impacting the IAQ, i.e. CO2, indoor plants can be exploited. In this paper, we present the low-cost prototype of a system we designed with the intent to foster user’s awareness about IAQ exploiting an IoT-enabled home gardening system, able to sense information about the ambient conditions and plant health.

Autonomous cars have been in the making for over 15 years. Skepticism has taken the place of initial hype and enthusiasm. Current autonomous driving systems give no guarantee of 100% correctness and reliability, and users are not willing to take a chance on a car that is unable to cope with all the possible driving scenarios. Robotic drivers are expected to be perfect. Major players such as Tesla and Waymo rely on highly detailed maps and very large sensor data in a race to build the ultimate robotic driver to cope with all possible driving scenarios. This approach optimizes for safety but delays the dream of fully autonomous cars. In this paper we consider robot-drivers as teen-drivers eager to learn how to drive but prone to mistakes in the beginning. The question we are trying to investigate is "what if we allow autonomous cars to make mistakes like young human drives do?" In this paper, we explore reinforcement learning for small size autonomous vehicles fusing information from several sensors including a camera, color sensors, and sonar sensors. The robot-drivers have initially no information about the driving scenarios they learn with experience through a mechanism of rewards designed to quickly help our robot-teen to learn its driving skills.

Mobility and ambient conditions are key factors in urban environments, affecting well-being and quality of life. In this context, sensors, smart mobility, networks, connectivity can play a significant and strategic role, being exploited with the aim of improving data and information available to public administration and to each citizen. In this way, they can be supported in having more sustainable and aware behaviours and in getting useful information and services, improving their daily activities. In this paper, we present a prototype of smart bike eco-system, designed with the aim of collecting, aggregating and sharing data about air pollution and about the urban environment, which can be exploited in a smart mobility context thanks to sensor and vehicular networks.

Environmental contaminations such as fine particulate matter (PM2.5) and ultraviolet (UV) are expanding public health concerns globally. Existing centralized monitoring stations, however, are unable to properly estimate human exposure due to the low resolution spatiotemporal data. Wireless Sensor Network (WSN) uses real-time capable instrumentation could not only provide fine-grained readings of multiple environmental factors, but also create real-time mapping of the contaminants. In this paper we present a WSN system which is capable of sensing multiple environmental factors, collecting data from multiple dispersed sensor nodes and displaying the aggregated data in real-time. Each individual sensor node is capable of probing multiple factors, including temperature, humidity, atmospheric pressure, PM2.5, UV radiation, and geographical location. Sensor data are transmitted to a server, which then be stored into a database through Wi-Fi networks. Each sensor node is portable enough to be carried for personal use, enabling broad potential application of our system.