Unsupervised Anomaly Detection in Multivariate Spatio-Temporal Data Using Deep Learning: Early Detection of COVID-19 Outbreak in Italy

Abstract

Unsupervised anomaly detection for spatio-temporal data has extensive use in a wide variety of applications such as earth science, traffic monitoring, fraud and disease outbreak detection. Most realworld time series data have a spatial dimension as an additional context which is often expressed in terms of coordinates of the region of interest (such as latitude -longitude information). However, existing techniques are limited to handle spatial and temporal contextual attributes in an integrated and meaningful way considering both spatial and temporal dependency between observations. In this paper, a hybrid deep learning framework is proposed to solve the unsupervised anomaly detection problem in multivariate spatio-temporal data. The proposed framework works with unlabeled data and no prior knowledge about anomalies are assumed. As a case study, we use the public COVID-19 data provided by the Italian Department of Civil Protection. Northern Italy regions' COVID-19 data are used to train the framework; and then any abnormal trends or upswings in COVID-19 data of central and southern Italian regions are detected. The proposed framework detects early signals of the COVID-19 outbreak in test regions based on the reconstruction error. For performance comparison, we perform a detailed evaluation of 15 algorithms on the COVID-19 Italy dataset including the state-of-the-art deep learning architectures. Experimental results show that our framework shows significant improvement on unsupervised anomaly detection performance even in data scarce and high contamination ratio scenarios (where the ratio of anomalies in the data set is more than 5%). It achieves the earliest detection of COVID-19 outbreak and shows better performance on tracking the peaks of the COVID-19 pandemic in test regions. As the timeliness of detection is quite important in the fight against any outbreak, our framework provides useful insight to suppress the resurgence of local novel coronavirus outbreaks as early as possible.

The Role of Artificial Intelligence in Fighting the COVID-19 Pandemic

“…As a first attempt, AI can support video-surveillance systems to check the correct usage of masks or enhance public health monitoring through IoT data (Zhu et al 2020). Models able to identify irregular patterns and early signs of the outbreak of the pandemic, as in Karadayi et al (2020), are also developed to design strategies to prevent the collapse of healthcare facilities (Car et al 2020). Furthermore, simulation toolkits, providing estimates of the epidemiological parameters combined with components seeking the optimal trade-off policies between the decision makers' constraints and goals (Ghamizi et al 2020), evidence how AI models can be used to model problems connected with the spread and impact of infectious diseases based on geographical and time data as inputs.…”

Anomaly Detection in COVID-19 Time-Series Data

“…The existing anomaly detection techniques in COVID-19 data focus only on outbreak detection [5][6][7][8] in the COVID-19 tracking cases across the world. Karadayi et al [5] used a hybrid autoencoder network composed of a 3D convolutional neural network (CNN) and an autocorrelation based network for outbreak detection from spatio-temporal COVID-19 data provided by the Italian Department of Civil Protection. Jombart et al [6] used linear regression, generalised linear models (GLMs), and Bayesian regression to detect sudden changes in potential COVID-19 cases in England.…”

“…Existing anomaly detection techniques for COVID-19 data (some based on machine learning) focus only on outbreak detection [5][6][7][8] in the COVID-19 tracking cases across the world. Machine learning approaches [9][10][11] have also been used for data quality assurance in other domains.…”

“…This approach, called IDEAL, builds upon our previous work on data quality assessment approach [19] that uses an LSTM-autoencoder [10] network to find anomalies in unsupervised data. Anomalies are data records or subsequences of data records whose behaviors (i.e., attribute values or change in the values over time) are significantly different from the majority of records and subsequences in a timeseries data set [5]. IDEAL automatically (1) discovers different types of constraints from the sequence data, (2) marks subsequences and records that violate the constraints as suspicious, and (3) explains the violations.…”

Developing a Data-Driven Unsupervised Pattern Recognition Approach for Sensor Signal Anomaly Detection

“…Anomaly detection refers to the tasks of identifying the rare events or observations that are deviated from its normal frequent behaviour. Today, anomaly detection is broadly used in many research areas such as health monitoring [1], [2], [3], [4], [5], [6] for example heart disease diagnosis [1] and neuromuscular disorders diagnosis [5], environment monitoring such as sewer pipeline fault identification [7] and solar farms anomalies detection [8], and machine condition monitoring [9], [10] for example machinery fault diagnosis [11], [12], [13], [14], [9]. Depending on the anomaly detection problem, it is required to design algorithms which are able to identify anomalies in different types of data such as image [15], [2], [16], [17], video [7], sound signal [9] speech signal [18], sensor signal [19], [5], text [20], spatio-temporal data [4], streaming data [21] and time-series [22], [23].…”