TOKYO, Jul 16, 2021 – (JCN Newswire) – Fujitsu Limited and Inria, the French national research institute for digital sciences and technologies, today announced the development of a new AI technology capable of identifying factors contributing to anomalies in time series data.
In recent years, various types of time series data collected from fields such as healthcare, social infrastructure, and manufacturing have been exploited by AI to perform situational judgment and detect anomalies. In the case of time series data, however, there is a wide range of factors that can contribute to decision making in AI. This means that even experts find it difficult to notice what kind of change in the data contributed to an anomaly detection, making it difficult to take appropriate action to prevent their occurrence.
Fujitsu and Inria, more precisely Inria’s DATASHAPE project-team led by Frédéric Chazal in France, have now successfully developed a new technology based on topological data analysis (TDA) (1) which makes it possible to identify factors contributing to AI anomaly detections for given time series and visualizing differences in AI decisions under normal and abnormal circumstances.
Fujitsu and Inria predict that this will contribute to the analysis of the causes of anomalies in time series data for various phenomena, clarifying the mechanism surrounding the occurrence of anomalies, as well as the discovery of new solutions to them.
This technology will be featured as one of only 3% of total papers submitted as a Long Talk presentation at the 38th International Conference on Machine Learning (ICML), the leading international conference in the field of machine learning. machine learning, which opens virtually from July 18, 2021.
Newly developed technology
Fujitsu and Inria have developed AI technology capable of determining the cause of anomalies in time series data, consisting of the following key functionalities.
1) Using analysis technology developed by Fujitsu that extracts features that affect judgment from time series data and detects anomalies (2), features that led to the abnormal judgment as well as features not Linked data that has been deemed abnormal by the AI is mapped to a plane (TDA space).
2) The technology transforms the point data of the feature that is the cause closer to the point data group of the feature that is not the cause on the plane.
3) The time series data is distorted based on the conversion of the characteristics of the point data, and the data deemed normal is generated.
This allows the waveform of normal and abnormal time series data to be compared and allows the user to visually investigate the cause of the anomaly.
The newly developed technology was applied to test the possibility of detecting symptoms of delirium (3) using actual electroencephalographic (EEG) (4) data collected in strict accordance with ethical guidelines. Using the newly developed technology, it was confirmed that the brain wave characteristics of the time series data coincide with the phenomenon of ‘slowing down’ (5) that sometimes accompanies the state of delirium. These results offer the possibility of helping healthcare professionals interpret the data to help determine the cause of these symptoms. This could one day contribute to important medical developments, including the ability to discover possible precursors of diseases that are difficult to identify with conventional techniques, as well as the discovery of preventive treatments. The technology could also be applied to shed light on disease mechanisms that are not yet well understood.
Comments from Dr Gen Shinozaki, ASSOCIATE PROFESSOR OF PSYCHIATRICS AND BEHAVIORAL SCIENCES, Stanford University School of Medicine
Due to the nature of random signals, it has proven difficult to use EEG data quantitatively and accurately to identify certain disorders. In recent years, advances in data processing technologies, such as AI, have provided insight into the characteristic changes in subtle brain waves. These advances are important not only for diagnosing various disorders, but also for understanding the response to treatment and the pathophysiological mechanism. Technology developed by Fujitsu and Inria has successfully captured the unique characteristics of brain waves in patients with delirium. In addition to verifying this, we anticipate that further improvements and practical uses of this technology will eventually offer the potential to achieve accurate diagnosis, monitoring of response to treatment, and elucidating the pathophysiology of other disorders.
Fujitsu and Inria plan to encourage the use of the jointly developed technology in field work and experiments in companies and research institutes, and are verifying the technology.
(1) Topological Data Analysis (TDA):
A method of data analysis is provided in which the data is arranged at a group of points in space and geometric data is extracted from the group.
(2) Analysis technology developed by Fujitsu that categorizes time series data by characteristics and detects anomalies:
Fujitsu and French company Inria jointly develop technology to automatically create AI models for anomaly detection (Press release: 2020/3/16)
a syndrome, or a group of symptoms, caused by a disturbance in the normal functioning of the brain.
(4) real EEG data:
the newly developed technology was applied to electroencephalographic data from approximately 600 patients, who consented to participate in research with Dr. Shinozaki, at the University of Iowa. Professor Shinozaki has been an Associate Professor at Stanford University since June 2021.
(5) slowdown phenomenon:
a phenomenon that occurs frequently in the EEG data of patients with delirium.
Fujitsu is the Japanese leader in information and communication technologies (ICT) offering a full range of technology products, solutions and services. Approximately 126,000 Fujitsu people support customers in more than 100 countries. We use our experience and the power of ICT to shape the future of society with our customers. Fujitsu Limited (TSE: 6702) reported consolidated sales of 3.6 trillion yen (US $ 34 billion) for the fiscal year ended March 31, 2021. For more information, please see www.fujitsu.com.
Inria is the French national research institute for digital sciences and technologies. World-class research, technological innovation and entrepreneurial risk are its DNA. Within 200 project-teams, most of which are shared with major research universities, more than 3,500 researchers and engineers are exploring new avenues, often in an interdisciplinary manner and in collaboration with industrial partners, to meet ambitious challenges.
As a technological institute, Inria supports the diversity of avenues of innovation: from publishing open source software to the creation of technological start-ups (Deeptech).
Copyright 2021 JCN Newswire. All rights reserved. www.jcnnewswire.com