Ultrasound Anomaly Detection Based on Variational Autoencoders

Milkovic, Fran; Filipović, Branimir; Subašić, Marko; Petković, Tomislav; Lončarić, Sven; Budimir, Marko

doi:10.1109/ispa52656.2021.9552041

Cited by 9 publications

(6 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The concept of anomaly detection using artificial intelligence has been approached with various methods from the research community because of the numerous challenges involved, such as the vast amount and diversity of data to be analyzed. In recent years, we observed that the community has been researching various autoencoder combinations to solve complex problems effectively [ 15 , 16 ]. Because of the depth and richness of information, the universality of applications, and the difficulty of monitoring processes, this research is becoming increasingly important.…”

Section: Literature Reviewmentioning

confidence: 99%

“…For evaluating ultrasonic testing (UT) data, Milkovic et al [ 15 ] suggested a variational autoencoder (VAE). In standard UT data, the VAE was applied to characterize the distributions.…”

Section: Literature Reviewmentioning

confidence: 99%

“…This is an innovative model of artificial vision, for optimal decision making, regarding the recognition of anomalies in the industrial environment. Specifically, we present a novel methodology using fuzzy entropy and Euclidean fuzzy similarity measurement for the first time in the literature, in order to maximize the possibility of using nonlinear transformation through deterministic functions, thus creating an entirely realistic and highly reliable machine vision detection system for Industry 4.0 based on variational fuzzy autoencoder [ 15 , 20 ].…”

Section: Proposed Machine Vision-based Anomaly Detection Systemmentioning

confidence: 99%

See 2 more Smart Citations

A Machine Vision Anomaly Detection System to Industry 4.0 Based on Variational Fuzzy Autoencoder

Jiang

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

From a technological point of view, Industry 4.0 evolves and operates in a smart environment in which the real and virtual worlds come together through smart cyber-physical systems. These devices that control each other autonomously activate innovative functions that enhance the production process. However, the industrial environment in which the most modern digital automation and information technologies are integrated is an ideal target for large-scale targeted cyberattacks. Implementing an integrated and effective security strategy in the Industrial 4.0 ecosystem presupposes a vertical inspection process at regular intervals to address any new threats and vulnerabilities throughout the production line. This view should be accompanied by the deep conviction of all stakeholders that all systems of modern industrial infrastructure are a potential target of cyberattacks and that the slightest rearrangement of mechatronic systems can lead to generalized losses. Accordingly, given that there is no panacea in designing a security strategy that fully ensures the infrastructure in question, advanced high-level solutions should be adopted, effectively implementing security perimeters without direct dependence on human resources. One of the most important methods of active cybersecurity in Industry 4.0 is the detection of anomalies, i.e., the identification of objects, observations, events, or behaviors that do not conform to the expected pattern of a process. The theme of this work is the identification of defects in the production line resulting from cyberattacks with advanced machine vision methods. An original variational fuzzy autoencoder (VFA) methodology is proposed. Using fuzzy entropy and Euclidean fuzzy similarity measurement maximizes the possibility of using nonlinear transformation through deterministic functions, thus creating an entirely realistic vision system. The final finding is that the proposed system can evaluate and categorize anomalies in a highly complex environment with significant accuracy.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

“…For evaluating ultrasonic testing (UT) data, Milkovic et al [ 15 ] suggested a variational autoencoder (VAE). In standard UT data, the VAE was applied to characterize the distributions.…”

Section: Literature Reviewmentioning

confidence: 99%

Section: Proposed Machine Vision-based Anomaly Detection Systemmentioning

confidence: 99%

See 1 more Smart Citation

A Machine Vision Anomaly Detection System to Industry 4.0 Based on Variational Fuzzy Autoencoder

Jiang

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…Therefore, we decided to opt for an anomaly detection task that requires fewer data for training. Interestingly, Variational Autoencoders have been successfully employed for anomaly/change/novelty detection tasks, such as bird species [19], ultrasounds [16], time series [14], computer networks [11], etc.…”

Section: # Of Observationsmentioning

confidence: 99%

Variational Autoencoders for Anomaly Detection in Respiratory Sounds

Cozzatti¹,

Simonetta²,

Ntalampiras³

2022

Preprint

View full text Add to dashboard Cite

This paper proposes a weakly-supervised machine learningbased approach aiming at a tool to alert patients about possible respiratory diseases. Various types of pathologies may affect the respiratory system, potentially leading to severe diseases and, in certain cases, death. In general, effective prevention practices are considered as major actors towards the improvement of the patient's health condition. The proposed method strives to realize an easily accessible tool for the automatic diagnosis of respiratory diseases. Specifically, the method leverages Variational Autoencoder architectures permitting the usage of training pipelines of limited complexity and relatively small-sized datasets. Importantly, it offers an accuracy of 57%, which is in line with the existing strongly-supervised approaches.

show abstract

“…In [6] authors developed a novel convolutional neural network that outperforms all current models on ultrasonic defect detection. In [7,8] authors managed to develop a deep learning approach that utilizes only non-anomalous data for training the defect, or anomaly, detector network. In [8] they show that the proposed network successfully detects all defects and outperforms standard classification approaches when a small number of anomalous data is present.…”

mentioning

confidence: 99%

Improvement Possibilities for Nuclear Power Plants Inspections by Adding Deep Learning-based Assistance Algorithms Into a Classic Ultrasound NDE Acquisition and Analysis Software

Pavlović¹,

Budimir²,

Milkovic

et al. 2023

joe

View full text Add to dashboard Cite

The safety of nuclear power plants has always been one of the most important security issues in the industry in general. Numerous standards, techniques, and tools have been developed to deal specifically with the safety of nuclear power plants – one has specialised probes, robotized systems, electronics, and software. Although seen as a mature (or slowly evolving) industry, this notion about nuclear safety is a bit misleading – the area is developing in many promising new directions. Some recent global events will speed up this development even more. On the other hand, the industry is currently going through digital transformation, which brings networking of devices, equipment, computers, and humans. This fourth industrial revolution promises speed, reliability, and efficiencies not possible up until now. In the NDE sector, new production techniques and traditional manufacturing lines are getting to be lights-out operations (near-total automation). The same is most probably going to happen with the safety inspections and quality insurance. Robotics and automation are improving worker safety and reducing human error. The well-being of inspectors working in a hazardous environment is being taken care of. Most experts agree that the digitalization of NDE offers unprecedented opportunities to the world of inspection for infrastructure safety, inspector well-being, and even product design improvements. While the community tends to agree on the value proposition of digital transformation of NDE, it also recognizes the challenges associated with such a major shift in a well-established and regulated sector. The work presented in this paper shows a part of the project that aims to develop a modular ultrasound diagnostic NDE system (consisting of exchangeable transducers, electronics, and acquisition/analysis software algorithms), for applications in hazardous environments within nuclear power plants. The paper will show how the software part of this system can reach near-total automation by implementing various deep learning algorithms as its features and, then, testing those algorithms on laboratory samples, showing encouraging results and promises of online monitoring applications. Furthermore, future general prospects of this technology are discussed, and how this technology can affect the well-being of nuclear power plant inspectors and contribute to overall plant safety.

show abstract

Ultrasound Anomaly Detection Based on Variational Autoencoders

Cited by 9 publications

References 10 publications

A Machine Vision Anomaly Detection System to Industry 4.0 Based on Variational Fuzzy Autoencoder

A Machine Vision Anomaly Detection System to Industry 4.0 Based on Variational Fuzzy Autoencoder

Variational Autoencoders for Anomaly Detection in Respiratory Sounds

Improvement Possibilities for Nuclear Power Plants Inspections by Adding Deep Learning-based Assistance Algorithms Into a Classic Ultrasound NDE Acquisition and Analysis Software

Contact Info

Product

Resources

About