Application of Artificial Intelligence and Machine Learning Methods to NDT Signal and Image Analysis

Dear Colleagues,

Non-destructive testing (NDT) plays a critical role in ensuring the structural integrity, safety, and longevity of materials and components across diverse sectors, including aerospace, civil infrastructure, energy production, manufacturing, and cultural heritage preservation. Conventional NDT techniques—such as ultrasonic testing, eddy current inspection, acoustic emission monitoring, radiography, infrared thermography, and visual inspection—generate rich datasets in the form of signals and images. However, the increasing complexity of modern materials, the demand for real-time monitoring, and the sheer volume of data produced in industrial settings have exposed limitations in traditional manual analysis, which is often subjective, labor-intensive, and prone to variability.

The integration of Artificial Intelligence (AI) and Machine Learning (ML) has emerged as a transformative paradigm in NDT. Techniques such as deep convolutional neural networks (CNNs) for image segmentation and defect detection, recurrent and transformer-based models for sequential signal analysis, hybrid multimodal fusion frameworks, and explainable AI (XAI) approaches are enabling automated, accurate, and scalable interpretation of NDT data. These advancements draw from broader progress in computer vision, signal processing, and data science, while addressing domain-specific challenges like noisy environments, limited labeled datasets, and the need for physical interpretability.

This research area is of paramount importance because it directly contributes to Industry 4.0/5.0 goals: predictive maintenance, reduced downtime, enhanced safety, and sustainability. AI-enhanced NDT can significantly improve defect detection rates, reduce false positives/negatives, accelerate inspections, and support digital twins and smart sensor networks. As global infrastructure ages and regulatory standards tighten, robust AI/ML solutions for NDT signal and image analysis are essential for preventing failures, optimizing resource use, and advancing non-invasive diagnostics in complex real-world scenarios.

The aim of this Special Issue is to collect and disseminate cutting-edge research on the development, application, and evaluation of AI and ML methods specifically tailored to the analysis of signals and images produced by NDT technologies. We seek contributions that bridge the gap between advanced computational intelligence and practical NDT challenges, fostering innovation in data processing, modeling, fusion, and interpretation.

This topic aligns closely with the scope of NDT—Journal of Non-Destructive Testing, an international, open-access, peer-reviewed journal dedicated to NDT science, technology, and applications. The journal emphasizes the collection, processing, modeling, fusion, and interpretation of multisource, multiscale, and multitemporal data to enhance standalone and combined NDT methods. It also promotes the design and implementation of state-of-the-art technological solutions, including ICT for data management and visualization, as well as contributions to standards and best practices.

By focusing on AI/ML for signal and image analysis, this Special Issue directly supports the journal’s goals of advancing inclusive, interdisciplinary research and strengthening conventional and emerging NDT capabilities in new and complex environments.

We welcome high-quality submissions addressing the following (non-exhaustive) themes:

• Novel deep learning architectures and hybrid models for NDT image analysis (e.g., defect segmentation, classification, and reconstruction in radiography, thermography, or shearography).
• Machine learning techniques for advanced signal processing and feature extraction in ultrasonic, acoustic emission, eddy current, vibration, or electromagnetic NDT data.
• Multimodal data fusion combining signals and images for improved diagnostic reliability and uncertainty quantification.
• Explainable AI, physics-informed ML, and interpretable models to enhance trust and integration with domain knowledge.
• Transfer learning, few-shot/zero-shot learning, active learning, and domain adaptation strategies for data-scarce or varying industrial NDT conditions.
• AI-driven real-time monitoring, predictive analytics, and digital twin integration in NDT systems.
• Performance benchmarking, standardization, regulatory aspects, and case studies in key industries (aerospace, energy, civil engineering, etc.).
• Ethical considerations, robustness to adversarial conditions, and sustainability impacts of AI in NDT.

Accepted Article Types: Research articles, review papers, technical notes, and short communications. All submissions must present original, unpublished work with sufficient methodological detail for reproducibility, in line with the journal’s standards. No formal length restrictions apply, provided manuscripts are concise, well-structured, and scientifically sound.

Dr. Giacomo Veneri
Dr. Remo Ribichini
Dr. Hossein Taheri
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 250 words) can be sent to the Editorial Office for assessment.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-anonymized peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. NDT is an international peer-reviewed open access quarterly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1000 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• non-destructive testing (NDT)
• artificial intelligence
• machine learning
• deep learning
• signal processing
• image analysis
• defect detection
• multimodal data fusion
• explainable AI (XAI)
• nondestructive evaluation (NDE)