Editorial Topical Collection - Explainable and Augmented Machine Learning for Biosignals and Biomedical Images

Aug 3, 2026

Upload by Violet

Editorial introduction to the Sensors 2023 topical collection on explainable and augmented machine learning for biosignals and biomedical images. Highlights the motivation for addressing the “black-box” opacity of machine learning in biomedical decision-making and the role of explainable AI in improving interpretability and clinician trust. Frames the collection within the growing availability of medical data from biosensors and IoT networks, and the use of augmented learning via techniques such as GANs, then previews ten peer-reviewed contributions across biosignal and medical imaging applications.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages4

Reading Time10 min

AccessFree to download

Introduction and motivation

Explainability to address ML opacity

Augmented learning with generative models

Topical collection scope

Overview of the accepted papers

sensors Editorial Editorial Topical Collection: “Explainable and Augmented Machine Learning for Biosignals and Biomedical Images” Cosimo Ieracitano 1, *, Mufti Mahmud 2,3,4, Maryam Doborjeh 5 and Aim² Lay-Ekuakille 6 Citation: Ieracitano, C.; Mahmud, M.; Doborjeh, M.; Lay-Ekuakille, A. Editorial Topical Collection: “Explainable and Augmented Machine Learning for Biosignals and Biomedical Images”. Sensors 2023, 23, 9722. [https://doi.org/10.3390/](https://doi.org/10.3390/)[ ](https://doi.org/10.3390/)s23249722 Received: 15 November 2023 Accepted: 28 November 2023 Published: 9 December 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ([https://](https://)[ ](https://)[creativecommons.org/licenses/by/](creativecommons.org/licenses/by/)[ ](creativecommons.org/licenses/by/)[4.0/](4.0/)) . 1 DICEAM Department, University Mediterranea of Reggio Calabria, Via Zehender, Feo di Vito, 89122 Reggio Calabria, Italy 2 Department of Computer Science, Nottingham Trent University, Nottingham NG11 8NS, UK; [mufti.mahmud@ntu.ac.uk](mufti.mahmud@ntu.ac.uk) 3 Computing and Informatics Research Centre, Nottingham Trent University, Nottingham NG11 8NS, UK 4 Medical Technologies Innovation Facility, Nottingham Trent University, Nottingham NG11 8NS, UK 5 Computer Science and Software Engineering, Auckland University of Technology, Auckland 1010, New Zealand; [maryam.gholami.doborjeh@aut.ac.nz](maryam.gholami.doborjeh@aut.ac.nz) 6 Department of Innovation Engineering, University of Salento, 73100 Lecce, Italy; [aime.lay.ekuakille@unisalento.it](aime.lay.ekuakille@unisalento.it) * Correspondence: cosimo.ieracitano@unirc.it Machine learning (ML) is a well-known subﬁeld of artiﬁcial intelligence (AI) that aims at developing algorithms and statistical models able to empower computer systems to automatically adapt to a speciﬁc task through experience or learning from data [1] . ML techniques have been demonstrating remarkable breakthroughs in the ﬁeld of biomedical research, especially in predictive analytics and classiﬁcation tasks [2,3] . However, the success of ML in this domain has been accompanied by a challenge, i.e., the inherent opaqueness of ML algorithms [4] . Indeed, despite their efﬁcacy, ML algorithms lack transparency in their decision-making processes and are often seen as black boxes. This lack of transparency raises concerns, especially in critical domains, where understanding the rationale behind machine decisions is important for fostering trust indecision-making [5] . In this regard, the emergence of explainable artiﬁcial intelligence (xAI) techniques has become a pivotal focus within this ﬁeld. In particular, xAI methods strive to unveil the internal mechanisms of the AI algorithms, aiming to shed light on the outcomes, predictions, decisions, and recommendations generated by such models [6,7] . The primary objective is to enhance the interpretability and transparency of machine decisions. This is of paramount importance in medical applications, where such enhanced comprehension could have asigniﬁcant impact on clinicians' ﬁnal decision-making [8] . In this context, several xAI-based approaches have been emerging in clinical applications, for example, rehabilitation systems based on brain–computer interfaces [9], detection of neurological disorders [10], breast cancers [11,12], and medical imaging analysis [13] . Furthermore, the escalating availability of medical and clinical data, collected from an expanding network of interconnected biosensors within the Internet of Things (IoT) framework, provides a rich source for training and reﬁning ML models [14,15] . In addition, recent advances in augmented techniques, i.e., generative adversarial networks (GANs), have enhanced the decision-making capabilities of ML algorithms. Indeed, generativ

FAQs

Why does the editorial emphasize explainable AI in biomedical machine learning?

Because machine learning models are often opaque in their decision-making, which raises concerns in critical medical contexts. Explainable AI aims to reveal how outcomes and recommendations are produced to support transparency and trust in decision-making.

How does data availability influence the need for augmented and explainable approaches?

The editorial links the expanding availability of medical and clinical data from interconnected biosensors in IoT settings to the need for training and refining ML models. It also notes that augmented techniques like GANs can generate synthetic samples to mitigate data scarcity and improve generalization.

What is the topical collection focused on?

The collection presents ten peer-reviewed papers on the latest advancements in explainable and augmented machine learning applied to biosignals and biomedical images, with each contribution undergoing a rigorous multi-reviewer revision process.

Aug 3, 2026

Upload by Violet