Sensors

Research

Jump to: Review

19 pages, 2258 KiB

Open AccessArticle

A Multidimensional Particle Swarm Optimization-Based Algorithm for Brain MRI Tumor Segmentation

by Zsombor Boga, Csanád Sándor and Péter Kovács

Sensors 2025, 25(9), 2800; https://doi.org/10.3390/s25092800 - 29 Apr 2025

Particle Swarm Optimization (PSO) has been extensively applied to optimization tasks in various domains, including image segmentation. In this work, we present a clustering-based segmentation algorithm that employs a multidimensional variant of PSO. Unlike conventional methods that require a predefined number of segments, [...] Read more.

Particle Swarm Optimization (PSO) has been extensively applied to optimization tasks in various domains, including image segmentation. In this work, we present a clustering-based segmentation algorithm that employs a multidimensional variant of PSO. Unlike conventional methods that require a predefined number of segments, our approach automatically selects an optimal segmentation granularity based on specified similarity criteria. This strategy effectively isolates brain tumors by incorporating both grayscale intensity and spatial information across multiple MRI modalities, allowing the method to be reliably tuned using a limited amount of training data. We further demonstrate how integrating these initial segmentations with a random forest classifier (RFC) enhances segmentation precision. Using MRI data from the RSNA-ASNR-MICCAI brain tumor segmentation (BraTS) challenge, our method achieves robust results with reduced reliance on extensive labeled datasets, offering a more efficient path toward accurate, clinically relevant tumor segmentation. Full article

(This article belongs to the Special Issue Sensors and Machine-Learning Based Signal Processing)

► Show Figures

Figure 1

19 pages, 5181 KiB

Open AccessArticle

Electric Motor Vibration Signal Classification Using Wigner–Ville Distribution for Fault Diagnosis

by Jian-Da Wu, Wen-Jun Luo and Kai-Chao Yao

Sensors 2025, 25(4), 1196; https://doi.org/10.3390/s25041196 - 15 Feb 2025

Viewed by 669

Abstract

Noise and vibration signal classification can be applied to fault diagnosis in mechanical and electronic systems such as electric vehicles. Traditional signal classification technology uses signal time and frequency domain characteristics as the identification basis. This study proposes a technique for visualizing sound [...] Read more.

Noise and vibration signal classification can be applied to fault diagnosis in mechanical and electronic systems such as electric vehicles. Traditional signal classification technology uses signal time and frequency domain characteristics as the identification basis. This study proposes a technique for visualizing sound signals using the Wigner–Ville distribution (WVD) method to extract vibration signal characteristics and artificial neural networks as the signal classification basis. A brushless motor is used as the machinery power source to verify the feasibility of this method to classify different signal vibration characteristics. In this experimental work, six states in various brushless motor revolutions were deliberately designed for measuring vibration signals. The brushless motor vibration signal is imaged using the WVD analysis method to extract the vibration signal characteristics. Through the WVD method, the brushless motor data is converted, and the YOLO (you only look once) deep coiling machine neural method is used to identify and classify the brushless motor WVD images. The Wagener analysis method parameters and recognition rates are discussed, thereby improving accurate motor fault diagnostic capabilities. This research provides a method for fault diagnosis that can be accurately performed without dismantling the brushless motor. The proposed approach can improve the reliability and stability of brushless motor applications. Full article

(This article belongs to the Special Issue Sensors and Machine-Learning Based Signal Processing)

► Show Figures

Figure 1

13 pages, 2233 KiB

Open AccessArticle

High-Quality Text-to-Speech Implementation via Active Shallow Diffusion Mechanism

by Junlin Deng, Ruihan Hou, Yan Deng, Yongqiu Long and Ning Wu

Sensors 2025, 25(3), 833; https://doi.org/10.3390/s25030833 - 30 Jan 2025

Viewed by 1027

Abstract

Denoising diffusion probabilistic models (DDPMs) have proven to be useful in text-to-speech (TTS) tasks; however, it has been a challenge for traditional diffusion models to carry out real-time processing because of the need for hundreds of sampling steps during the iteration. In this [...] Read more.

Denoising diffusion probabilistic models (DDPMs) have proven to be useful in text-to-speech (TTS) tasks; however, it has been a challenge for traditional diffusion models to carry out real-time processing because of the need for hundreds of sampling steps during the iteration. In this work, a two-stage fast inference and efficient diffusion-based acoustic model of TTS, the Cascaded MixGAN-TTS (CMG-TTS), is proposed to address this problem. An active shallow diffusion mechanism is adopted to divide the CMG-TTS training process into two stages. Specifically, a basic acoustic model in the first stage is trained to provide valuable a priori knowledge for the second stage, and for the underlying acoustic modeling, a mixture combination mechanism-based linguistic encoder is introduced to work with pitch and energy predictors. In the following stage of processing, a post-net is used to optimize the mel-spectrogram reconstruction performance. The CMG-TTS is evaluated on datasets such as the AISHELL3 and LJSpeech, and the experiments show that the CMG-TTS achieves satisfactory results in both subjective and objective evaluation metrics with only one denoising step. Compared to other TTS models based on diffusion modeling, the CMG-TTS obtains a leading score in the real time factor (RTF), and both stages of the CMG-TTS are effective in the ablation studies. Full article

(This article belongs to the Special Issue Sensors and Machine-Learning Based Signal Processing)

► Show Figures

Figure 1

16 pages, 3058 KiB

Open AccessArticle

Creating Refined Datasets for Better Chaos Detection

by Dariusz R. Augustyn, Katarzyna Harężlak, Agnieszka Szczęsna, Henryk Josiński, Paweł Kasprowski and Adam Świtoński

Sensors 2025, 25(3), 796; https://doi.org/10.3390/s25030796 - 28 Jan 2025

Viewed by 514

Abstract

In recent years, the analysis of signal properties (especially biomedical signals) has become an important research direction. One interesting feature of signals is their potential to be chaotic. This article concerns the issues of classification of real signals or synthetic ones in the [...] Read more.

In recent years, the analysis of signal properties (especially biomedical signals) has become an important research direction. One interesting feature of signals is their potential to be chaotic. This article concerns the issues of classification of real signals or synthetic ones in the context of detecting chaotic properties. In previous works, datasets of synthetic signals were created based on well-known chaotic and non-chaotic dynamical systems. They were published and used to train classifiers. This paper extends the previous studies and proposes a method for obtaining/extracting signals to force classifiers to learn to detect chaos. The proposed method allows the generation of groups of signals with similar initial conditions. The property of chaotic dynamical systems was used here, which consists of the strong dependence of the signal courses on a small change in the initial conditions. This method is based on reconstructing multidimensional phase space and data clustering. An additional goal of the work is to create referential datasets with so-called refined signals using the described method and to make them publicly available. The usefulness of the new datasets was confirmed during a simple experiment with the usage of the LSTM neural network. Full article

(This article belongs to the Special Issue Sensors and Machine-Learning Based Signal Processing)

► Show Figures

Figure 1

12 pages, 485 KiB

Open AccessArticle

Opportunities and Challenges for Clinical Practice in Detecting Depression Using EEG and Machine Learning

by Damir Mulc, Jaksa Vukojevic, Eda Kalafatic, Mario Cifrek, Domagoj Vidovic and Alan Jovic

Sensors 2025, 25(2), 409; https://doi.org/10.3390/s25020409 - 12 Jan 2025

Viewed by 1258

Abstract

Major depressive disorder (MDD) is associated with substantial morbidity and mortality, yet its diagnosis and treatment rates remain low due to its diverse and often overlapping clinical manifestations. In this context, electroencephalography (EEG) has gained attention as a potential objective tool for diagnosing [...] Read more.

Major depressive disorder (MDD) is associated with substantial morbidity and mortality, yet its diagnosis and treatment rates remain low due to its diverse and often overlapping clinical manifestations. In this context, electroencephalography (EEG) has gained attention as a potential objective tool for diagnosing depression. This study aimed to evaluate the effectiveness of EEG in identifying MDD by analyzing 140 EEG recordings from patients diagnosed with depression and healthy volunteers. Using various machine learning (ML) classification models, we achieved up to 80% accuracy in distinguishing individuals with MDD from healthy controls. Despite its promise, this approach has limitations. The variability in the clinical and biological presentations of depression, as well as patient-specific confounding factors, must be carefully considered when integrating ML technologies into clinical practice. Nevertheless, our findings suggest that an EEG-based ML model holds potential as a diagnostic aid for MDD, paving the way for further refinement and clinical application. Full article

(This article belongs to the Special Issue Sensors and Machine-Learning Based Signal Processing)

► Show Figures

Figure 1

26 pages, 4835 KiB

Open AccessArticle

Optimization of Imaging Reconnaissance Systems Using Super-Resolution: Efficiency Analysis in Interference Conditions

by Marta Bistroń and Zbigniew Piotrowski

Sensors 2024, 24(24), 7977; https://doi.org/10.3390/s24247977 - 13 Dec 2024

Cited by 2 | Viewed by 958

Abstract

Image reconnaissance systems are critical in modern applications, where the ability to accurately detect and identify objects is crucial. However, distortions in real-world operational conditions, such as motion blur, noise, and compression artifacts, often degrade image quality, affecting the performance of detection systems. [...] Read more.

Image reconnaissance systems are critical in modern applications, where the ability to accurately detect and identify objects is crucial. However, distortions in real-world operational conditions, such as motion blur, noise, and compression artifacts, often degrade image quality, affecting the performance of detection systems. This study analyzed the impact of super-resolution (SR) technology, in particular, the Real-ESRGAN model, on the performance of a detection model under disturbed conditions. The methodology involved training and evaluating the Faster R-CNN detection model with original and modified data sets. The results showed that SR significantly improved detection precision and mAP in most interference scenarios. These findings underscore SR’s potential to improve imaging systems while identifying key areas for future development and further research. Full article

(This article belongs to the Special Issue Sensors and Machine-Learning Based Signal Processing)

► Show Figures

Figure 1

21 pages, 6786 KiB

Open AccessArticle

Bearing-DETR: A Lightweight Deep Learning Model for Bearing Defect Detection Based on RT-DETR

by Minggao Liu, Haifeng Wang, Luyao Du, Fangsong Ji and Ming Zhang

Sensors 2024, 24(13), 4262; https://doi.org/10.3390/s24134262 - 30 Jun 2024

Cited by 8 | Viewed by 3124

Abstract

Detecting bearing defects accurately and efficiently is critical for industrial safety and efficiency. This paper introduces Bearing-DETR, a deep learning model optimised using the Real-Time Detection Transformer (RT-DETR) architecture. Enhanced with Dysample Dynamic Upsampling, Efficient Model Optimization (EMO) with Meta-Mobile Blocks (MMB), and [...] Read more.

Detecting bearing defects accurately and efficiently is critical for industrial safety and efficiency. This paper introduces Bearing-DETR, a deep learning model optimised using the Real-Time Detection Transformer (RT-DETR) architecture. Enhanced with Dysample Dynamic Upsampling, Efficient Model Optimization (EMO) with Meta-Mobile Blocks (MMB), and Deformable Large Kernel Attention (D-LKA), Bearing-DETR offers significant improvements in defect detection while maintaining a lightweight framework suitable for low-resource devices. Validated on a dataset from a chemical plant, Bearing-DETR outperformed the standard RT-DETR, achieving a mean average precision (mAP) of 94.3% at IoU = 0.5 and 57.5% at IoU = 0.5–0.95. It also reduced floating-point operations (FLOPs) to 8.2 G and parameters to 3.2 M, underscoring its enhanced efficiency and reduced computational demands. These results demonstrate the potential of Bearing-DETR to transform maintenance strategies and quality control across manufacturing environments, emphasising adaptability and impact on sustainability and operational costs. Full article

(This article belongs to the Special Issue Sensors and Machine-Learning Based Signal Processing)

► Show Figures

Figure 1

Review

Jump to: Research

25 pages, 309 KiB

Open AccessReview

Causality, Machine Learning, and Feature Selection: A Survey

by Asmae Lamsaf, Rui Carrilho, João C. Neves and Hugo Proença

Sensors 2025, 25(8), 2373; https://doi.org/10.3390/s25082373 - 9 Apr 2025

Viewed by 577

Abstract

Causality, which involves distinguishing between cause and effect, is essential for understanding complex relationships in data. This paper provides a review of causality in two key areas: causal discovery and causal inference. Causal discovery transforms data into graphical structures that illustrate how variables [...] Read more.

Causality, which involves distinguishing between cause and effect, is essential for understanding complex relationships in data. This paper provides a review of causality in two key areas: causal discovery and causal inference. Causal discovery transforms data into graphical structures that illustrate how variables influence one another, while causal inference quantifies the impact of these variables on a target outcome. The models are more robust and accurate with the integration of causal reasoning into machine learning, improving applications like prediction and classification. We present various methods used in detecting causal relationships and how these can be applied in selecting or extracting relevant features, particularly from sensor datasets. When causality is used in feature selection, it supports applications like fault detection, anomaly detection, and predictive maintenance applications critical to the maintenance of complex systems. Traditional correlation-based methods of feature selection often overlook significant causal links, leading to incomplete insights. Our research highlights how integrating causality can be integrated and lead to stronger, deeper feature selection and ultimately enable better decision making in machine learning tasks. Full article

(This article belongs to the Special Issue Sensors and Machine-Learning Based Signal Processing)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Sensors and Machine-Learning Based Signal Processing

Share This Special Issue

Special Issue Editors

Special Issue Information

Keywords

Benefits of Publishing in a Special Issue

Published Papers (8 papers)

Research

Review

Further Information

Guidelines

MDPI Initiatives

Follow MDPI