Applied Sciences

16 pages, 1716 KB

Open AccessArticle

Between-Limb Asymmetry Assessment During a Double-Leg Drop Jump Task After Anterior Cruciate Ligament Reconstruction—A Musculoskeletal Modelling Approach

by Rodrigo B. Mateus, Sílvia Cabral, Chris Richter and António P. Veloso

Appl. Sci. 2025, 15(22), 12347; https://doi.org/10.3390/app152212347 - 20 Nov 2025

Viewed by 1082

Abstract

Approximately two-thirds of athletes who are submitted to Anterior Cruciate Ligament Reconstruction (ACLR) never return to their preinjury level of performance, potentially due to muscle strength deficiencies or altered loading patterns during landing or jumping tasks. This study aimed to estimate individual muscle [...] Read more.

Approximately two-thirds of athletes who are submitted to Anterior Cruciate Ligament Reconstruction (ACLR) never return to their preinjury level of performance, potentially due to muscle strength deficiencies or altered loading patterns during landing or jumping tasks. This study aimed to estimate individual muscle forces during a double-leg drop jump task, and assess sagittal plane between-limb asymmetries in muscle forces and ground reaction forces using a musculoskeletal modelling approach, in athletes who underwent ACLR. Thirty male field-sport athletes (age: 18–35 years; mass: 84.3 ± 12.3 kg; height: 180.2 ± 8.4 cm) post-ACLR (39.8 ± 3.9 weeks) using patellar or quadriceps tendon grafts were tested. Scaled musculoskeletal models were implemented in OpenSim, and muscle forces were estimated using the Computed Muscle Control optimization method. The contralateral limb exhibited greater vertical ground reaction forces across most of the rebound phase (d = 2.01). Compared with the contralateral limb, the ACLR limb showed reduced quadriceps (d = 1.72), soleus (d = 0.95), and gluteus maximus (d = 0.83) forces, indicating deficits in knee extensor, plantarflexor, and hip extensor neuromuscular function. Smaller asymmetries were found for the gluteus medius (d = 0.60) and hamstrings (d = 0.72), while other muscles showed symmetrical activation patterns. These results reveal persistent between-limb asymmetries in muscle recruitment and loading up to nine months post-ACLR, emphasizing the importance of targeted rehabilitation to restore symmetrical neuromuscular control during explosive movements. Full article

(This article belongs to the Special Issue Novel Approaches of Physical Therapy-Based Rehabilitation)

► Show Figures

Figure 1

15 pages, 4120 KB

Open AccessArticle

Case Study on Compression of Vibration Data for Distributed Wireless Condition Monitoring Systems

by Rick Pandey, Felix Grimm, Dominik Nille, Christoph Böckenhoff, Jonathan Gamez, Sebastian Uziel, Albert Dorneich, Tino Hutschenreuther and Silvia Krug

Appl. Sci. 2025, 15(22), 12346; https://doi.org/10.3390/app152212346 - 20 Nov 2025

Viewed by 893

Abstract

To build robust condition monitoring solutions, it is important to identify signals that capture relevant information. However, how a degradation affects a given part of machinery might not be clear at the beginning. As a result, exploration measurement campaigns collecting large amounts of [...] Read more.

To build robust condition monitoring solutions, it is important to identify signals that capture relevant information. However, how a degradation affects a given part of machinery might not be clear at the beginning. As a result, exploration measurement campaigns collecting large amounts of data are needed for initial evaluation. Vibration signals are typical examples of such data. Although, for explorative measurement campaigns, the battery-powered wireless node brings extra flexibility in terms of positioning the sensor at the desired location and facilitates retrofitting, the limited energy posed by them is the major downside. Sending high-sampled data over wireless channels is costly energy-wise if all samples are to be sent. When multiple sensor nodes transmit real-time measurement data concurrently over a wireless channel, the risk of channel saturation increases significantly. Avoiding this requires identifying an optimal balance between sampling time, transmission duration, and payload size. This can be done by processing and compressing data before transmission, on the sensor node close to the data acquisition and later reconstructing the received samples on the central node. In this paper, we analyze two compression mechanisms to ensure a good compression ratio and still allow good signal reconstruction for later analysis. We study two approaches, one based on the Fast Fourier Transform and one on Singular Value Decomposition, and discuss the pros and cons of each variant. Full article

(This article belongs to the Special Issue Advances in Machinery Fault Diagnosis and Condition Monitoring)

► Show Figures

Figure 1

16 pages, 3494 KB

Open AccessArticle

Comparison of Pretreatment Methods for Obtaining Collagen Hydrolysates from the Swim Bladder of Totoaba macdonaldi and Their Negative Impact on Cancer Cells

by Evelin Martínez-Benavidez, Angélica María Vergara-Pineda, Jesús Cervantes-Martínez, José Leonardo Puch-Sánchez, Sandra Daniela Bravo, Ofelia Yadira Lugo-Melchor, Hugo S. García and Inocencio Higuera-Ciapara

Appl. Sci. 2025, 15(22), 12345; https://doi.org/10.3390/app152212345 - 20 Nov 2025

Viewed by 953

Abstract

The search for therapeutic bioactive peptides has led to the utilization of marine byproducts as collagen sources. This study evaluated the effect of collagen hydrolysates (CH) obtained from the swim bladder (SB) of Totoaba macdonaldi on breast (MCF-7) and colorectal (Caco-2) adenocarcinoma cells [...] Read more.

The search for therapeutic bioactive peptides has led to the utilization of marine byproducts as collagen sources. This study evaluated the effect of collagen hydrolysates (CH) obtained from the swim bladder (SB) of Totoaba macdonaldi on breast (MCF-7) and colorectal (Caco-2) adenocarcinoma cells and on human dermal fibroblasts (CRL-1474), considering the need for less invasive and less toxic treatment alternatives. Two pretreatment methods for the SB were compared: (1) NaOH and butanol (SBPT), and (2) hexane (SBDF). The pretreated tissues underwent direct enzymatic hydrolysis using bromelain. The resulting hydrolysates were characterized by SDS-PAGE, Raman spectroscopy, and chromatographic profiling. Both pretreatments preserved the structure of type I collagen. Bromelain hydrolysis was efficient, yielding peptides with molecular weights below 20 kDa for CH-SBPT and below 10 kDa for CH-SBDF. CH of Totoaba macdonaldi significantly reduced MCF-7 and Caco-2 cells viability, particularly at 20 mg/mL. In CRL-1474 fibroblasts, CH-SBDF stimulated cell proliferation, while CH-SBPT had neutral effects. Hexane pretreatment is a viable alternative to NaOH, reducing processing steps without compromising yield or bioactivity. CH derived from Totoaba macdonaldi exhibit promising anticancer and regenerative properties, suggesting potential biomedical applications. Further research is needed to isolate specifically active peptides and elucidate their mechanisms of action. Full article

(This article belongs to the Section Applied Biosciences and Bioengineering)

► Show Figures

Figure 1

14 pages, 1143 KB

Open AccessArticle

Vibration Analysis of Cantilever Beam with Free End Resting on 3D-Printed Spring and Considering the Effect of Accelerometer and Exciter Masses

by Hassan H. Mahdi, Sami A. Nama, Marwan T. Mezher and Tomasz Trzepieciński

Appl. Sci. 2025, 15(22), 12344; https://doi.org/10.3390/app152212344 - 20 Nov 2025

Viewed by 1575

Abstract

A cantilever beam is a mechanical structure fixed at one end and free at the other. It converts the applied external forces into bending and shear force; therefore, it should be designed to resist deflection. The fundamental natural frequency of the cantilever beam [...] Read more.

A cantilever beam is a mechanical structure fixed at one end and free at the other. It converts the applied external forces into bending and shear force; therefore, it should be designed to resist deflection. The fundamental natural frequency of the cantilever beam depends on its material properties, geometry, and supporting conditions. This work studied the effect of adding an accelerometer and a motor, which represent multiple masses, on the fundamental natural frequency of a cantilever beam. The beam is also supported by a 3D-printed spring at its free end. Three-dimensional-printed springs with different infill percentages (20%, 50%, 80%, and 100%) were used with different infill patterns (concentric, grid, and triangle) to study the effect of these parameters on natural frequency. The results showed that the triangle pattern gives the best results for fundamental natural frequency and resulting force values. In addition to that, the triangle pattern with 80% infill percentage is preferred for printing as compared with 100% infill percentage because it gives better vibration results. Full article

(This article belongs to the Section Additive Manufacturing Technologies)

► Show Figures

Figure 1

31 pages, 3054 KB

Open AccessArticle

Outlier Detection in EEG Signals Using Ensemble Classifiers

by Agnieszka Duraj, Natalia Łukasik and Piotr S. Szczepaniak

Appl. Sci. 2025, 15(22), 12343; https://doi.org/10.3390/app152212343 - 20 Nov 2025

Viewed by 891

Abstract

Epilepsy is one of the most prevalent neurological disorders, affecting over 50 million people worldwide. Accurate detection and characterization of epileptic activity are clinically critical, as seizures are associated with substantial morbidity, mortality, and impaired quality of life. Electroencephalography (EEG) remains the gold [...] Read more.

Epilepsy is one of the most prevalent neurological disorders, affecting over 50 million people worldwide. Accurate detection and characterization of epileptic activity are clinically critical, as seizures are associated with substantial morbidity, mortality, and impaired quality of life. Electroencephalography (EEG) remains the gold standard for epilepsy assessment; however, its manual interpretation is time-consuming, subjective, and prone to inter-rater variability, emphasizing the need for automated analytical approaches. This study proposes an automated ensemble classification framework for outlier detection in EEG signals. Three interpretable baseline models—Support Vector Machine (SVM), k-Nearest Neighbors (k-NN), and decision tree (DT-CART)—were screened. Ensembles were formed only from base models that had a pre-registered meta-selection rule (

F_{1}

on the outlier-class

> 0.60

). Under this criterion, DT-CART did not qualify and was excluded from all ensembles; final ensembles combined SVM and k-NN. The framework was evaluated on two publicly available datasets with distinct acquisition conditions. The Bonn EEG dataset comprises 500 artifact-free single-channel recordings from healthy subjects and epilepsy patients under controlled laboratory settings. In contrast, the Guinea-Bissau and Nigeria Epilepsy (GBNE) dataset contains multi-channel EEG recordings from 97 participants acquired in field conditions using low-cost equipment, reflecting real-world diagnostic challenges such as motion artifacts and signal variability. The ensemble framework substantially improved outlier detection performance, with stacking achieving up to a 95.0%

F_{1}

-score (accuracy 95.0%) on the Bonn dataset and 85.5%

F_{1}

-score (accuracy 85.5%) on the GBNE dataset. These findings demonstrate that the proposed approach provides a robust, interpretable, and generalizable solution for EEG analysis, with strong potential to enhance reliable, efficient, and scalable epilepsy detection in both laboratory and resource-limited clinical environments. Full article

(This article belongs to the Special Issue EEG Signal Processing in Medical Diagnosis Applications)

► Show Figures

Figure 1

16 pages, 3005 KB

Open AccessArticle

Assessment of Oncology Patients’ Knowledge on Skin Care During and After Radiotherapy Treatment

by Emilia Klimaszewska, Joanna Jońska, Marta Ogorzałek, Paulina Marcula, Monika Maj, Dawid Aleksandrowicz and Ryszard Tomasiuk

Appl. Sci. 2025, 15(22), 12342; https://doi.org/10.3390/app152212342 - 20 Nov 2025

Viewed by 880

Abstract

Radiation therapy is one of the methods of cancer treatment, using ionizing radiation, which can lead to many negative skin changes. Therefore, its proper care is critical. This provided the impetus for an attempt to assess oncology patients’ knowledge of skin care during [...] Read more.

Radiation therapy is one of the methods of cancer treatment, using ionizing radiation, which can lead to many negative skin changes. Therefore, its proper care is critical. This provided the impetus for an attempt to assess oncology patients’ knowledge of skin care during and after radiation therapy treatment. Methods: An anonymous and voluntary questionnaire survey was conducted among 100 patients during and after radiotherapy treatment at the Bohaterów Radomskiego Czerwca 76 Radom Oncology Center in Poland. The chi-square independence test was used to examine the independence of certain features. In case it was not possible to verify the hypothesis of the independence of features without the Yates correction, the chi-square independence test, with this correction, was used. Results: Fifty-three men and 47 women were included in the study, with prostate cancer (39%) and breast cancer (35%) being the most common, respectively. Most of the respondents were aged 56+. 67% of respondents indicated having negative skin lesions during or after radiation therapy treatment, which occurred most often in the areas of the head and neck (32%), chest (10%), and upper extremities (9%). 57% of respondents did not visit medical personnel or a cosmetologist for skin care. In contrast, 24% indicated a belief in a high level of knowledge about skin care during and after radiation therapy treatment. Conclusions: Analysis of the obtained results may lead to further development of certain issues in medicine. They are, among others, the need to increase awareness of society that has contact with oncological therapy on various actions which impact the maintenance of good skin condition, improvement of its appearance, and alleviation of symptoms which appear during or after radiotherapy. Full article

► Show Figures

Figure 1

33 pages, 17069 KB

Open AccessArticle

Development of a CAD–FEA Integrated Automation Add-In for DfAM-Aware Topology Optimization: A Case Study on an Additively Manufactured Pusher Duct Support Bracket for a Novel UAV Prototype

by H. Kursat Celik, Ali Elham, Recep Cinar, M. Ali Erbil, Robert Entwistle, Allan E. W. Rennie and Ibrahim Akinci

Appl. Sci. 2025, 15(22), 12341; https://doi.org/10.3390/app152212341 - 20 Nov 2025

Cited by 2 | Viewed by 1477

Abstract

The integration of additive manufacturing (AM) and topology optimization (TO) is transforming mechanical design and prototyping practices across multiple engineering sectors, including agricultural and aerospace applications. This study presents the development of TODfAM, a bespoke SOLIDWORKS add-in that automates TO workflows and embeds [...] Read more.

The integration of additive manufacturing (AM) and topology optimization (TO) is transforming mechanical design and prototyping practices across multiple engineering sectors, including agricultural and aerospace applications. This study presents the development of TODfAM, a bespoke SOLIDWORKS add-in that automates TO workflows and embeds Design for Additive Manufacturing (DfAM) principles directly within a parametric CAD environment. The tool integrates parametric modelling, finite element analysis (FEA)-based structural evaluation, and TO in a unified platform, enabling automated generation and assessment of design iterations with respect to both mechanical performance and AM-specific manufacturability constraints. A case study on a pusher-duct support bracket for an Unmanned Aerial Vehicle (UAV) was conducted to demonstrate the functionality of the developed workflow. The optimized bracket achieved a 13.77% mass reduction while maintaining structural integrity under representative loading conditions. The CAD-integrated framework reduces toolchain hand-offs and allows early manufacturability evaluation within the design environment, thereby improving workflow continuity and consistency. The principal novelty of this work lies in the establishment of a fully CAD-native, DfAM-aware optimization framework that consolidates the design-to-manufacturing process into a single automated environment. This approach not only streamlines pre- and post-processing tasks but also promotes wider industrial adoption of AM by providing a practical, designer-oriented route to lightweight and manufacturable structures. Full article

(This article belongs to the Section Additive Manufacturing Technologies)

► Show Figures

Figure 1

34 pages, 466 KB

Open AccessArticle

biLorentzFM: Hyperbolic Multi-Objective Deep Learning for Reciprocal Recommendation

by Kübra Karacan Uyar and Yücel Batu Salman

Appl. Sci. 2025, 15(22), 12340; https://doi.org/10.3390/app152212340 - 20 Nov 2025

Cited by 1 | Viewed by 1190

Abstract

Reciprocal recommendation requires satisfying preferences on both sides of a match, which differs from standard one-sided settings and often involves hierarchical structure (e.g., skills, seniority, education). We present biLorentzFM, which is a multi-objective framework that integrates hyperbolic geometry into factorization machine architectures using [...] Read more.

Reciprocal recommendation requires satisfying preferences on both sides of a match, which differs from standard one-sided settings and often involves hierarchical structure (e.g., skills, seniority, education). We present biLorentzFM, which is a multi-objective framework that integrates hyperbolic geometry into factorization machine architectures using Lorentz embeddings with learnable curvature and manifold-aware optimization. The approach addresses whether a geometric structure aligned with hierarchical relationships can improve reciprocal matching without requiring major architectural changes. On a large-scale recruitment dataset from Kariyer.Net (1,150,302 interactions, 229,805 candidates), the model achieves candidate and company AUCs of 0.9964 and 0.9913 respectively, representing 6.6% and 6.0% improvements over the strongest Euclidean baseline while maintaining practical inference latency (2.1 ms per batch). Cross-validation analysis confirms robustness (5-fold: 0.9813 ± 0.0002; 3-seed: 0.9964 ± 0.0012) with very large effect sizes (Cohen’s d = 2.89–3.08). Although the per-epoch training time increases by 23.5% due to manifold operations, faster convergence (12 vs. 18 epochs) reduces the total training time by 17.8%. Cross-domain evaluation on Speed Dating data demonstrates generalization beyond explicit hierarchies with a 2.8% AUC improvement despite lacking structured taxonomies. Learned curvature parameters differ by entity type, providing interpretable indicators of hierarchical structure strength. Ablation studies isolate contributions from geometric structure (6.6%), learnable curvature (4.7%), multi-objective learning (2.1%), and explicit feature interactions (0.6%). A systematic comparison reveals that Lorentz embeddings outperform Poincaré ball implementations by 4.4% AUC under identical conditions, which is attributed to numerical stability advantages. The results indicate that pairing standard recommendation architectures with geometry reflecting hierarchical relationships can provide consistent improvements for reciprocal matching, while limitations including cold-start performance, computational overhead at an extreme scale, and static hierarchy assumptions suggest directions for future work on adaptive curvature, fairness constraints, and dynamic taxonomies. Full article

► Show Figures

Figure 1

21 pages, 13964 KB

Open AccessArticle

Cutting-Load Characteristics of Excavation Machine Picks in Hydraulic-Precracked Coal–Rock

by Qingguo Dong, Hongmei Liu and Yi Xu

Appl. Sci. 2025, 15(22), 12339; https://doi.org/10.3390/app152212339 - 20 Nov 2025

Viewed by 436

Abstract

Hydraulic pre-fracturing is a rock-weakening technique applied in hard-rock excavation. To investigate the effects of hydraulic pre-fracturing on crack propagation in excavation roadwalls and on the cutting loads experienced by excavation machine picks when cutting precracked rock, a two-way fluid–solid couplinproach (CFD–DEM) was [...] Read more.

Hydraulic pre-fracturing is a rock-weakening technique applied in hard-rock excavation. To investigate the effects of hydraulic pre-fracturing on crack propagation in excavation roadwalls and on the cutting loads experienced by excavation machine picks when cutting precracked rock, a two-way fluid–solid couplinproach (CFD–DEM) was employed to simulate the three-dimensional crack propagation process of a rock face under hydraulic fracturing. The results indicate that crack propagation under hydraulic fracturing evolves through four stages: (1) initiation of the primary crack; (2) further development of the primary crack, accompanied by the emergence of fine subsidiary cracks; (3) retardation of the primary crack growth, concurrent with propagation of secondary cracks; and (4) further expansion of secondary cracks. The influences of borehole aperture and injection pressure on crack propagation were analyzed; within the investigated ranges, increasing either aperture or injection pressure produced a nonlinear increase in crack development. When the hydraulic-fracture borehole diameter increased from 85 mm to 100 mm, the number of broken bonds increased by 56.2%; when the injection pressure increased from 25 MPa to 40 MPa, the number of broken bonds increased by 153.9%. The cutting force experienced by picks when cutting precracked rock decreased by 9.05% compared with cutting intact (non-precracked) rock; after precracking, the mean forces in the Z and Y directions decreased by 11.46% and 7.20%, respectively, whereas the mean force in the X direction increased by 5.49%. The findings provide reference data for the practical implementation of hydraulic pre-fracturing in hard-rock excavation. Full article

► Show Figures

Figure 1

21 pages, 2076 KB

Open AccessArticle

Groundwater Quality Near Riverbanks and Its Suitability for Agricultural Use in Semi-Arid Regions

by Layth Saleem Salman Al-Shihmani, Ali Jawad Al-Sarraji, Ahmed Abed Gatea Al-Shammary, Jesús Fernández-Gálvez and Andrés Caballero-Calvo

Appl. Sci. 2025, 15(22), 12338; https://doi.org/10.3390/app152212338 - 20 Nov 2025

Viewed by 728

Abstract

Water scarcity has become one of the most pressing challenges to agricultural sustainability, particularly in arid and semi-arid regions where climate change, dam construction, and rapid population growth have intensified the pressure on water and food resources. Groundwater adjacent to rivers represents a [...] Read more.

Water scarcity has become one of the most pressing challenges to agricultural sustainability, particularly in arid and semi-arid regions where climate change, dam construction, and rapid population growth have intensified the pressure on water and food resources. Groundwater adjacent to rivers represents a potential supplementary resource that can reduce reliance on restricted surface water supplies. This study assessed the hydrochemical characteristics and agricultural suitability of shallow groundwater located near the Tigris River, Iraq. Fieldwork involved monitoring four active wells and collecting samples over six periods from October 2022 to May 2023, combined with twelve soil samples from surrounding agricultural fields. Laboratory analyses determined key water and soil properties, including pH, electrical conductivity, major cations and anions, and a range of salinity and sodicity indices such as total dissolved solids (TDS), sodium adsorption ratio (SAR), residual sodium carbonate (RSC), potential salinity (PS), magnesium ratio, Simpson ratio (SR), Jones ratio (JR), and sodium percentage (Na%). Results indicated that groundwater levels fluctuated seasonally in tandem with the Tigris River, which directly influenced salinity levels. SI values were positive, TDS values were in the high salinity class, RSC values were consistently negative, PS values were in the medium to poor category, Na% values and MR values were within acceptable limits for irrigation, and SR values were moderately to highly contaminated. Groundwater quality, according to the U.S. Salinity Laboratory classification, was categorized between the C4S1 class (very high salinity, low sodium) and the C3S1 (high salinity, low sodium). Soil analyses showed predominantly light-textured soils with moderate Ec and SAR values below sodicity thresholds. The combination of soil permeability and groundwater characteristics suggests that irrigation is feasible under specific management practices. The study concludes that groundwater adjacent to rivers can serve as a valuable supplementary source for agriculture in semi-arid regions. Its use is most effective when applied to salt-tolerant crops, supported by leaching requirements, or blended with fresh water. These findings emphasize the importance of integrated groundwater management for enhancing agricultural resilience and sustainable land use under water-scarce conditions. Excessive extraction of groundwater near rivers can also pose long-term sustainability challenges. Full article

(This article belongs to the Special Issue New Approaches to Water Treatment: Challenges and Trends, 2nd Edition)

► Show Figures

Figure 1

19 pages, 2710 KB

Open AccessArticle

Internet of Things-Based Electromagnetic Compatibility Monitoring (IEMCM) Architecture for Biomedical Devices

by Chiedza Hwata, Gerard Rushingabigwi, Omar Gatera, Didacienne Mukalinyigira, Celestin Twizere, Bolaji N. Thomas and Diego H. Peluffo-Ord’onez

Appl. Sci. 2025, 15(22), 12337; https://doi.org/10.3390/app152212337 - 20 Nov 2025

Cited by 1 | Viewed by 1045

Abstract

Electromagnetic compatibility is the capability of electrical and electronic equipment to function properly around devices radiating electromagnetic energy, without mutual disturbance. Hospital environments contain numerous devices operating simultaneously and sharing resources. Undetected electromagnetic interference can cause medical devices’ malfunctions, exposing patients and staff. [...] Read more.

Electromagnetic compatibility is the capability of electrical and electronic equipment to function properly around devices radiating electromagnetic energy, without mutual disturbance. Hospital environments contain numerous devices operating simultaneously and sharing resources. Undetected electromagnetic interference can cause medical devices’ malfunctions, exposing patients and staff. Traditional monitoring is time-consuming and relies on expert interpretation. An Internet of Things-enabled embedded system architecture for remote and real-time monitoring of electromagnetic fields from medical devices is proposed. It integrates frequency probes, a Raspberry Pi 4, and a communication module. A three-month study conducted at Muhima District Hospital, Kigali, Rwanda, demonstrated the system’s effectiveness in monitoring electromagnetic field levels and cloud transmission. The signals were benchmarked against International Electrotechnical Commission and Rwanda Standards Board standards. Alerts are triggered when thresholds are exceeded, with results plotted on website and mobile interfaces. Emissions were highest at noon when the equipment was most active and lower after 1:30 PM, indicating reduced activity. The sample recorded statistics of electric fields include mean (1.0028), minimum (0.7228), and maximum (1.3515). Among the five filters evaluated, the Savitzky–Golay performed better, with MSE (0.235) and SNR (9.308). A 412 ms average latency and 24 h operation was achieved, offering a portable solution for hospital safety and equipment optimization. Full article

► Show Figures

Figure 1

29 pages, 3446 KB

Open AccessArticle

QRetinex-Net: A Quaternion Retinex Framework for Bio-Inspired Color Constancy

by Sos Agaian and Vladimir Frants

Appl. Sci. 2025, 15(22), 12336; https://doi.org/10.3390/app152212336 - 20 Nov 2025

Cited by 1 | Viewed by 818

Abstract

Color constancy, the ability to perceive consistent object colors under varying illumination, is a core function of the human visual system and a persistent challenge in machine vision. Retinex theory models this process by decomposing an image

S

into reflectance (

R

) [...] Read more.

Color constancy, the ability to perceive consistent object colors under varying illumination, is a core function of the human visual system and a persistent challenge in machine vision. Retinex theory models this process by decomposing an image

S

into reflectance (

R

) and illumination (

I

) components (

S^{'} = R I

). However, conventional Retinex methods suffer from key limitations: independent RGB processing that disrupts inter-channel correlations, weak grounding in color perception models, non-invertible decomposition (

S^{'} \neq S

), and limited biological plausibility. We propose QRetinex-Net, a unified Retinex framework formulated in the quaternion domain—

S = R \otimes I

, where

\otimes

denotes the Hamilton product. Representing RGB channels as pure quaternions enables holistic color processing, biologically inspired modeling, and invertible image reconstruction. We further introduce the Reflectance Consistency Index (RCI) to quantitatively assess illumination invariance and reflectance stability. Experiments on low-light crack detection, infrared–visible fusion, and face detection under varying lighting demonstrate that QRetinex-Net outperforms RetinexNet, KIND++, U-RetinexNet, and Diff-Retinex, achieving up to 11% performance gains, LPIPS ≈ 0.0001, and RCI ≈ 0.988. Full article

(This article belongs to the Section Computing and Artificial Intelligence)

► Show Figures

Figure 1

21 pages, 9492 KB

Open AccessArticle

Integration of Geophysical Methods to Obtain a Geoarchaeological Model of the Santa Lucia di Mendola Site (Southeastern Sicily—Italy)

by Gabriele Morreale, Sabrina Grassi, Carlos José Araque-Pérez, Angelo Gilotti, Rosa Lanteri, Ermelinda Storaci, Teresa Teixidó and Sebastiano Imposa

Appl. Sci. 2025, 15(22), 12335; https://doi.org/10.3390/app152212335 - 20 Nov 2025

Viewed by 858

Abstract

Geophysical prospecting has increasingly become a fundamental tool in archaeological research thanks to its ability to rapidly investigate large areas and detect underground structures without impacting the ground. In this study, an integrated geophysical approach was applied to the early Christian archaeological site [...] Read more.

Geophysical prospecting has increasingly become a fundamental tool in archaeological research thanks to its ability to rapidly investigate large areas and detect underground structures without impacting the ground. In this study, an integrated geophysical approach was applied to the early Christian archaeological site of Santa Lucia di Mendola, located in southeastern Sicily (Italy). The site is characterised by a complex stratigraphy developed through the exploitation of existing karst features within the limestone lithotype and includes a dating back to the 4th century A.D. hypogeum, surmounted by the remains of a Byzantine Basilica and a small sacristy carved into the rock. A comprehensive geophysical survey was performed to determine a geoarchaeological model of the area. To evaluate and compare the geophysical responses, some of the main geophysical methods used in archaeology were applied: seismic refraction method (SRT), geoelectric method (ERT), frequency domain electromagnetic method (FDEM) and magnetic survey (MAG). The anomalies identified suggest the presence of additional structures dug into the subsoil, probably connected to those currently accessible. This hypothesis is supported by presence of the remains of a wall located at the northern end of the sacristy corridor, which separates this part of the passage from another area visibly filled with rubble. Full article

(This article belongs to the Special Issue Advances in Applied Geophysics for Environmental Monitoring and Cultural Heritage)

► Show Figures

Figure 1

22 pages, 9007 KB

Open AccessArticle

Numerical Analysis of Aerodynamic Drag Reduction for a DrivAer Automobile Model Using Rear Air Jets

by Shun Liu, Tao Chen and Wenjie Zhou

Appl. Sci. 2025, 15(22), 12334; https://doi.org/10.3390/app152212334 - 20 Nov 2025

Viewed by 1082

Abstract

This paper presents a numerical investigation into aerodynamic drag reduction by air jets for a realistic DrivAer estateback vehicle model. Numerical simulations are conducted based on Reynolds-Averaged Navier–Stokes equations with a shear stress transport k-ω turbulence model, for optimizing the drag reduction with [...] Read more.

This paper presents a numerical investigation into aerodynamic drag reduction by air jets for a realistic DrivAer estateback vehicle model. Numerical simulations are conducted based on Reynolds-Averaged Navier–Stokes equations with a shear stress transport k-ω turbulence model, for optimizing the drag reduction with seven individual rear slot jets and their combination. The results demonstrate that the jets located at the upper and lower edges of the rear end could achieve the highest individual drag reduction of up to 4.82%, by suppressing recirculation bubbles, delaying flow separation, and promoting pressure recovery. The jet positioned at the lower lateral side of vehicle base reduces the drag by 4.14% through the control of the underbody vortex. Moderate performance is observed for other individual jets within the wake flow. The underlying mechanisms are elucidated by detailed analyses of wake flow fields and rear-end surface pressure distributions. On this basis, optimal performance is obtained by a multi-jet combination, incorporating the best vertical jet and three better horizontal jets, which collectively yield a remarkable 11.80% drag reduction with high energy efficiency. This work confirms that the active flow control by the rear air jets can greatly improve the aerodynamic efficiency for realistic vehicles, providing a practical approach for drag reduction in modern automotive applications. Full article

(This article belongs to the Special Issue Recent Advances in Engineering Applications of Computational Fluid Dynamics: 2nd Edition)

► Show Figures

Figure 1

15 pages, 7926 KB

Open AccessArticle

The Ragweed Finder: A Citizen-Science Project to Inform Pollen Allergy Sufferers About Ambrosia artemisiifolia Populations in Austria

by Lukas Dirr, Katharina Bastl, Maximilian Bastl, Uwe Edwin Berger, Johannes Martin Bouchal, Andreja Kofol Seliger, Donát Magyar, Jana Ščevková, Tamás Szigeti and Friðgeir Grímsson

Appl. Sci. 2025, 15(22), 12333; https://doi.org/10.3390/app152212333 - 20 Nov 2025

Cited by 1 | Viewed by 885

Abstract

Ambrosia artemisiifolia (ragweed) is a highly invasive species that produces large amounts of allergenic pollen. This poses a serious health risk to allergy sufferers. The “Ragweed Finder” is an Austrian citizen science platform (website and app) that enables the public to report occurrences [...] Read more.

Ambrosia artemisiifolia (ragweed) is a highly invasive species that produces large amounts of allergenic pollen. This poses a serious health risk to allergy sufferers. The “Ragweed Finder” is an Austrian citizen science platform (website and app) that enables the public to report occurrences of ragweed, which are then verified by experts. Over 90% of reports are confirmed as positive, with most originating from eastern Austria, where ragweed is widespread. The number of reports has generally increased over time, except in 2020 during the pandemic. Report frequency does not directly correlate with daily pollen concentrations, but peaks before and during pollen season. Most observations occur along traffic routes, likely due to seed dispersal by vehicle airflow or easier accessibility for users. Verified observations are displayed on an interactive map, helping allergy sufferers to avoid exposure and informing local authorities of the need for targeted control actions. The data are also used to raise awareness among policymakers and help to enact the first law for the control and prevention of ragweed in Burgenland (Austria), in 2021: the “Burgenland Ragweed Control Act”. This demonstrates the success of the “Ragweed Finder” as an important tool for monitoring this invasive species in Austria. Full article

(This article belongs to the Section Environmental Sciences)

► Show Figures

Figure 1

23 pages, 2770 KB

Open AccessArticle

Unsteady Lifting-Line Free-Wake Aerodynamic Modeling for Rotors in Hovering and Axial Flight

by Gregorio Frassoldati, Riccardo Giansante, Giovanni Bernardini and Massimo Gennaretti

Appl. Sci. 2025, 15(22), 12332; https://doi.org/10.3390/app152212332 - 20 Nov 2025

Cited by 1 | Viewed by 945

Abstract

A time-stepping, lifting-line, computationally efficient tool for preliminary design applications is developed to predict the unsteady aerodynamic loads of rotors operating in hovering and axial flight. The velocity field induced by wake vorticity is computed using a free-wake vortex-lattice model, while sectional aerodynamic [...] Read more.

A time-stepping, lifting-line, computationally efficient tool for preliminary design applications is developed to predict the unsteady aerodynamic loads of rotors operating in hovering and axial flight. The velocity field induced by wake vorticity is computed using a free-wake vortex-lattice model, while sectional aerodynamic loads are evaluated through the application of Küssner and Schwarz’s airfoil theory. The vorticity released by the trailing edge is related to the distribution of bound circulation and is convected downstream to form the vortex-lattice wake. The local bound circulation is determined by applying the Kutta–Joukowski theorem for unsteady flows. The proposed unsteady aerodynamic solver is successfully validated by comparison with both experimental data available in the literature and numerical results obtained by a three-dimensional boundary element method computational tool for potential flow. It does not apply to rotors in edgewise flight conditions and when compressibility effects are not negligible. Full article

(This article belongs to the Section Acoustics and Vibrations)

► Show Figures

Figure 1

24 pages, 8957 KB

Open AccessArticle

Utilizing VR Technology in Foundational Welding Skill Development

by Nuri Furkan Koçak, Ali Saygın and Fuat Türk

Appl. Sci. 2025, 15(22), 12331; https://doi.org/10.3390/app152212331 - 20 Nov 2025

Cited by 2 | Viewed by 1680

Abstract

Traditional approaches to welder training demand substantial investments in equipment, consumable materials, and workshop facilities, while also exposing novice learners to considerable safety risks. This study investigates the effectiveness of a virtual reality (VR)-based welding training system developed with Unity for the Meta [...] Read more.

Traditional approaches to welder training demand substantial investments in equipment, consumable materials, and workshop facilities, while also exposing novice learners to considerable safety risks. This study investigates the effectiveness of a virtual reality (VR)-based welding training system developed with Unity for the Meta Quest 2 platform, designed to deliver safe and immersive instruction in fundamental welding techniques. A total of twenty participants with no prior welding experience completed structured VR training sessions over two weeks. The program focused on developing competencies in welding machine operation (including start-up procedures and parameter adjustments), controlling shielding gas flow, and accurately regulating torch-to-workpiece distance, torch angle, and travel speed. Real-time feedback was integrated into the system to support accurate control and positioning of the welding torch. Quantitative assessments demonstrated significant improvements in both technical proficiency and trainee confidence and anxiety levels. Knowledge test scores increased from 45.3 to 85.1, while machine adjustment accuracy rose from 28.7 to 92.3. In parallel, participant confidence levels increased substantially, and anxiety scores decreased from 4.0–4.5 to 1.1–1.5 on standardized scales. These findings provide experimental evidence that VR-based training can enhance fundamental welding education by offering a safe, repeatable, and effective practice environment that simultaneously improves technical performance, strengthens learner confidence, and reduces training-related anxiety. Full article

(This article belongs to the Special Issue Recent Advances and Application of Virtual Reality)

► Show Figures

Figure 1

26 pages, 1409 KB

Open AccessArticle

Bow-Tie-Based Risk Assessment of Fishing Vessel Marine Accidents in the Open Sea Using IMO GISIS Data

by Seung-Hyun Lee, Su-Hyung Kim, Kyung-Jin Ryu, Soo-Yeon Kwon and Yoo-Won Lee

Appl. Sci. 2025, 15(22), 12330; https://doi.org/10.3390/app152212330 - 20 Nov 2025

Viewed by 932

Abstract

Open-sea fishing vessel accidents are difficult to assess systematically because no state holds exclusive jurisdiction, and reporting and investigative duties are not applied consistently. This study analyzed 67 officially reported accidents from the International Maritime Organization (IMO) Global Integrated Shipping Information System (GISIS) [...] Read more.

Open-sea fishing vessel accidents are difficult to assess systematically because no state holds exclusive jurisdiction, and reporting and investigative duties are not applied consistently. This study analyzed 67 officially reported accidents from the International Maritime Organization (IMO) Global Integrated Shipping Information System (GISIS) using a bow-tie framework combining fault tree analysis (FTA), Firth logistic regression, event tree analysis (ETA), and quantitative risk assessment (QRA). COLREG violations and watchkeeping failures dominated collisions; overload and stability issues caused capsizes; pump capacity, hull leakage, and vessel aging (≥30 years) caused sinkings. Firth regression confirmed older vessels and high beam-to-length ratios (≥0.30) significantly increased sinking likelihood. ETA and QRA estimated probabilities of 0.522 for collisions, 0.090 for capsizes, and 0.388 for sinkings, with risks of R = 0.155, 0.048, and 0.036. Because open-sea accident data rely on limited and voluntary reporting, results are preliminary. However, the bow-tie framework effectively identifies dominant causal factors and high-severity event pathways in open-sea fishing operations. Full article

(This article belongs to the Special Issue Risk and Safety of Maritime Transportation)

► Show Figures

Figure 1

20 pages, 822 KB

Open AccessArticle

Integrating Sentiment Analysis into Agile Feedback Loops for Continuous Improvement

by Diogo Marçal, José Metrôlho and Fernando Ribeiro

Appl. Sci. 2025, 15(22), 12329; https://doi.org/10.3390/app152212329 - 20 Nov 2025

Cited by 1 | Viewed by 1457

Abstract

The pursuit of continuous improvement is a defining feature of agile software development, yet its success depends on the systematic collection and interpretation of team members’ feedback. Conventional mechanisms, such as retrospectives and surveys, provide valuable insights but are often constrained by their [...] Read more.

The pursuit of continuous improvement is a defining feature of agile software development, yet its success depends on the systematic collection and interpretation of team members’ feedback. Conventional mechanisms, such as retrospectives and surveys, provide valuable insights but are often constrained by their episodic nature and susceptibility to subjective interpretation. This study examines the potential of Artificial Intelligence (AI), and in particular sentiment analysis, to complement feedback-driven practices and strengthen continuous improvement in agile contexts. Two literature reviews were conducted: one on applications of AI across software engineering domains and another focusing specifically on sentiment analysis in agile environments. Based on these insights, a prototype tool was developed to integrate sentiment analysis into task management workflows, enabling the structured collection and analysis of developers’ perceptions of task descriptions. Semi-structured interviews with experienced project managers confirmed the relevance of this approach, highlighting its capacity to improve task clarity and foster more transparent and inclusive feedback processes. Participants emphasized the value of the proposed approach in generating rapid, automated insights, while also identifying potential limitations related to response fatigue and the reliability of AI-generated outcomes. The findings suggest that incorporating sentiment analysis into agile practices is both feasible and advantageous, providing a pathway to align technical objectives with developer experiences while enhancing motivation, collaboration, and operational efficiency. Full article

(This article belongs to the Special Issue Intelligent Computing in Software Engineering)

► Show Figures

Figure 1

42 pages, 4993 KB

Open AccessArticle

Locating Causes of Inconsistency in a Variability Model for Software Product Line

by Younghun Han, Sungwon Kang and Jihyun Lee

Appl. Sci. 2025, 15(22), 12328; https://doi.org/10.3390/app152212328 - 20 Nov 2025

Viewed by 640

Abstract

One of the central activities of software product line development is variability modeling for a product family. Because variability models are needed at various stages of software product line development, determining whether a variability model has been modeled correctly is an essential activity [...] Read more.

One of the central activities of software product line development is variability modeling for a product family. Because variability models are needed at various stages of software product line development, determining whether a variability model has been modeled correctly is an essential activity for successful software product line development. Existing studies proposed various methods for analysis of various aspects of correctness of a variability model. In particular, analyzing whether a variability model is consistent or not is considered the most important analysis perspective since it is impossible to configure products from such a model. There are few studies in the software product line field that locate causes of inconsistency in a variability model. Furthermore, these existing methods cannot locate the exact causes of inconsistency due to the fact that the feature model they are based on allows ambiguity in its parent–child relationship or due to the fact that they are designed to produce explanations rather than locations of causes, resulting in producing long and complex explanations as the size of the feature model increases. In this work, we propose a method that determines whether or not a variability model has an inconsistency and identifies the exact locations of its causes if it has an inconsistency. To evaluate the proposed method, we developed a tool that automatically performs all the steps of the method and used it to conduct experiments with 49 models, including real-world variability models. As a result, the proposed method accurately identified all models with an inconsistency and located all causes of inconsistency in them. Full article

(This article belongs to the Section Computing and Artificial Intelligence)

► Show Figures

Figure 1

27 pages, 3487 KB

Open AccessArticle

Stability Assessment of Unilateral External Fixator Configurations for Open Tibial Fractures: An Experimental Study

by Elmedin Mešić, Nedim Pervan, Adil Muminović, Edvin Rahman and Bakir Muminović

Appl. Sci. 2025, 15(22), 12327; https://doi.org/10.3390/app152212327 - 20 Nov 2025

Viewed by 1280

Abstract

The primary objective of external fixation is to stabilize bone fractures, with the mechanical characteristics of the fixation system playing a critical role in shaping the biomechanical environment of the fracture and, consequently, the healing process. This study presents an experimental investigation of [...] Read more.

The primary objective of external fixation is to stabilize bone fractures, with the mechanical characteristics of the fixation system playing a critical role in shaping the biomechanical environment of the fracture and, consequently, the healing process. This study presents an experimental investigation of the stability of eight unilateral external fixation configurations applied to an open tibial fracture. The stiffness of each configuration was evaluated under axial compression, anterior–posterior (AP) bending, medial–lateral (ML) bending, and torsional loading. In addition, the effects of structural parameters—such as the number of half-pins, planarity of the configuration, and interfragmentary distance—on fixator stiffness and generated stresses were examined. The results revealed a linear relationship between applied load and both bone segment displacement and principal stresses. Biomechanical tests demonstrated that biplanar configurations provide sufficient stability for open tibial fractures, while simultaneously offering an optimal structural design for the fixation system. Moreover, the number of half-pins was identified as a statistically significant factor influencing configuration stiffness under axial loading and torsion, with biplanar configurations proving particularly effective in torsional scenarios. However, in AP and ML bending tests, neither configuration type nor any individual parameter produced statistically significant differences in bending stiffness. Interestingly, interfragmentary distance did not exert a statistically significant effect on configuration stiffness under any loading condition. Furthermore, neither configuration type nor the analyzed parameters had a notable influence on the principal stresses measured at the control points. Full article

► Show Figures

Figure 1

26 pages, 12950 KB

Open AccessArticle

Qualitative Assessment of Point Cloud from SLAM-Based MLS for Quarry Digital Twin Creation

by Ľudovít Kovanič, Patrik Peťovský, Branislav Topitzer, Peter Blišťan and Ondrej Tokarčík

Appl. Sci. 2025, 15(22), 12326; https://doi.org/10.3390/app152212326 - 20 Nov 2025

Cited by 2 | Viewed by 1265

Abstract

Quarries represent critical sites for raw material extraction, for which regular monitoring and mine surveying documentation, along with its updating, is essential to ensuring safety, environmental protection, and effective management of the mining process. This article aims to evaluate the modern approach to [...] Read more.

Quarries represent critical sites for raw material extraction, for which regular monitoring and mine surveying documentation, along with its updating, is essential to ensuring safety, environmental protection, and effective management of the mining process. This article aims to evaluate the modern approach to quarry surveying and the creation of a base mining map using advanced laser scanning methods, such as terrestrial laser scanning (TLS) and simultaneous localization and mapping (SLAM)-based mobile laser scanning (MLS). Particular attention is given to the analysis of noise generated using TLS and SLAM-based MLS methods. An analysis of mutual differences between point clouds is presented to compare the spatial accuracy of the point clouds obtained using MLS technology against those from the reference TLS method on both horizontally and vertically oriented test areas. To assess the quality and usability of data obtained using the TLS and MLS methods, a selected section of the mining wall was analyzed based on the distance between points (Cloud-to-Cloud analysis), cross-section analysis, and volume calculations based on 3D mesh models generated from stage edges and point clouds. The findings offer valuable insights into the effective use of each method in quarry surveying, contributing to the development of innovative approaches to spatial data collection as a base for creating Digital Twins of quarries. The article also evaluates the efficiency of both measurement approaches in terms of accuracy, measurement speed, and practical applicability in mining practices. The results show that the point cloud obtained by the TLS Leica RTC360 device, compared to that by the MLS method using the FARO Orbis device (FARO Technologies, Inc., Lakemary, FL, USA), achieves better values in terms of average noise level, standard deviation, interval of highest point density, and RMSD (Root Mean Square Deviation) in test areas. Our conclusions highlight the high potential of laser scanning for the modernization of mining documentation and the improvement of surveying processes in the smart mining industry, particularly for updating Digital Elevation Models (DEMs), Digital Terrain Models (DTMs), Digital Surface Models (DSMs), and other 3D models of quarries for the creation of their Digital Twins. Full article

(This article belongs to the Special Issue Surface and Underground Mining Technology and Sustainability)

► Show Figures

Figure 1

17 pages, 4006 KB

Open AccessArticle

The Study of Triangular Flow Regulators

by Marek Ochowiak, Magdalena Olszewska, Daniel Janecki, Sylwia Włodarczak, Andżelika Krupińska and Magdalena Matuszak

Appl. Sci. 2025, 15(22), 12325; https://doi.org/10.3390/app152212325 - 20 Nov 2025

Viewed by 537

Abstract

The paper presents the results of tests of flow regulators with a triangular prismatic cross-section chamber. The aim of the paper was to experimentally and numerically (CFD) assess the water flow for different design variants of the regulators. Three variants of flow regulators [...] Read more.

The paper presents the results of tests of flow regulators with a triangular prismatic cross-section chamber. The aim of the paper was to experimentally and numerically (CFD) assess the water flow for different design variants of the regulators. Three variants of flow regulators with a triangular prismatic cross-section chamber were analysed: without a barrier (FRWB), with a short barrier (FRSB), and with a long barrier (FRLB). Simultaneously, numerical simulations were carried out using the SST k-ω turbulence model (CFD). The obtained flow characteristics Q = f(H) showed the clear effect of the presence and length of the barrier on the efficiency of flow restriction. The regulators with a long barrier (FRLB) provided the highest flow damping and the highest stability of the system’s operation, which was confirmed by the experimental tests and CFD analyses. The regulators without a barrier (FRWB) were characterized by the highest liquid flow rate and the lowest damping efficiency. The use of a long barrier allowed for an increased control efficiency and improved predictability of the device’s operation. The triangular cross-section of the chamber favours the formation of a stable vortex flow and increases the efficiency of the regulator’s operation. Full article

(This article belongs to the Special Issue Advances in Computational and Experimental Fluid Dynamics)

► Show Figures

Figure 1

14 pages, 3307 KB

Open AccessArticle

A Novel Low-Illumination Image Enhancement Method Based on Convolutional Neural Network with Retinex Theory

by Haixia Mao, Wei Peng, Yan Tian and Xiaochun Zhu

Appl. Sci. 2025, 15(22), 12324; https://doi.org/10.3390/app152212324 - 20 Nov 2025

Viewed by 935

Abstract

Low-illumination images can seriously affect or even limit the performance of the human eye or a computer vision system, making image enhancement processing necessary. Traditional image enhancement methods, such as those based on image fusion, frequency thresholding, or spatial domain processing, lack robustness. [...] Read more.

Low-illumination images can seriously affect or even limit the performance of the human eye or a computer vision system, making image enhancement processing necessary. Traditional image enhancement methods, such as those based on image fusion, frequency thresholding, or spatial domain processing, lack robustness. Existing state-of-the-art methods, including GLADNet and MSR-Net, also suffer from color bias or the square effect in the recovered images. To address these issues, we propose a novel low-illumination image enhancement method based on a convolutional neural network (CNN) that incorporates the Retinex theory, namely Retinex-CNN. A decomposition sub-network is designed to transform the original image into a reflectance map and a light map, and then they are further optimized by a reflectance map refinement sub-network and a light map enhancement sub-network, respectively. Finally, according to Retinex theory, the refined reflectance map and the enhanced light map are synthesized to obtain the fusion result with better visual sense. We used synthetic and real low-illumination image datasets for training, testing, and comparison with other methods. In the synthetic scene, Retinex-CNN demonstrates superior performance with higher PSNR, MSE, and SSIM. In the real scene, Retinex-CNN has the best VIF score in six public datasets (over 0.75) and the best NIQE score in four of them. The experiments demonstrate that Retinex-CNN can not only effectively improve the brightness of the image but also enhance the clarity of the details and mitigate the serious color distortion and halo phenomenon. Additionally, the image enhancement process is less time-consuming. Full article

► Show Figures

Figure 1

20 pages, 15584 KB

Open AccessArticle

Physics-Informed Weighting Multi-Scale Deep Learning Inversion for Deep-Seated Fault Feature Identification: A Case Study of Aeromagnetic Data in the Dandong Region

by Haihua Ju, Zhong Xia, Jie Yang, Longran Zhou, Bo Dai, Jian Jiao, Duo Wang and Runqi Wang

Appl. Sci. 2025, 15(22), 12323; https://doi.org/10.3390/app152212323 - 20 Nov 2025

Viewed by 934

Abstract

Magnetic inversion through three-dimensional (3D) susceptibility reconstruction can effectively identify the deep extension characteristics and structural variations in faults. Therefore, the reliability of inversion results from magnetic anomaly data is a key issue that must be addressed in fault detection and quantitative evaluation [...] Read more.

Magnetic inversion through three-dimensional (3D) susceptibility reconstruction can effectively identify the deep extension characteristics and structural variations in faults. Therefore, the reliability of inversion results from magnetic anomaly data is a key issue that must be addressed in fault detection and quantitative evaluation of fault activity. In recent years, deep neural network-driven magnetic data inversion methods have rapidly become a research focus in the field of geophysical magnetic data inversion. However, existing methods primarily rely on convolutional neural networks (CNNs), whose inherent local feature extraction capabilities limit their ability to model the spatial continuity of large-scale subsurface magnetic structures. Moreover, the general lack of prior physical constraints in these network models often leads to unreliable inversion results. To address these limitations, this paper proposes a physics-informed multi-scale deep learning inversion method for magnetic anomaly data. The method designs a dual-stream Transformer-CNN fusion module (TCFM). It leverages the self-attention mechanism in Transformers to model global susceptibility correlations while efficiently capturing local geological features through CNN convolutional operations. This enables collaborative modeling of multi-scale subsurface magnetic structures, significantly enhancing inversion accuracy. Furthermore, by incorporating deep physical priors, we design a depth-aware weighted loss function. By strengthening optimization constraints in deep regions, it effectively improves the vertical resolution of inversion models for deep magnetic structures. Comparative experiments with U-Net++ and Transformer demonstrate that the proposed method achieves smaller errors and higher inversion accuracy. Applied to measured aeromagnetic data from the Dandong region of China, the method yields reliable inversion results. Variations in magnetic susceptibility within these results successfully delineate the spatial distribution of fault zones, providing a geophysical basis for regional seismic hazard monitoring and assessment. Full article

(This article belongs to the Section Earth Sciences)

► Show Figures

Figure 1

16 pages, 1863 KB

Open AccessArticle

Superpoint Network-Based Video Stabilization Technology for Mine Rescue Robots

by Shuqi Wang, Zhaowenbo Zhu and Yikai Jiang

Appl. Sci. 2025, 15(22), 12322; https://doi.org/10.3390/app152212322 - 20 Nov 2025

Viewed by 659

Abstract

Mine rescue robots operate in extremely adverse subterranean environments, where the acquired video data are frequently affected by severe jitter and motion distortion. Such instability leads to the loss of critical visual information, thereby reducing the reliability of rescue decision-making. To address this [...] Read more.

Mine rescue robots operate in extremely adverse subterranean environments, where the acquired video data are frequently affected by severe jitter and motion distortion. Such instability leads to the loss of critical visual information, thereby reducing the reliability of rescue decision-making. To address this issue, a dual-channel visual stabilization framework based on the SuperPoint network is proposed, extending the traditional ORB descriptor framework. Here, dual-channel refers to two configurable and mutually exclusive feature extraction paths—an ORB-based path and a SuperPoint-based path—that can be flexibly switched according to scene conditions and computational requirements, rather than operating simultaneously on the same frame. The subsequent stabilization pipeline remains unified and consistent across both modes. The method employs an optimized detector head that integrates deep feature extraction, non-maximum suppression, and boundary filtering to enable precise estimation of inter-frame motion. When combined with smoothing filters, the approach effectively attenuates vibrations induced by irregular terrain and dynamic operational conditions. Experimental evaluations conducted across diverse scenarios demonstrate that the proposed algorithm achieves an average improvement of 27.91% in Peak Signal-to-Noise Ratio (PSNR), a 55.04% reduction in Mean Squared Error (MSE), and more than a twofold increase in the Structural Similarity Index (SSIM) relative to pre-stabilized sequences. Moreover, runtime analysis indicates that the algorithm can operate in near-real-time, supporting its practical deployment on embedded mine rescue robot platforms.These results verify the algorithm’s robustness and applicability in environments requiring high visual stability and image fidelity, providing a reliable foundation for enhanced visual perception and autonomous decision-making in complex disaster scenarios. Full article

► Show Figures

Figure 1

29 pages, 4498 KB

Open AccessArticle

The Effect of Data Augmentation on Performance of Custom and Pre-Trained CNN Models for Crack Detection

by Tope Moses Omoniyi, Barnabas Abel, Oluwaseun Omoebamije, Zuberu Mark Onimisi, Jose C. Matos, Joaquim Tinoco and Tran Quang Minh

Appl. Sci. 2025, 15(22), 12321; https://doi.org/10.3390/app152212321 - 20 Nov 2025

Viewed by 2177

Abstract

Data augmentation is one of the effective solutions to improve the performance of machine learning models in general and deep learning in particular. Data augmentation techniques bring different effects to each model, but very few studies have considered this issue. This study investigated [...] Read more.

Data augmentation is one of the effective solutions to improve the performance of machine learning models in general and deep learning in particular. Data augmentation techniques bring different effects to each model, but very few studies have considered this issue. This study investigated the effect of five distinct data augmentation strategies on a custom-built Convolutional Neural Network (CNN) and nine pre-trained CNN models for crack detection. All ten models were initially trained on a reference dataset of unaugmented images, followed by separate experiments using the augmented datasets. The results show that the pre-trained models, especially VGG-16, EfficientNet-B7, Xception, DenseNet-201, and EfficientNet-B0, consistently achieved greater than 98% in accuracy across all augmentation techniques. Meanwhile, the custom-built CNN was very sensitive to illumination changes and noise. Image rotation and cropping have minimal negative impact and sometimes improve performance. The findings demonstrate that combining data augmentation with state-of-the-art pre-trained models offers a powerful and efficient alternative to the reliance on large-scale datasets for accurate crack detection using CNNs. Full article

(This article belongs to the Section Civil Engineering)

► Show Figures

Figure 1

17 pages, 1707 KB

Open AccessArticle

Investigation of the Wheel Power and State of Charge of Plug-In Hybrid Electric Vehicles (PHEVs) on a Chassis Dynamometer in Various Driving Test Cycles

by Wasan Palasai, Pongskorn Tepsorn, Taweesak Katthiyawan, Prathan Srichai and Isara Chaopisit

Appl. Sci. 2025, 15(22), 12320; https://doi.org/10.3390/app152212320 - 20 Nov 2025

Viewed by 990

Abstract

The purpose of this study is to monitor the battery performance of plug-in hybrid electric vehicles (PHEVs) on a chassis dynamometer using the US06, NEDC, and EPA highway driving cycles. The chassis dynamometer simulates vehicle operation and driving conditions and allows for precise [...] Read more.

The purpose of this study is to monitor the battery performance of plug-in hybrid electric vehicles (PHEVs) on a chassis dynamometer using the US06, NEDC, and EPA highway driving cycles. The chassis dynamometer simulates vehicle operation and driving conditions and allows for precise simulation of pre-defined driving cycles, including simulations of acceleration, deceleration, stopping, and re-acceleration on the road. In the case of the US06 driving cycle, the results for (EV mode) compared with energy consumption during electric testing revealed a consistent decrease in the SOC (state of charge) due to the rapid response of the electric motor distribution to the changing power, as well as electric power fluctuations during driving conditions. Under the NEDC, the test results for electric power (EV) compared with energy consumption during electric testing revealed that the SOC gradually decreased at the start of the test due to low driving speeds. Towards the end, at around 800 s, an increase in driving speed resulted in a noticeable drop in SOC. The electric power varied during the driving cycle in this test due to the motor’s rapid response to changes in power distribution while driving. For the EPA Highway driving cycle test, the test results for electric power (EV) compared with energy consumption during continuous electric testing indicated a gradual decrease in the SOC at first due to low driving speeds. As the driving speed increased after about 300 s, the SOC rapidly decreased. Because of the motor’s quick response to changes in the power distribution while driving, the electric power varied according to the driving cycle. Full article

► Show Figures

Figure 1

21 pages, 894 KB

Open AccessArticle

Development of Wheat Bread Fortified with Sea Buckthorn (Hippophae rhamnoides L.) Pomace: Nutritional Enhancement, Technological Properties, and Consumer Acceptance

by Anna Wirkijowska, Piotr Zarzycki and Konrad Terpiłowski

Appl. Sci. 2025, 15(22), 12319; https://doi.org/10.3390/app152212319 - 20 Nov 2025

Viewed by 875

Abstract

Sea buckthorn (Hippophae rhamnoides L.) is a rich source of bioactive compounds, including vitamin C, polyphenols, carotenoids, and dietary fiber. In this study, sea buckthorn pomace, an underutilized by-product of oil processing, was incorporated into wheat bread at levels of 0, 3, [...] Read more.

Sea buckthorn (Hippophae rhamnoides L.) is a rich source of bioactive compounds, including vitamin C, polyphenols, carotenoids, and dietary fiber. In this study, sea buckthorn pomace, an underutilized by-product of oil processing, was incorporated into wheat bread at levels of 0, 3, 6, 9, and 12% (based on flour weight). The technological performance (dough yield, baking loss, loaf volume, texture, and color), nutritional composition (protein, fat, dietary fiber fractions, mineral content, and caloric value), and sensory attributes of the resulting breads were comprehensively evaluated. Pomace addition markedly increased the protein content of bread (from 13.5% to 16.8%) and more than doubled total dietary fiber (from 5.4% to 11.6%), while reducing caloric value by approximately 5.6%. Increasing pomace levels also affected dough behavior and bread structure: water absorption rose from 59.9% to 68.9%, specific loaf volume decreased by 11–28%, and crumb hardness increased from 3.8 N (control) to 12.4 N (12% addition). Sensory acceptability remained high up to 6% pomace incorporation (acceptability index > 90%), whereas breads containing 9–12% received significantly lower scores, mainly due to darker crumb color and intensified sour or bitter notes. Overall, sea buckthorn pomace can be effectively used as a nutritionally enriching, value-added ingredient in wheat bread, enhancing fiber and protein content while maintaining desirable technological and sensory properties at moderate substitution levels. Full article

(This article belongs to the Special Issue Functional Foods: Product Development, Technological Trends and Safety)

► Show Figures

Figure 1

17 pages, 2644 KB

Open AccessArticle

Numerical Simulation of Clay Layer Permeability Failure Under Loose Strata: Effects of Mining-Induced Fracture Width

by Yuan Hang, Jinwei Li, Shichong Yuan, Dengkui Zhang and Chuanyong Wei

Appl. Sci. 2025, 15(22), 12318; https://doi.org/10.3390/app152212318 - 20 Nov 2025

Viewed by 533

Abstract

Based on the problem of water and sand inrush caused by the infiltration and failure of the clay layer at the bottom of the loose layer in shallow coal seam mining in eastern China, this study adopts the Particle Flow Code numerical simulation [...] Read more.

Based on the problem of water and sand inrush caused by the infiltration and failure of the clay layer at the bottom of the loose layer in shallow coal seam mining in eastern China, this study adopts the Particle Flow Code numerical simulation method to conduct multi-physics field coupling analysis. Based on the geological conditions of Taiping Coal Mine in Shandong Province, a two-dimensional water sand clay coupling model was constructed to systematically simulate the entire process of permeability failure of clay layers under different mining crack widths (5–20 mm). The permeability failure mechanism was revealed through porosity distribution, particle contact number, and contact force evolution laws. The numerical simulation results show that with the increase in crack width, the speed of contact reduction is faster, the speed of water and inrush is faster, and the time is shorter. The process of infiltration failure can be divided into two stages: the first stage is the clay infiltration deformation stage, and the second stage is the water inrush and sand collapse stage. In addition, the larger the width of the crack, the greater the contact force, and the shorter the time of infiltration failure and water and sand bursting experienced. The quantitative relationship between the width of mining induced cracks and permeability failure was revealed, and a critical discrimination index for permeability failure in clay layers was established, providing theoretical support for optimizing safe mining parameters and preventing water and sand inrush disasters in porous aquifers. Full article

(This article belongs to the Special Issue Hydrogeology and Regional Groundwater Flow)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Appl. Sci., Volume 15, Issue 22 (November-2 2025) – 498 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI