Technologies

24 pages, 11329 KiB

Open AccessArticle

An End-to-End Lightweight Multi-Scale CNN for the Classification of Lung and Colon Cancer with XAI Integration

by Mohammad Asif Hasan, Fariha Haque, Saifur Rahman Sabuj, Hasan Sarker, Md. Omaer Faruq Goni, Fahmida Rahman and Md Mamunur Rashid

Technologies 2024, 12(4), 56; https://doi.org/10.3390/technologies12040056 - 21 Apr 2024

Cited by 2 | Viewed by 2912

Abstract

To effectively treat lung and colon cancer and save lives, early and accurate identification is essential. Conventional diagnosis takes a long time and requires the manual expertise of radiologists. The rising number of new cancer cases makes it challenging to process massive volumes [...] Read more.

To effectively treat lung and colon cancer and save lives, early and accurate identification is essential. Conventional diagnosis takes a long time and requires the manual expertise of radiologists. The rising number of new cancer cases makes it challenging to process massive volumes of data quickly. Different machine learning approaches to the classification and detection of lung and colon cancer have been proposed by multiple research studies. However, when it comes to self-learning classification and detection tasks, deep learning (DL) excels. This paper suggests a novel DL convolutional neural network (CNN) model for detecting lung and colon cancer. The proposed model is lightweight and multi-scale since it uses only 1.1 million parameters, making it appropriate for real-time applications as it provides an end-to-end solution. By incorporating features extracted at multiple scales, the model can effectively capture both local and global patterns within the input data. The explainability tools such as gradient-weighted class activation mapping and Shapley additive explanation can identify potential problems by highlighting the specific input data areas that have an impact on the model’s choice. The experimental findings demonstrate that for lung and colon cancer detection, the proposed model was outperformed by the competition and accuracy rates of 99.20% have been achieved for multi-class (containing five classes) predictions. Full article

(This article belongs to the Special Issue The Future of Healthcare: Biomedical Technology and Integrated Artificial Intelligence)

► Show Figures

Figure 1

25 pages, 4707 KiB

Open AccessArticle

Digital Twin Models for Personalised and Predictive Medicine in Ophthalmology

by Miruna-Elena Iliuţă, Mihnea-Alexandru Moisescu, Simona-Iuliana Caramihai, Alexandra Cernian, Eugen Pop, Daniel-Ioan Chiş and Traian-Costin Mitulescu

Technologies 2024, 12(4), 55; https://doi.org/10.3390/technologies12040055 - 18 Apr 2024

Cited by 3 | Viewed by 2654

Abstract

This article explores the integration of Digital Twins in Systems and Predictive Medicine, focusing on eye diagnosis. By utilizing the Digital Twin models, the proposed framework can support early diagnosis and predict evolution after treatment by providing customized simulation scenarios. Furthermore, a structured [...] Read more.

This article explores the integration of Digital Twins in Systems and Predictive Medicine, focusing on eye diagnosis. By utilizing the Digital Twin models, the proposed framework can support early diagnosis and predict evolution after treatment by providing customized simulation scenarios. Furthermore, a structured architectural framework comprising five levels has been proposed, integrating Digital Twin, Systems Medicine, and Predictive Medicine for managing eye diseases. Based on demographic parameters, statistics were performed to identify potential correlations that may contribute to predispositions to glaucoma. With the aid of a dataset, a neural network was trained with the goal of identifying glaucoma. This comprehensive approach, based on statistical analysis and Machine Learning, is a promising method to enhance diagnostic accuracy and provide personalized treatment approaches. Full article

► Show Figures

Figure 1

22 pages, 3814 KiB

Open AccessArticle

Experimental and Numerical Analysis of a Novel Cycloid-Type Rotor versus S-Type Rotor for Vertical-Axis Wind Turbine

by José Eli Eduardo González-Durán, Juan Manuel Olivares-Ramírez, María Angélica Luján-Vega, Juan Emigdio Soto-Osornio, Juan Manuel García-Guendulain and Juvenal Rodriguez-Resendiz

Technologies 2024, 12(4), 54; https://doi.org/10.3390/technologies12040054 - 17 Apr 2024

Viewed by 2053

Abstract

The performance of a new vertical-axis wind turbine rotor based on the mathematical equation of the cycloid is analyzed and compared through simulation and experimental testing against a semicircular or S-type rotor, which is widely used. The study examines three cases: equalizing the [...] Read more.

The performance of a new vertical-axis wind turbine rotor based on the mathematical equation of the cycloid is analyzed and compared through simulation and experimental testing against a semicircular or S-type rotor, which is widely used. The study examines three cases: equalizing the diameter, chord length and the area under the curve. Computational Fluid Dynamics (CFD) was used to simulate these cases and evaluate moment, angular velocity and power. Experimental validation was carried out in a wind tunnel that was designed and optimized with the support of CFD. The rotors for all three cases were 3D printed in resin to analyze their experimental performance as a function of wind speed. The moment and Maximum Power Point (MPP) were determined in each case. The simulation results indicate that the cycloid-type rotor outperforms the semicircular or S-type rotor by 15%. Additionally, experimental evidence confirms that the cycloid-type rotor performs better in all three cases. In the MPP analysis, the cycloid-type rotor achieved an efficiency of 10.8% which was 38% better than the S-type rotor. Full article

(This article belongs to the Section Environmental Technology)

► Show Figures

Figure 1

20 pages, 1541 KiB

Open AccessArticle

Developing a Performance Evaluation Framework Structural Model for Educational Metaverse

by Elena Tsappi, Ioannis Deliyannis and George Nathaniel Papageorgiou

Technologies 2024, 12(4), 53; https://doi.org/10.3390/technologies12040053 - 16 Apr 2024

Cited by 1 | Viewed by 2021

Abstract

In response to the transformative impact of digital technology on education, this study introduces a novel performance management framework for virtual learning environments suitable for the metaverse era. Based on the Structural Equation Modeling (SEM) approach, this paper proposes a comprehensive evaluative model, [...] Read more.

In response to the transformative impact of digital technology on education, this study introduces a novel performance management framework for virtual learning environments suitable for the metaverse era. Based on the Structural Equation Modeling (SEM) approach, this paper proposes a comprehensive evaluative model, anchored on the integration of the Theory of Planned Behavior (TPB), the Unified Theory of Acceptance and Use of Technology (UTAUT), and the Community of Inquiry Framework (CoI). The model synthesizes five Key Performance Indicators (KPIs)—content delivery, student engagement, metaverse tool utilization, student performance, and adaptability—to intricately assess academic avatar performances in virtual educational settings. This theoretical approach marks a significant stride in understanding and enhancing avatar efficacy in the metaverse environment. It enriches the discourse on performance management in digital education and sets a foundation for future empirical studies. As virtual online environments gain prominence in education and training, this research study establishes the basic principles and highlights the key points for further empirical research in the new era of the metaverse educational environment. Full article

(This article belongs to the Topic Theoretical and Applied Problems in Human-Computer Intelligent Systems)

► Show Figures

Figure 1

21 pages, 5536 KiB

Open AccessArticle

A Comparison of Machine Learning-Based and Conventional Technologies for Video Compression

by Lesia Mochurad

Technologies 2024, 12(4), 52; https://doi.org/10.3390/technologies12040052 - 15 Apr 2024

Cited by 1 | Viewed by 2652

Abstract

The growing demand for high-quality video transmission over bandwidth-constrained networks and the increasing availability of video content have led to the need for efficient storage and distribution of large video files. To improve the latter, this article offers a comparison of six video [...] Read more.

The growing demand for high-quality video transmission over bandwidth-constrained networks and the increasing availability of video content have led to the need for efficient storage and distribution of large video files. To improve the latter, this article offers a comparison of six video compression methods without loss of quality. Particularly, H.255, VP9, AV1, convolutional neural network (CNN), recurrent neural network (RNN), and deep autoencoder (DAE). The proposed decision is to use a dataset of high-quality videos to implement and compare the performance of classical compression algorithms and algorithms based on machine learning. Evaluations of the compression efficiency and the quality of the received images were made on the basis of two metrics: PSNR and SSIM. This comparison revealed the strengths and weaknesses of each approach and provided insights into how machine learning algorithms can be optimized in future research. In general, it contributed to the development of more efficient and effective video compression algorithms that can be useful for a wide range of applications. Full article

► Show Figures

Figure 1

14 pages, 6188 KiB

Open AccessArticle

Monitoring of Hip Joint Forces and Physical Activity after Total Hip Replacement by an Integrated Piezoelectric Element

by Franziska Geiger, Henning Bathel, Sascha Spors, Rainer Bader and Daniel Kluess

Technologies 2024, 12(4), 51; https://doi.org/10.3390/technologies12040051 - 9 Apr 2024

Viewed by 2128

Abstract

Resultant hip joint forces can currently only be recorded in situ in a laboratory setting using instrumented total hip replacements (THRs) equipped with strain gauges. However, permanent recording is important for monitoring the structural condition of the implant, for therapeutic purposes, for self-reflection, [...] Read more.

Resultant hip joint forces can currently only be recorded in situ in a laboratory setting using instrumented total hip replacements (THRs) equipped with strain gauges. However, permanent recording is important for monitoring the structural condition of the implant, for therapeutic purposes, for self-reflection, and for research into managing the predicted increasing number of THRs worldwide. Therefore, this study aims to investigate whether a recently proposed THR with an integrated piezoelectric element represents a new possibility for the permanent recording of hip joint forces and the physical activities of the patient. Hip joint forces from nine different daily activities were obtained from the OrthoLoad database and applied to a total hip stem equipped with a piezoelectric element using a uniaxial testing machine. The forces acting on the piezoelectric element were calculated from the generated voltages. The correlation between the calculated forces on the piezoelectric element and the applied forces was investigated, and the regression equations were determined. In addition, the voltage outputs were used to predict the activity with a random forest classifier. The coefficient of determination between the applied maximum forces on the implant and the calculated maximum forces on the piezoelectric element was R² = 0.97 (p < 0.01). The maximum forces on the THR could be determined via activity-independent determinations with a deviation of 2.49 ± 13.16% and activity-dependent calculation with 0.87 ± 7.28% deviation. The activities could be correctly predicted using the classification model with 95% accuracy. Hence, piezoelectric elements integrated into a total hip stem represent a promising sensor option for the energy-autonomous detection of joint forces and physical activities. Full article

► Show Figures

Figure 1

19 pages, 1747 KiB

Open AccessReview

Past, Present, and Future of New Applications in Utilization of Eddy Currents

by Nestor O. Romero-Arismendi, Juan C. Olivares-Galvan, Jose L. Hernandez-Avila, Rafael Escarela-Perez, Victor M. Jimenez-Mondragon and Felipe Gonzalez-Montañez

Technologies 2024, 12(4), 50; https://doi.org/10.3390/technologies12040050 - 9 Apr 2024

Cited by 3 | Viewed by 3924

Abstract

Eddy currents are an electromagnetic phenomenon that represent an inexhaustible source of inspiration for technological innovations in the 21st century. Throughout history, these currents have been a subject of research and technological development in multiple fields. This article delves into the fascinating world [...] Read more.

Eddy currents are an electromagnetic phenomenon that represent an inexhaustible source of inspiration for technological innovations in the 21st century. Throughout history, these currents have been a subject of research and technological development in multiple fields. This article delves into the fascinating world of eddy currents, revealing their physical foundations and highlighting their impact on a wide range of applications, ranging from non-destructive evaluation of materials to levitation phenomena, as well as their influence on fields as diverse as medicine, the automotive industry, and aerospace. The nature of eddy currents has stimulated the imaginations of scientists and engineers, driving the creation of revolutionary technologies that are transforming our society. As we progress through this article, we will cover the main aspects of eddy currents, their practical applications, and challenges for future works. Full article

(This article belongs to the Collection Electrical Technologies)

► Show Figures

Figure 1

27 pages, 22078 KiB

Open AccessArticle

Numerical Study of the Influence of the Structural Parameters on the Stress Dissipation of 3D Orthogonal Woven Composites under Low-Velocity Impact

by Wang Xu, Mohammed Zikry and Abdel-Fattah M. Seyam

Technologies 2024, 12(4), 49; https://doi.org/10.3390/technologies12040049 - 5 Apr 2024

Cited by 1 | Viewed by 1767

Abstract

This study investigates the effects of the number of layers, x-yarn (weft) density, and z-yarn (binder) path on the mechanical behavior of E-glass 3D orthogonal woven (3DOW) composites during low-velocity impacts. Meso-level finite element (FE) models were developed and validated for 3DOW composites [...] Read more.

This study investigates the effects of the number of layers, x-yarn (weft) density, and z-yarn (binder) path on the mechanical behavior of E-glass 3D orthogonal woven (3DOW) composites during low-velocity impacts. Meso-level finite element (FE) models were developed and validated for 3DOW composites with different yarn densities and z-yarn paths, providing analyses of stress distribution within reinforcement fibers and matrix, energy absorption, and failure time. Our findings revealed that lower x-yarn densities led to accumulations of stress concentrations. Furthermore, changing the z-yarn path, such as transitioning from plain weaves to twill or basket weaves had a noticeable impact on stress distributions. The research highlights the significance of designing more resilient 3DOW composites for impact applications by choosing appropriate parameters in weaving composite designs. Full article

(This article belongs to the Section Innovations in Materials Processing)

► Show Figures

Figure 1

15 pages, 9991 KiB

Open AccessArticle

Carbon Fiber Polymer Reinforced 3D Printed Composites for Centrifugal Pump Impeller Manufacturing

by Gabriel Mansour, Vasileios Papageorgiou and Dimitrios Tzetzis

Technologies 2024, 12(4), 48; https://doi.org/10.3390/technologies12040048 - 3 Apr 2024

Viewed by 1847

Abstract

Centrifugal pumps are used extensively in various everyday applications. The occurrence of corrosion phenomena during operation often leads to the failure of a pump’s operating components, such as the impeller. The present research study examines the utilization of composite materials for fabricating centrifugal [...] Read more.

Centrifugal pumps are used extensively in various everyday applications. The occurrence of corrosion phenomena during operation often leads to the failure of a pump’s operating components, such as the impeller. The present research study examines the utilization of composite materials for fabricating centrifugal pump components using additive manufacturing as an effort to fabricate corrosion resistant parts. To achieve the latter two nanocomposite materials, carbon fiber reinforced polyamide and carbon fiber reinforced polyphenylene sulfide were compared with two metal alloys, cast iron and brass, which are currently used in pump impeller manufacturing. The mechanical properties of the materials are extracted by performing a series of experiments, such as uniaxial tensile tests, nanoindentation and scanning electron microscope (SEM) examination of the specimen’s fracture area. Then, computational fluid dynamics (CFD) analysis is performed using various impeller designs to determine the fluid pressure exerted on the impeller’s geometry during its operation. Finally, the maximum power rating of an impeller that can be made from such composites will be determined using a static finite element model (FEM). The FEM static model is developed by integrating the data collected from the experiments with the results obtained from the CFD analysis. The current research work shows that nanocomposites can potentially be used for developing impellers with rated power of up to 9.41 kW. Full article

(This article belongs to the Section Manufacturing Technology)

► Show Figures

Figure 1

16 pages, 371 KiB

Open AccessArticle

Impact Localization for Haptic Input Devices Using Hybrid Laminates with Sensoric Function

by René Schmidt, Alexander Graf, Ricardo Decker, Stephan Lede, Verena Kräusel, Lothar Kroll and Wolfram Hardt

Technologies 2024, 12(4), 47; https://doi.org/10.3390/technologies12040047 - 1 Apr 2024

Viewed by 3517

Abstract

The required energy savings can be achieved in all automotive domains through weight savings and the merging of manufacturing processes in production. This fact is taken into account through functional integration in lightweight materials and manufacturing in a process close to large-scale production. [...] Read more.

The required energy savings can be achieved in all automotive domains through weight savings and the merging of manufacturing processes in production. This fact is taken into account through functional integration in lightweight materials and manufacturing in a process close to large-scale production. In previous work, separate steps of a process chain for manufacturing a center console cover utilizing a sensoric hybrid laminate have been developed and evaluated. This includes the process steps of joining, forming and inline polarization as well as connecting to an embedded system. This work continues the research process by evaluating impact localization methods to use the center console as a haptic input device. For this purpose, different deep learning methods are derived from the state of the art and analyzed for their applicability in two consecutive studies. The results show that MLPs, LSTMs, GRUs and CNNs are suitable to localize impacts on the novel laminate with high localization rates of up to 99%, and thus the usability of the developed laminate as a haptic input device has been proven. Full article

(This article belongs to the Section Innovations in Materials Processing)

► Show Figures

Figure 1

16 pages, 2761 KiB

Open AccessArticle

Enhancing Patient Care in Radiotherapy: Proof-of-Concept of a Monitoring Tool

by Guillaume Beldjoudi, Rémi Eugène, Vincent Grégoire and Ronan Tanguy

Technologies 2024, 12(4), 46; https://doi.org/10.3390/technologies12040046 - 29 Mar 2024

Viewed by 1576

Abstract

Introduction: A monitoring tool, named Oncology Data Management (ODM), was developed in radiotherapy to generate structured information based on data contained in an Oncology Information System (OIS). This study presents the proof-of-concept of the ODM tool and highlights its applications to enhance patient [...] Read more.

Introduction: A monitoring tool, named Oncology Data Management (ODM), was developed in radiotherapy to generate structured information based on data contained in an Oncology Information System (OIS). This study presents the proof-of-concept of the ODM tool and highlights its applications to enhance patient care in radiotherapy. Material & Methods: ODM is a sophisticated SQL query which extracts specific features from the Mosaiq OIS (Elekta, UK) database into an independent structured database. Data from 2016 to 2022 was extracted to enable monitoring of treatment units and evaluation of the quality of patient care. Results: A total of 25,259 treatments were extracted. Treatment machine monitoring revealed a daily 11-treatement difference between two units. ODM showed that the unit with fewer daily treatments performed more complex treatments on diverse locations. In 2019, the implementation of ODM led to the definition of quality indicators and in organizational changes that improved the quality of care. As consequences, for palliative treatments, there was an improvement in the proportion of treatments prepared within 7 calendar days between the scanner and the first treatment session (29.1% before 2020, 40.4% in 2020 and 46.4% after 2020). The study of fractionation in breast treatments exhibited decreased prescription variability after 2019, with distinct patient age categories. Bi-fractionation once a week for larynx prescriptions of 35 × 2.0 Gy achieved an overall treatment duration of 47.0 ± 3.0 calendar days in 2022. Conclusions: ODM enables data extraction from the OIS and provides quantitative tools for improving organization of a department and the quality of patient care in radiotherapy. Full article

(This article belongs to the Topic Smart Healthcare: Technologies and Applications)

► Show Figures

Figure 1

21 pages, 799 KiB

Open AccessArticle

An Artificial Bee Colony Algorithm for Coordinated Scheduling of Production Jobs and Flexible Maintenance in Permutation Flowshops

by Asma Ladj, Fatima Benbouzid-Si Tayeb, Alaeddine Dahamni and Mohamed Benbouzid

Technologies 2024, 12(4), 45; https://doi.org/10.3390/technologies12040045 - 25 Mar 2024

Viewed by 1968

Abstract

This research work addresses the integrated scheduling of jobs and flexible (non-systematic) maintenance interventions in permutation flowshop production systems. We propose a coordinated model in which the time intervals between successive maintenance tasks as well as their number are assumed to be non-fixed [...] Read more.

This research work addresses the integrated scheduling of jobs and flexible (non-systematic) maintenance interventions in permutation flowshop production systems. We propose a coordinated model in which the time intervals between successive maintenance tasks as well as their number are assumed to be non-fixed for each machine on the shopfloor. With such a flexible nature of maintenance activities, the resulting joint schedule is more practical and representative of real-world scenarios. Our goal is to determine the best job permutation in which flexible maintenance activities are properly incorporated. To tackle the NP-hard nature of this problem, an artificial bee colony (ABC) algorithm is developed to minimize the total production time (Makespan). Experiments are conducted utilizing well-known Taillard’s benchmarks, enriched with maintenance data, to compare the proposed algorithm performance against the variable neighbourhood search (VNS) method from the literature. Computational results demonstrate the effectiveness of the proposed algorithm in terms of both solution quality and computational times. Full article

► Show Figures

Figure 1

18 pages, 2472 KiB

Open AccessArticle

Blood Pressure Measurement Device Accuracy Evaluation: Statistical Considerations with an Implementation in R

by Tanvi Chandel, Victor Miranda, Andrew Lowe and Tet Chuan Lee

Technologies 2024, 12(4), 44; https://doi.org/10.3390/technologies12040044 - 25 Mar 2024

Viewed by 1830

Abstract

Inaccuracies from devices for non-invasive blood pressure measurements have been well reported with clinical consequences. International standards, such as ISO 81060-2 and the seminal AAMI/ANSI SP10, define protocols and acceptance criteria for these devices. Prior to applying these standards, a sample size of [...] Read more.

Inaccuracies from devices for non-invasive blood pressure measurements have been well reported with clinical consequences. International standards, such as ISO 81060-2 and the seminal AAMI/ANSI SP10, define protocols and acceptance criteria for these devices. Prior to applying these standards, a sample size of N >= 85 is mandatory, that is, the number of distinct subcjects used to calculate device inaccuracies. Often, it is not possible to gather such a large sample. Many studies apply these standards with a smaller sample. The objective of the paper is to introduce a methodology that broadens the method first developed by the AAMI Sphygmomanometer Committee for accepting a blood pressure measurement device. We study changes in the acceptance region for various sample sizes using the sampling distribution for proportions and introduce a methodology for estimating the exact probability of the acceptance of a device. This enables the comparison of the accuracies of existing device development techniques even if they were studied with a smaller sample size. The study is useful in assisting BP measurement device manufacturers. To assist clinicians, we present a newly developed “bpAcc” package in R to evaluate acceptance statistics for various sample sizes. Full article

(This article belongs to the Special Issue The Future of Healthcare: Biomedical Technology and Integrated Artificial Intelligence)

► Show Figures

Figure 1

31 pages, 1533 KiB

Open AccessReview

Applied Deep Learning-Based Crop Yield Prediction: A Systematic Analysis of Current Developments and Potential Challenges

by Khadija Meghraoui, Imane Sebari, Juergen Pilz, Kenza Ait El Kadi and Saloua Bensiali

Technologies 2024, 12(4), 43; https://doi.org/10.3390/technologies12040043 - 24 Mar 2024

Cited by 4 | Viewed by 5752

Abstract

Agriculture is essential for global income, poverty reduction, and food security, with crop yield being a crucial measure in this field. Traditional crop yield prediction methods, reliant on subjective assessments such as farmers’ experiences, tend to be error-prone and lack precision across vast [...] Read more.

Agriculture is essential for global income, poverty reduction, and food security, with crop yield being a crucial measure in this field. Traditional crop yield prediction methods, reliant on subjective assessments such as farmers’ experiences, tend to be error-prone and lack precision across vast farming areas, especially in data-scarce regions. Recent advancements in data collection, notably through high-resolution sensors and the use of deep learning (DL), have significantly increased the accuracy and breadth of agricultural data, providing better support for policymakers and administrators. In our study, we conduct a systematic literature review to explore the application of DL in crop yield forecasting, underscoring its growing significance in enhancing yield predictions. Our approach enabled us to identify 92 relevant studies across four major scientific databases: the Directory of Open Access Journals (DOAJ), the Institute of Electrical and Electronics Engineers (IEEE), the Multidisciplinary Digital Publishing Institute (MDPI), and ScienceDirect. These studies, all empirical research published in the last eight years, met stringent selection criteria, including empirical validity, methodological clarity, and a minimum quality score, ensuring their rigorous research standards and relevance. Our in-depth analysis of these papers aimed to synthesize insights on the crops studied, DL models utilized, key input data types, and the specific challenges and prerequisites for accurate DL-based yield forecasting. Our findings reveal that convolutional neural networks and Long Short-Term Memory are the dominant deep learning architectures in crop yield prediction, with a focus on cereals like wheat (Triticum aestivum) and corn (Zea mays). Many studies leverage satellite imagery, but there is a growing trend towards using Unmanned Aerial Vehicles (UAVs) for data collection. Our review synthesizes global research, suggests future directions, and highlights key studies, acknowledging that results may vary across different databases and emphasizing the need for continual updates due to the evolving nature of the field. Full article

(This article belongs to the Section Information and Communication Technologies)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Technologies, Volume 12, Issue 4 (April 2024) – 14 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI