Recent Patents on Engineering - Volume 19, Issue 9, 2025

In order to assess how well conventional and hybrid pitch detection techniques perform in speech processing applications, a comparative analysis of the two types of methods is conducted.

A proposed hybrid approach, Proposed PEF+CEP, is examined alongside five traditional algorithms, namely Normalized Correlation Function (NCF), Pitch Estimation Filter (PEF), Log-Harmonic Summation (LHS), Summation of Residual Harmonics (SRH) and Cepstrum Pitch Determination (CEP). The effectiveness is evaluated using performance metrics like accuracy, specificity, sensitivity, and Gross Pitch Error (GPE).

Our findings show that the accuracy and specificity of the traditional methods are impressive; the accuracy and sensitivity of the suggested hybrid method surpass their performance, with an astounding 98.8% accuracy and 99.2% sensitivity.

Furthermore, the Proposed PEF+CEP method is a promising solution for accurate and dependable pitch detection in speech processing applications because it strikes a strong balance between computational efficiency and robustness. These results open up new avenues for research in the field of speech processing and demonstrate the potential of hybrid approaches.

A tumour is a kind of cell that grows uncontrollably and uniformly. Brain tumors (BT) are one of the leading causes of mortality in humans. In the United States, almost half of all patients die from primary BT each year. Brain tumors are diagnosed using electronic modalities. Magnetic resonance imaging (MRI) is one of the most widely used and popular electronic modalities for tumor detection.

In this research article, we included classification and optimization, with categorization using a deep convolution neural network (DCNN) and optimization using the AdaBound Optimizer (AO). When filtering input pictures, the aim of utilizing DCNN for classification is to maintain spatial connections. The spatial connection is critical for detecting the tumor-non-tumor area interface and assessing edge tissues. As a result, it categorizes the input with a higher likelihood, and for ease of implementation, the AO is utilized. It uses less memory, is more computationally efficient, and achieves superior nonstationary goals. AO also performs effectively with large datasets/parameters because of the noisy or sparse gradient requirement.

Finally, the results show improved accuracy, faster convergence, and better generalization compared to traditional optimizers. These findings highlight the potential of the model for more reliable and efficient automated medical diagnosis.

According to the simulated results, the training accuracy rate of the AO model is 91.17% and 96.87% for 5 and10 epochs, respectively. Relative to the training accuracy of the SGD optimizer models for 5 and10 epochs is 96.7% and 99%, respectively.

The blood-brain barrier (BBB) is a semipermeable, discerning barrier that keeps the CNS's internal environment stable. Constructing a concise statement on BBB permeability is challenging, yet it remains a crucial factor in developing central nervous system (CNS)-acting drugs. Clinical studies provide the most reliable assessment of BBB permeability, but they require substantial time and effort. Consequently, various computational approaches have been explored to estimate BBB permeability.

However, there has always been a problem with the precision of models used to predict BBB permeability. Using a dataset of 3912 chemicals, we trained a deep-learning (DL) and machine learning (ML) algorithm to better predict BBB permeability. There were 1,917 features stored for each compound; they included 1356 physicochemical (1D & 2D) attributes, 174 molecular-access-system (MACCS), & 311 substructure fingerprints.

We compared and contrasted the created models' prediction performance metrics. It was determined that the prediction accuracy of the DNN was 99.58%, the one-dimensional convolutional neural network (CNN-1D) was 98.36%, and the CNN by transfer learning (VGG16) achieved 97.23%. The “DeePred-BBB” framework, which predicts the BBB-permeability of substances utilizing their simplified-molecular-input-line-entry-system (SMILES) notations, was built using the top-performing DNN-based model. It might be helpful in the early phases of medication development for screening compounds according to their BBB permeability.

This advancement highlights its potential utility in early CNS drug development. Additionally, “DeePred-BBB” integrates a user-friendly interface, enabling seamless compound screening for researchers and drug developers.

DDoS attacks, where numerous compromised systems overwhelm a target with traffic, are significant threats to online services' stability. These attacks exploit the fundamentals of internet communication, using botnets to flood targets and deplete their resources, severely reducing performance. The strength of DDoS attacks lies in their distributed nature, which complicates the differentiation between legitimate and malicious traffic. As digital reliance grows, so does the significance of these attacks, which impact businesses, governments, and public services crucial for operations.

The paper aims to provide a comprehensive understanding of DDoS attacks, categorizing them into bandwidth and resource depletion, infrastructure, and zero-day attacks. It seeks to highlight the complexity and impact of these attacks, particularly those utilizing IoT botnets, on internet reliability and security. The study emphasizes the limitations of current defense mechanisms, advocating for improved strategies that consider the distributed nature of these threats. Through this analysis, the paper aims to foster a deeper understanding of DDoS attacks, their consequences, and the need for more effective mitigation and prevention techniques.

The patent study employs an in-depth literature review to classify DDoS attacks and explore various mitigation strategies. It provides a detailed examination of attack mechanisms, including bandwidth depletion, resource depletion, infrastructure attacks, and zero-day vulnerabilities. The paper discusses several defense techniques, such as filtering, intrusion detection systems, and advanced AI and machine learning approaches. It emphasizes the role of IoT devices in amplifying DDoS attacks and the challenges of defending against these evolving threats.

The paper identifies four main categories of DDoS attacks and describes their operational mechanisms, impacts, and mitigation challenges. It reveals that due to inadequate security measures, IoT devices significantly contribute to the scale and impact of DDoS attacks. Despite the various defense mechanisms discussed, the paper points out their limitations in effectively countering the evolving nature of DDoS attacks. It emphasizes the need for more robust, adaptive strategies incorporating technological advancements and better security practices in IoT device manufacturing.

DDoS attacks, particularly those leveraging IoT botnets, pose increasingly sophisticated threats to digital infrastructure. The paper underscores the urgent need for more effective defense mechanisms, highlighting the importance of technological advancements, better IoT security, and collaborative efforts among stakeholders. It calls for future research focused on developing AI-driven systems for real-time prediction and mitigation of attacks, as well as the formulation of international cyber-security policies to address the growing menace of DDoS attacks in a globally connected environment.

Free Space Optics (FSO) is a wireless data transmission method for infrastructure that uses laser beam energy to transmit information waves through the atmosphere. Furthermore, due to its high bandwidth potential and simple deployment, FSO has garnered considerable interest. However, atmospheric turbulence and misalignment present obstacles to establishing dependable and effective FSO links.

For systems varying from space invariant to totally space variant, the optimal design of free-space optical connectivity systems using diffractive optics is found from an engineering perspective. Parameters such as the light's wavelength, the system's total number of optical sources and detectors, their sizes, and their spacing are used to determine the system's volume. Another important parameter is the diffractive lens's f-number. Diffraction Optical Elements (DOEs) have emerged as a promising means of addressing these difficulties. Also, the patent related to automated honey beehive box gives the insight of its monitoring system.

This paper provides an overview of the implementation and advancements of FSO systems utilizing DOEs, including the fundamental principles, design considerations, and performance improvements. The study discusses the basics of diffraction and the role of DOEs in FSO systems. It explores the diffraction grating equation and the Huygens-Fresnel principle to understand wave propagation and interference phenomena. Design considerations for FSO systems equipped with DOEs are discussed, including the selection of appropriate DOEs and evaluation of performance benefits. The study also investigates the application of AI methods, such as machine learning and deep learning, in optimizing FSO systems with DOEs.

A thorough overview of Free Space Optics (FSO) systems utilizing Diffraction Optical Elements (DOEs) is given in this review study. It examines diffraction theory and DOE use in FSO, emphasizing their potential for beam forming, beam steering, and adaptive optics. The study examines FSO with DOE design concerns, performance improvements, applications, and future approaches. FSO systems may overcome problems with air turbulence, misalignment, and fading by using the characteristics of DOEs, opening the door for dependable and effective wireless communication.

In conclusion, the effect of DOEs on BER efficiency is also modified by the obscuration ratio. Transmission power is increased when more DOEs are used by an amount defined by their obscuration ratios. Additionally, because of the increased power complement in these systems, the effect of DOEs is more pronounced. The integration of AI further enhances FSO capabilities by providing adaptive optimization, fault detection, predictive maintenance, and improved security. Future research directions may include exploring advanced AI techniques and conducting practical implementations of FSO systems with DOEs for various applications, particularly in Internet of Things (IoT) scenarios.

The escalating expansion of global telecom towers since the 1990s has led to an ageing infrastructure, with towers exceeding 30 years. Rust, a pervasive threat, jeopardizes structural integrity, posing risks of collapse and harm to riggers. Recognizing the importance of innovation in addressing this challenge, rust detection and tower maintenance patents play a crucial role in advancing the field.

This patent paper focuses on detecting and evaluating rust levels on ageing telecom towers. Beyond addressing the immediate concerns, the objective includes exploring patented technologies in rust detection and prevention. The aim is to amalgamate innovative solutions into the algorithmic framework for automated rust detection, thereby enhancing the robustness and novelty of the proposed approach.

The study involves a comprehensive literature review to identify patented technologies related to rust detection on structures like telecom towers. By integrating patented methods into the image processing code developed in Python, the algorithm gains an edge in accuracy and efficiency. The three-tier classification of rust severity aligns with patented preventative maintenance strategies, ensuring a holistic approach to tower care.

The amalgamation of patented technologies with the image processing code enhances the accuracy of rust detection on telecom towers. The results guide targeted treatments, informed by patented preventative maintenance strategies, including patented rust inhibitors, protective coatings, and corrosion-resistant materials.

This paper introduces an innovative algorithm for rust detection and emphasizes the integration of patented technologies. Incorporating patented solutions into preventative maintenance strategies makes the approach more effective and technologically advanced. This holistic method not only benefits telecom companies in accident prevention and cost reduction but also positions the developed algorithm as a significant contribution to patented rust detection and tower maintenance technologies.

Choosing an e-learning website is crucial for researchers and developers in today's fiercely competitive world. The concept of an e-learning website is expanding quickly these days. When a substantial number of selection criteria are considered across a wide range of options, the process of selecting and ranking e-learning websites becomes a complex and challenging task.

The present research aims on the evaluation and selection of E-learning websites by modelling it as a multi-criteria decision making (MCDM) problem due the dependency of multiple conflicting attributes. This study targets on the upcoming problem of evaluating, ranking, and selecting E-learning websites, having a significant impact on the educational industry.

The present research work proposes the W-CODAS strategy for optimizing e-learning websites, which combines the Shannon Entropy approach which is used to calculate the weights of the selection criteria with the CODAS technique. The results obtained will be protected using patents.

According to the method and framework that is proposed, the E-learning website exhibiting the lowest preference index value is assigned the top rank or position, whereas the website which has the highest value is ranked or placed at last.

This study was conducted to rank the various E-learning websites. In particular, an integrated MCDM method has been developed by combining entropy approach and distance based approach. It is established that once a complete set of selection criteria and their respective priority weights have been identified, this CoDAS method can be applied to rank them. The interdependencies/priority weights of the selection criteria have been given due consideration in the CoDAS method. The proposed methodology i.e. W-CoDAS in the present research is based on the simple mathematical matrix operations that reduce the complexity of the evaluation distance-based process as well as implementation time. The results are verified by comparing the W-CODAS ranking findings to established MCDM systems like TOPSIS and VIKOR, and employing Spearman's rank correlation test, enhances the robustness and credibility of the research.

Litchi is a famous fruit with a large planting area and high output, which has become a pillar industry in many regions. Due to many uncontrollable factors in the production process, such as picking time, fruit size, weather conditions, etc., there are large differences in the quality of litchi, which directly affect the sales price.

This paper selected the intelligent sorting system based on image recognition technology by investigating different types of intelligent sorting systems in the market. The sorting system could identify litchi with small, fragile, and irregular shapes. By applying image recognition technology to the field of litchi logistics, the problem of information asymmetry between fruit farmers and logistics enterprises could be effectively solved. This paper designed an intelligent sorting system based on a convolutional neural network (CNN), which used image recognition and classification technology to recognize litchi. By comparing the processing effects of different algorithms, the CNN network suitable for the system architecture was selected for training.

The research results showed that under the same other conditions, the passing rate of the intelligent litchi sorting system under the YOLOv5 was 11.1%, and the failing rate was 88.9%. The passing rate of the intelligent litchi sorting system based on the CNN algorithm was 95.6%, and the failing rate was 4.4%.

It shows the positive relationship between the CNN algorithm and the litchi intelligent sorting system, and shows that the system can effectively identify and sort litchi with different shapes.

Utilizing highly concentrated slurry is recommended due to its ability to reduce both operational expenses and water use. In the previous studies, the pressure drop analysis on coal, sand, and coal ashes was investigated. However, there is a scarcity of research on the pressure drop properties of iron ore slurry, particularly when it comes to highly concentrated slurries.

This patent aims to replicate the iron ore flow in a hydro-slurry pipe, specifically focusing on predicting the features of pressure drop, distribution of volume fraction, and behaviour of solid particles.

This patent presents the CFD modelling of pressure drop characteristics of iron ore-water multiphase flow inside a hydro-slurry pipeline.

A granular flow was represented using an Eulerian technique based on kinetic theory to depict multiphase phenomena. Simulations were carried out on a pipeline with a 50 mm diameter. The velocity ranged from 2 to 5 m/s, whereas the efflux concentration varied between 20% and 60%. An analysis was conducted on the impact of granular size at a greater concentration. The numerical code was validated using experimental findings, and it was determined that the RNG k-ε turbulent model exhibited satisfactory validation with the experimental data.

Results show that the augmentation in pressure drop is non-linearly correlated with both the granular concentration and the velocity. The size of the efflux concentration zone expands as the concentration rises, but this zone shrinks as the velocity increases. The variation in volume fraction at the lower periphery of the pipe decreases with an increase in velocity and increases with the size of particle and granular concentration. The turbulent intensity of the mixture was affected marginally with an increase in concentration but highly by velocity. The variation in granular size increased turbulence as large particles caused additional turbulence. The velocity profile recorded marginal variation in the pattern of solid phase flow with variations in granular concentration, granular size, and velocity. The change in velocity resulted in particle shifting.

As per this patent, the RNG k-ε turbulent model is superior to the other multiphase models for ore-water flow analysis.

An infinite vertical sheet is the medium of inquiry for studying the copper-water nanofluid behavior with an inclined magnetic field and Cogley radiation. The research contains significant aspects. The use of this nanofluid is of great practical importance, especially in the areas of energy efficiency, thermal management, radiative cooling for space technology, environmental impact reduction, and improving heat transfer efficiency in electronic device cooling operations. Moreover, the research enhances the understanding and manipulation of sophisticated materials. The uttermost significance of this resides in its capacity to uncover issues inherent in the stages of planning, developing, analyzing, and improving manufacturing processes, possibly securing new solutions via patent protection.

This research aims to provide a comprehensive analysis of the unsteady natural convection of a nanofluid along an infinite vertical sheet. The study focuses on understanding how an inclined magnetic field, Cogley radiation, heat source, and varying nanoparticle volume concentrations impact the velocity, thermal, and concentration profiles of the nanofluid. Additionally, the investigation seeks to evaluate the Nusselt number, Sherwood number, and skin friction coefficient across different concentration profiles to derive meaningful insights.

The Laplace transform (L.T.) method, in combination with the MATLAB symbolic computing program, is used to develop numerical solutions for the main boundary value issues. The Laplace transform technique is used to clarify dimensionless partial differential equations (PDEs) and their limiting situations. An in-depth study is conducted on the profiles of momentum, energy, and concentration for various kinds of nanofluids with changing volume concentrations of nanoparticles. This research provides a comprehensive examination and also identifies prospective prospects in this field that might be protected by patents.

The work provides an in-depth understanding of how inclined magnetic fields, Cogley radiation, heat generation, and nanoparticle volume concentration affect the velocity, energy, and mass profiles of the nanofluid. Graphical representations depict these effects, offering a visual comprehension of the system's behavior. The Nusselt number and skin friction coefficient are determined by analytical and numerical methods. These findings demonstrate that the Nusselt number and skin friction coefficient have an inverse relationship with changes in concentration profiles.

The work improves our comprehension of the fluctuating natural convection of nanofluids across an unbounded vertical surface, taking into account variables such as an inclined magnetic field, heat source, Cogley radiation, and nanoparticle volume concentration. The practical implications of the discoveries, together with the analytical and numerical results, enhance our comprehension of the dynamics of nanofluid systems. Ensuring the protection and widespread distribution of patented technology is essential for promoting sustainable advancements.

Electro-hydraulic position servo system is widely used, in order to cope with complex working conditions, the system is required to have high dynamic response and accurate position control ability, while the characteristics of the system and its control strategy need to be studied in depth.

To review the characteristics of electro-hydraulic position servo systems and typical control strategies under complex working conditions, and summarize their advantages and disadvantages to provide a basis for the design and optimization of electro-hydraulic position servo systems.

Literature and patents related to electrohydraulic position servo system characteristics and control strategies are collected, synthesized and summarized.

The characteristics of electro-hydraulic position servo system under complex working conditions are summarized, especially the performance of the system and the application effect of the control strategy under the working conditions such as load disturbance and gap nonlinearity. In addition, the application of sliding mode control, adaptive control and neural network control in electrohydraulic servo systems is summarized.

In different complex working conditions, the selection of appropriate control strategies can improve the performance and stability of electro-hydraulic position servo systems. The study of the characteristics of electro-hydraulic position servo systems and typical control strategies is of great significance to the development of electro-hydraulic position servo systems.

As the demand for strawberries increases, the growing area of strawberries is also increasing. However, the main method of harvesting strawberries is the traditional artificial harvest, so the harvest is the most time-consuming and laborious operation in the production chain.

By summarizing the recent years of strawberry harvest research, some valuable conclusions have been drawn to predict the future development of strobing machinery, providing reference to researchers in relevant fields.

This paper examines the patents of recent strawberry harvesting machinery, by comparing the characteristics of various mechanical structures, to plan the research route of the future strawberry machine.

The emerging problems and future directions of technology are identified through analysis and discussion of existing technologies.

Improvements in its structure alone do not meet the demands of modernization, and strawberry harvesting machines need to be more intelligent and integrated and use more efficient ways. More relevant patents are needed to be invented.

Tracheotomy is a surgical technique in which the trachea in the front of the neck is cut to construct an artificial airway. Bypassing blockage or lesions in the upper airway, this opening permits oxygen provision and carbon dioxide exhalation straight through the trachea. A more profound comprehension of the pathophysiologic role of the airway has made tracheotomy a crucial component of treatment for several illnesses. Medical devices used both during and after a tracheotomy are known as auxiliary devices for tracheotomies. These devices, which include dilators, cannulas, plugging devices for cannulas, and immobilization devices, ensure the procedure is completed safely and the patient recovers quickly.

This paper aims to identify the key issues, summarize the many tracheotomy auxiliary device constructions in recent years, and serve as a resource for researchers working on related topics.

Assess and evaluate the benefits and drawbacks of the tracheotomy auxiliary devices covered by the patents filed in the last several years and identify areas for development.

This patent research examines the shortcomings of current tracheotomy auxiliary devices and speculates about future directions for their advancement.

Although they are challenging to use, the tracheotomy auxiliary devices now in use provide good stability and dependability. Enhancing the mechanical structure of tracheotomy auxiliary devices and strengthening clinical trials for various patient populations and intricate surgical settings are essential improvements. Furthermore, pertinent patents for auxiliary equipment used in tracheotomies have not yet been created.

With advances in medical technology and an aging population, the number of patients with upper limb movement disorders has increased, who are facing difficulties in self-care and occupational integration. Traditional rehabilitation methods have limitations, such as unstable results, long cycle times, and insufficient resources. Therefore, upper limb rehabilitation robotics has emerged, combining robotics, medicine, and other fields to simulate upper limb movement for targeted rehabilitation training in order to improve effectiveness, shorten the cycle, and reduce the burden on therapists.

This study aimed to introduce the classification, advantages and disadvantages, and development trend in existing upper limb rehabilitation robots and provide a basis for other researchers to understand the current status of their development and future trends.

Various studies and patents on upper limb rehabilitation robots were reviewed, revealing the structural characteristics, along with the advantages and disadvantages, of typical robotic arms used for upper limb rehabilitation.

Through the analysis of various upper limb rehabilitation robots, the characteristics and problems of upper limb rehabilitation robots were identified, and the development trend of upper limb rehabilitation robots was prospected.

Upper limb rehabilitation robots have many advantages, such as good adaptability, personalized customization, safety and reliability, etc., but they also have some disadvantages, such as difficult control, high manufacturing cost, and short service life. In the future, upper limb rehabilitation robots will develop towards simpler structures, greater comfort, affordability, diverse functions, better efficacy, and increased safety.

The planting technology of rice mulching and rice transplanting can effectively control weed growth, reduce fertilizer loss, maintain surface temperature, improve fertilizer utilization, reduce the use of pesticides and chemical fertilizer, and is a new mode of green rice cultivation. The rice mulching device is a key component, and the quality of its mulching directly affects the productivity and planting quality of rice. Additionally, obtaining an efficient rice transplanter with an integrated rice transplanter for transplanting and mulching is the hot spot of research in this field.

By analyzing and discussing the patents of rice mulching cultivation technology research and mechanical mulching devices, their development trend is summarized, which provides a reference for the development of rice transplanter mulching devices.

The research status of rice transplanter mulching devices in recent years is summarized, and various mulching devices' structure types and applications are discussed.

By summarizing the patents on rice transplanter mulching devices, the current status of patent applications was obtained, and the problems and future research direction of rice transplanter mulching devices were discussed.

In-depth study of various types of rice transplanter mulching devices, the existing rice transplanter mulching device needs further improvement regarding the laying mechanism, the film-breaking mechanism, and the structural type of the mulching device. With the application of GPS, sensor detection, information processing, automatic control, and other technologies in agricultural machinery, rice transplanter filming is gradually developing in the direction of high speed, automation, and intelligence.

This patent study investigates the application of friction stir welding (FSW) to join aluminium alloy 6061 with high-strength interstitial free galvannealed steel. The effects of tool rotational speed, traverse speed, and shoulder diameter on the mechanical properties of the joints were examined.

Optimal conditions were identified, with a tool rotational speed of 1200 rpm and a traverse speed of 100 mm/min, significantly enhancing the ultimate tensile strength (UTS) and impact strength of the joints. The optimized parameters resulted in UTS values reaching up to 320 MPa. Microstructural analysis revealed substantial grain refinement in the stir zone (SZ), leading to increased micro-hardness values of up to 150 HV, compared to the base materials (BM) and heat-affected zones (HAZ).

Micro-hardness often rises as a result of grain refining brought on by tool rotation in the SZ and TMAZ. However, as the indentation distance from the SZ grows, the micro-hardness declines, especially in the HAZ and the BM's grain-coarsened region. The distribution of micro-hardness also shows that dynamic recrystallization in the FSW leads to considerable grain refinement in the SZ. SZ is, hence, harder on the microscale than BM and HAZ.

The findings highlight the critical role of process parameter optimization in improving the mechanical performance of aluminium-steel dissimilar joints.

As the COVID-19 pandemic develops, there is a lot of false information floating around on social media, and the risks are huge. This is why identifying and countering disinformation campaigns is so important. Modern deep learning models that employ Natural-Language-Processing (NLP) approaches, such as Bidirectional-Encoder-Representations-from-Transformers (BERT), have been quite effective at identifying disinformation.

To tackle the spread of false information on COVID-19, we present an explainable NLP approach that utilizes DistilBERT and Shapley-Additive-exPlanations (SHAP), two powerful and efficient frameworks. A dataset consisting of 984 assertions regarding COVID-19 that were factually verified was initially compiled. The DistilBERT model achieved superior performance in spotting COVID-19 disinformation after we doubled the dataset's sample size using back-translation.

Compared to more conventional machine learning models, it performed better on both datasets. Additionally, we used SHAP to enhance the explainability of the models, which were then tested in amid-subjects experimentation with three constraints: text(T), text+SHAP explanation(TSE), and text+SHAP explanation+source and evidence(TSESE). The goal was to increase public trust in the models' predictions.

Compared to the T condition, the TSE &TSESE constraints showed a substantial increase in participants' confidence and sharing of COVID-19-related information. Improving public trust and identifying COVID-19 disinformation were two important outcomes of our study.