Improvement In Accuracy Research Articles

Comprehensive Improvement of Binocular Structured Light Calibration Method Based on Radical-Conservative Cooperative Particle Swarm.

To achieve high-precision 3D reconstruction, a comprehensive improvement has been made to the binocular structured light calibration method. During the calibration process, the calibration object's imaging quality and the camera parameters' nonlinear optimization effect directly affect the caibration accuracy. Firstly, to address the issue of poor imaging quality of the calibration object under tilted conditions, a pixel-level adaptive fill light method was designed using the programmable light intensity feature of the structured light projector, allowing the calibration object to receive uniform lighting and thus improve the quality of the captured images. Then, collaborative Particle Swarm Optimization was studied to optimize the camera parameters. Compared with other optimization algorithms, this algorithm has higher global search capability and can obtain more accurate camera parameters. Under comprehensive improvement, the 3D reconstruction accuracy of binocular structured light is 0.053 mm, showing a 36.33% improvement in reconstruction accuracy compared to mainstream calibration methods.

AI model using CT-based imaging biomarkers to predict hepatocellular carcinoma in patients with chronic hepatitis B.

Various hepatocellular carcinoma (HCC) prediction models have been proposed for patients with chronic hepatitis B (CHB) using clinical variables. We aimed to develop an artificial intelligence (AI)-based HCC prediction model by incorporating imaging biomarkers derived from abdominal computed tomography (CT) images along with clinical variables. An AI prediction model employing a gradient-boosting machine algorithm was developed utilizing imaging biomarkers extracted by DeepFore, a deep learning-based CT auto-segmentation software. The derivation cohort (n=5,585) was randomly divided into the training and internal validation sets at a 3:1 ratio. The external validation cohort included 2,883 patients. Six imaging biomarkers (i.e., abdominal visceral fat-total fat volume ratio, total fat-trunk volume ratio, spleen, and liver volume; liver-spleen Hounsfield unit [HU] ratio; and muscle HU) and eight clinical variables were selected as the main variables of our model, PLAN-B-DF. In the internal validation set (median follow-up duration=7.4 years), PLAN-B-DF demonstrated an excellent predictive performance with a c-index of 0.91 and good calibration function (P=0.78 by the Hosmer-Lemeshow test). In the external validation cohort (median follow-up duration=4.6 years), PLAN-B-DF showed a significantly better discrimination function compared to previous models including PLAN-B, PAGE-B, modified PAGE-B, and CU-HCC (c-index, 0.89 vs. 0.65-0.78; all P<0.001) and maintained a good calibration function (P=0.42 by the Hosmer-Lemeshow test). When patients were classified into four groups according to the risk probability calculated by PLAN-B-DF, the 10-year cumulative HCC incidence was 0.0%, 0.4%, 16.0%, and 46.2% in the minimal-, low-, intermediate-, and high-risk groups, respectively. This AI prediction model, integrating deep learning-based auto-segmentation of CT images, offers improved performance in predicting HCC risk among patients with CHB compared to previous models. The AI-driven HCC prediction model (PLAN-B-DF), employing an automated CT segmentation algorithm, demonstrates a significant improvement in predictive accuracy and risk stratification among patients with CHB. Using a gradient-boosting algorithm and CT metrics such as visceral fat volume and myosteatosis, PLAN-B-DF outperforms previous models based solely on clinical and demographic data. This model not only shows a higher c-index compared to previous models, but also effectively classifies CHB patients into different risk groups. This model uses machine learning to analyze the complex relationships among various risk factors contributing to HCC occurrence, thereby offering more personalized surveillance for CHB patients.

TDMSANet: A Tri-Dimensional Multi-Head Self-Attention Network for Improved Crop Classification from Multitemporal Fine-Resolution Remotely Sensed Images

Accurate and timely crop distribution data are crucial for governments, in order to make related policies to ensure food security. However, agricultural ecosystems are spatially and temporally dynamic systems, which poses a great challenge for accurate crop mapping using fine spatial resolution (FSR) imagery. This research proposed a novel Tri-Dimensional Multi-head Self-Attention Network (TDMSANet) for accurate crop mapping from multitemporal fine-resolution remotely sensed images. Specifically, three sub-modules were designed to extract spectral, temporal, and spatial feature representations, respectively. All three sub-modules adopted a multi-head self-attention mechanism to assign higher weights to important features. In addition, the positional encoding was adopted by both temporal and spatial submodules to learn the sequence relationships between the features in a feature sequence. The proposed TDMSANet was evaluated on two sites utilizing FSR SAR (UAVSAR) and optical (Rapid Eye) images, respectively. The experimental results showed that TDMSANet consistently achieved significantly higher crop mapping accuracy compared to the benchmark models across both sites, with an average overall accuracy improvement of 1.40%, 3.35%, and 6.42% over CNN, Transformer, and LSTM, respectively. The ablation experiments further showed that the three sub-modules were all useful to the TDMSANet, and the Spatial Feature Extraction Module exerted larger impact than the remaining two sub-modules.

Effective Stemmers Using Trie Data Structure for Enhanced Processing of Gujarati Text

Stemming plays a crucial role in natural language processing and information retrieval. It is challenging for the Gujarati language due to the complex morphology of several stemming algorithms for the Gujarati language that have been developed using rule-based, dictionary-based, or hybrid approaches. However, they are computationally expensive, produce more over-stemming errors and have limited accuracy. This paper introduces three novel optimized Gujarati stemmers using a trie data structure to overcome the above-mentioned limitations. The significant contributions to this paper are as follows. First, three optimized Gujarati stemmers, namely Optimized Gujarati Stemmer using Suffix Stripping Approach (OGS_SSA), Optimized Gujarati Stemmer using Rule-Based Approach (OGS_RBA), and Optimized Gujarati Stemmer using Re-parsing Based Approach (OGS_RPA), are proposed. Second, a novel algorithm to create a Gujarati dictionary using the trie data structure is proposed. Third, the proposed stemmers are rigorously assessed using three standard datasets, namely entertainment, health, and agriculture. The performance of the proposed stemmers is measured using evaluation parameters such as precision, recall, F1 score, accuracy, number of stemming errors and processing time. The results show that OGS_RPA consistently exceeds the OGS_SSA and OGS_RBA for precision, recall, F1 score, and accuracy. In addition, it exhibits a lower number of stemming errors. Moreover, the performance of the proposed stemmer is compared with the existing Gujarati hybrid stemmer. The results show a 14–16% improvement in accuracy and less processing time compared to the Gujarati hybrid stemmer. OGS_SSA demonstrated enhanced processing time, making it a feasible option for applications that prioritize prompt response time. Furthermore, it demonstrates 10–11% enhancement in accuracy and a reduction in processing time than the Gujarati hybrid stemmer. OGS_RBA exhibits moderate performance due to its rule-based methodology compared to OGS_RPA and OGS_SSA. However, it shows 10–13% improvement in accuracy than the Gujarati hybrid stemmer.

Dynamic Road Anomaly Detection: Harnessing Smartphone Accelerometer Data with Incremental Concept Drift Detection and Classification.

Effective monitoring of road conditions is crucial for ensuring safe and efficient transportation systems. By leveraging the power of crowd-sourced smartphone sensor data, road condition monitoring can be conducted in real-time, providing valuable insights for transportation planners, policymakers, and the general public. Previous studies have primarily focused on the use of pre-trained machine learning models and threshold-based methods for anomaly classification, which may not be suitable for real-world scenarios that require incremental detection and classification. As a result, there is a need for novel approaches that can adapt to changing data environments and perform effective classification without relying on pre-existing training data. This study introduces a novel, real-time road condition monitoring technique harnessing smartphone sensor data, addressing the limitations of pre-trained models that lack adaptability in dynamic environments. A hybrid anomaly detection method, combining unsupervised and supervised learning, is proposed to effectively manage concept drift, demonstrating a significant improvement in accuracy and robustness with a 96% success rate. The findings underscore the potential of incremental learning to enhance model responsiveness and efficiency in distinguishing various road anomalies, offering a promising direction for future transportation safety and resource optimization strategies.

Advancing Educational Management with the ATT-MR-WL Intelligent Question-answering Model

Higher education plays a critical role in cultivating talent, preserving culture, and promoting social progress. However, current challenges, such as inefficient information dissemination and low problem-solving efficiency among students, highlight the need for intelligent question-answering systems. These systems, leveraging artificial intelligence and natural language processing technologies, enable rapid and accurate responses to student queries, thereby providing intelligent support for higher education management. This study introduces the ATT-MR-WL model, a generative AI system integrating Mask R-CNN and Word2Vec+LSTM to enhance intelligent question-answering functionality. The model, customized to handle both text and visual data, is evaluated using the established VQA v2.0 dataset and a specially developed EM dataset reflecting university management scenarios. The ATT-MR-WL model demonstrates a 3% accuracy improvement over traditional methods and enhances its ability to handle multimodal queries. This research provides important insights for enhancing the efficiency and quality of higher education management and advancing the process of educational informatization.

Detection of source code vulnerabilities using Nature language processing and deep graph network

The software production sector gains advantages from automated code generating techniques, yet encounters issues related to vulnerabilities in the resulting code. This research presents a hybrid paradigm, termed GBD, for detecting vulnerabilities in software written in C and C++. It integrates Graph Convolution Network (GCN), Bidirectional Encoder Representations from Transformers (BERT), and Dropout. During Phase 2 of the GBD model, the subsequent tasks are executed concurrently: (i) obtaining node and edge features utilizing the GCN graph convolution network; (ii) deriving segment features employing the BERT model; (iii) constructing a source code profile via the Code Property Graph (CPG). Phase 3 of the model implements the Dropout strategy to mitigate overfitting. Phase 4 is the classifier that ascertains the presence of vulnerabilities in the source code. Experimental findings demonstrate the superiority of the proposed model relative to alternative methods, attaining a prediction accuracy of 61.21% for vulnerable code and 88.94% for normal files. Additionally, the classification outcomes demonstrate that with a token length of 512, the GBD model yields the most uniform results across all metrics: Accuracy (86.65%), Precision (38.59%), Recall (66.21%), and F1-score (48.76%). This corresponds with our analysis of the Verum experimental dataset, indicating that over 70% of the source code files have lengths exceeding 256 but less than 512. Furthermore, the GBD model exhibits strong performance across both individual and multiple datasets. For example, in the Verum dataset, the GBD model surpasses five alternative methodologies—REVEAL [1], Russell [2], VulDeePecker [3], SySeVR [4], and Devign [5] - by 4% in Accuracy and between 15% and 57% in Precision, Recall, and F1-score. In comparison to SySeVR [4], the GBD model exceeds it by 3% to 25% across all metrics. In comparison to Devign [5], GBD achieves improvements of 5% to 39% in Precision, Recall, and F1-score. Upon assessment of the FFmpeg+Qume dataset, the GBD model attains an Accuracy improvement ranging from 0.2% to 10% above all other studies. In terms of precision, GBD surpasses alternative methods by 0.3% to 9%. In terms of Recall, GBD is marginally worse than REVEAL by 1.5%, although surpasses all other methodologies by 10% to over 31%. In terms of F1-score, GBD is 0.3% inferior to REVEAL but surpasses other studies by 7% to 30%. The results indicate that the GBD model is effective on both individual and multiple datasets

Structural Attributes Injection Is Better: Exploring General Approach for Radar Image ATR with a Attribute Alignment Adapter

Nowadays, deep learning techniques are extensively applied in the field of automatic target recognition (ATR) for radar images. However, existing data-driven approaches frequently ignore prior knowledge of the target, leading to a lack of interpretability and poor performance of trained models. To address this issue, we first integrate the knowledge of structural attributes into the training process of an ATR model, providing both category and structural information at the dataset level. Specifically, we propose a Structural Attribute Injection (SAI) module that can be flexibly inserted into any framework constructed based on neural networks for radar image recognition. Our proposed method can encode the structural attributes to provide structural information and category correlation of the target and can further apply the proposed SAI module to map the structural attributes to something high-dimensional and align them with samples, effectively assisting in target recognition. It should be noted that our proposed SAI module can be regarded as a prior feature enhancement method, which means that it can be inserted into all downstream target recognition methods on the same dataset with only a single training session. We evaluated the proposed method using two types of radar image datasets under the conditions of few and sufficient samples. The experimental results demonstrate that our application of our proposed SAI module can significantly improve the recognition accuracy of the baseline models, which is equivalent to the existing state-of-the-art (SOTA) ATR approaches and outperforms the SOTA approaches in terms of resource consumption. Specifically, with the SAI module, our approach can achieve substantial accuracy improvements of 3.48%, 18.22%, 1.52%, and 15.03% over traditional networks in four scenarios while requiring 1/5 of the parameter count and just 1/14 of the FLOPs on average.

A Rice Leaf Area Index Monitoring Method Based on the Fusion of Data from RGB Camera and Multi-Spectral Camera on an Inspection Robot

Automated monitoring of the rice leaf area index (LAI) using near-ground sensing platforms, such as inspection robots, is essential for modern rice precision management. These robots are equipped with various complementary sensors, where specific sensor capabilities partially overlap to provide redundancy and enhanced reliability. Thus, leveraging multi-sensor fusion technology to improve the accuracy of LAI monitoring has become a crucial research focus. This study presents a rice LAI monitoring model based on the fused data from RGB and multi-spectral cameras with an ensemble learning algorithm. The results indicate that the estimation accuracy of the rice LAI monitoring model is effectively improved by fusing the vegetation index and textures from RGB and multi-spectral sensors. The model based on the LightGBM regression algorithm has the most improvement in accuracy, with a coefficient of determination (R2) of 0.892, a root mean square error (RMSE) of 0.270, and a mean absolute error (MAE) of 0.160. Furthermore, the accuracy of LAI estimation in the jointing stage is higher than in the heading stage. At the jointing stage, both LightGBM based on optimal RGB image features and Random Forest based on fused features achieved an R2 of 0.95. This study provides a technical reference for automatically monitoring rice growth parameters in the field using inspection robots.

Task-relevant stimulus design improves P300-based brain-computer interfaces.

In the pursuit of refining P300-based brain-computer interfaces (BCIs), our research aims to propose a novel stimulus design focused on selective attention and task relevance to address the challenges of P300-based BCIs, including the necessity of repetitive stimulus presentations, accuracy improvement, user variability, and calibration demands. In the oddball task for P300-based BCIs, we develop a stimulus design involving task-relevant dynamic stimuli implemented as finger-tapping to enhance the elicitation and consistency of event-related potentials (ERPs). We further improve the performance of P300-based BCIs by optimizing ERP feature extraction and classification in offline analyses. With the proposed stimulus design, online P300-based BCIs in 37 healthy participants achieve an accuracy of 91.2% and an information transfer rate (ITR) of 28.37 bits/min with two stimulus repetitions. With optimized computational modeling in BCIs, our offline analyses reveal the possibility of single-trial execution, showcasing an accuracy of 91.7% and an ITR of 59.92 bits/min. Furthermore, our exploration into the feasibility of across-subject zero-calibration BCIs through offline analyses, where a BCI built on a dataset of 36 participants is directly applied to a left-out participant with no calibration, yields an accuracy of 94.23% and the ITR of 31.56 bits/min with two stimulus repetitions and the accuracy of 87.75% and the ITR of 52.61 bits/min with single-trial execution. When using the finger-tapping stimulus, the variability in performance among participants is the lowest, and a greater increase in performance is observed especially for those showing lower performance using the conventional color-changing stimulus. Signficance. Using a novel task-relevant dynamic stimulus design, this study achieves one of the highest levels of P300-based BCI performance to date. This underscores the importance of coupling stimulus paradigms with computational methods for improving P300-based BCIs.

Securing Authentication and Detecting Malicious Entities in Drone Missions

This study proposes a hierarchical communication framework for drone swarms designed to enhance security and operational efficiency. Leveraging elliptic curve cryptography and space quanta concepts, the model ensures continuous authentication and risk assessment of participating entities. Experimental results demonstrate the framework’s effectiveness in mitigating security risks, achieving reliable communication even in adverse conditions. Key findings include significant improvement in threat detection accuracy and reduced computational overhead, validating the model’s applicability for real-world drone swarm operations. These contributions establish a robust foundation for secure and resilient drone coordination.

The Effect of Processing Techniques on the Classification Accuracy of Brain-Computer Interface Systems.

Background/Objectives: Accurately classifying Electroencephalography (EEG) signals is essential for the effective operation of Brain-Computer Interfaces (BCI), which is needed for reliable neurorehabilitation applications. However, many factors in the processing pipeline can influence classification performance. The objective of this study is to assess the effects of different processing steps on classification accuracy in EEG-based BCI systems. Methods: This study explores the impact of various processing techniques and stages, including the FASTER algorithm for artifact rejection (AR), frequency filtering, transfer learning, and cropped training. The Physionet dataset, consisting of four motor imagery classes, was used as input due to its relatively large number of subjects. The raw EEG was tested with EEGNet and Shallow ConvNet. To examine the impact of adding a spatial dimension to the input data, we also used the Multi-branch Conv3D Net and developed two new models, Conv2D Net and Conv3D Net. Results: Our analysis showed that classification accuracy can be affected by many factors at every stage. Applying the AR method, for instance, can either enhance or degrade classification performance, depending on the subject and the specific network architecture. Transfer learning was effective in improving the performance of all networks for both raw and artifact-rejected data. However, the improvement in classification accuracy for artifact-rejected data was less pronounced compared to unfiltered data, resulting in reduced precision. For instance, the best classifier achieved 46.1% accuracy on unfiltered data, which increased to 63.5% with transfer learning. In the filtered case, accuracy rose from 45.5% to only 55.9% when transfer learning was applied. An unexpected outcome regarding frequency filtering was observed: networks demonstrated better classification performance when focusing on lower-frequency components. Higher frequency ranges were more discriminative for EEGNet and Shallow ConvNet, but only when cropped training was applied. Conclusions: The findings of this study highlight the complex interaction between processing techniques and neural network performance, emphasizing the necessity for customized processing approaches tailored to specific subjects and network architectures.

Stellar atmospheric parameters and chemical abundances of ~ 5 million stars from S-PLUS multiband photometry

The APOGEE, GALAH, and LAMOST spectroscopic surveys have substantially contributed to our understanding of the Milky Way by providing a wide range of stellar parameters and chemical abundances. Complementing these efforts, photometric surveys that include narrowband and medium-band filters, such as Southern Photometric Local Universe Survey (S-PLUS), provide a unique opportunity to estimate the atmospheric parameters and elemental abundances for a much larger number of sources, compared to spectroscopic surveys. Our aim is to establish methodologies for extracting stellar atmospheric parameters and selected chemical abundances from S-PLUS photometric data, which cover approximately $3000$ square degrees, by applying seven narrowband and five broadband filters. We used all 66 S-PLUS colors to estimate parameters based on three different training samples from the LAMOST, APOGEE, and GALAH surveys, applying cost-sensitive neural network (NN) and random forest (RF) algorithms. We kept the stellar abundances that lacked corresponding absorption features in the S-PLUS filters to test for spurious correlations in our method. Furthermore, we evaluated the effectiveness of the NN and RF algorithms by using estimated $T_ eff $ and \( g\) values as the input features to determine other stellar parameters and abundances. The NN approach consistently outperforms the RF technique on all parameters tested. Moreover, incorporating $T_ eff $ and \( g\) leads to an improvement in the estimation accuracy by approximately $3&lt;!PCT!&gt;$. We kept only parameters with a goodness-of-fit higher than $50&lt;!PCT!&gt;$. Our methodology allowed us to obtain reliable estimates for fundamental stellar parameters ($T_ eff g\), and Fe/H ) and elemental abundance ratios such as alpha /Fe Al/Fe C/Fe Li/Fe and Mg/Fe for approximately five million stars across the Milky Way, with a goodness-of-fit above $60&lt;!PCT!&gt;$. We also obtained additional abundance ratios, including Cu/Fe O/Fe and Si/Fe . However, these ratios should be used cautiously due to their low accuracy or lack of a clear relationship with the S-PLUS filters. Validation of our estimations and methods was performed using star clusters, Transiting Exoplanet Survey Satellite (TESS) data and Javalambre Photometric Local Universe Survey (J-PLUS) photometry, further demonstrating the robustness and accuracy of our approach. By leveraging S-PLUS photometric data and advanced machine learning techniques, we have established a robust framework for extracting fundamental stellar parameters and chemical abundances from medium-band and narrowband photometric observations. This approach offers a cost-effective alternative to high-resolution spectroscopy. The estimated parameters hold significant potential for future studies, particularly when classifying objects within our Milky Way or gaining insights into its various stellar populations.

Cross-Field Road Markings Detection Based on Inverse Perspective Mapping.

With the rapid development of the autonomous vehicles industry, there has been a dramatic proliferation of research concerned with related works, where road markings detection is an important issue. When there is no public open data in a field, we must collect road markings data and label them by ourselves manually, which is huge labor work and takes lots of time. Moreover, object detection often encounters the problem of small object detection. The detection accuracy often decreases when the detection distance increases. This is primarily because distant objects on the road take up few pixels in the image and object scales vary depending on different distances and perspectives. For the sake of solving the issues mentioned above, this paper utilizes a virtual dataset and open dataset to train the object detection model and cross-field testing in the field of Taiwan roads. In order to make the model more robust and stable, the data augmentation method is employed to generate more data. Therefore, the data are increased through the data augmentation method and homography transformation of images in the limited dataset. Additionally, Inverse Perspective Mapping is performed on the input images to transform them into the bird's eye view, which solves the "small objects at far distance" problem and the "perspective distortion of objects" problem so that the model can clearly recognize the objects on the road. The model testing on the front-view images and bird's eye view images also shows a remarkable improvement of accuracy by 18.62%.

Accurate detection of arbitrary ship directions using SAR based on RTMDet

ABSTRACT Ship detection, an important branch of Synthetic Aperture Radar (SAR) images interpretation, is crucial for various maritime reconnaissance and surveillance but remains challenging due to background interference, dense ship alignments and high aspect ratios. To address these challenges, this letter proposes an arbitrary direction detection method based on Real-Time Models for object Detection (RTMDet), which is one of the fastest and most accurate single-stage detection detectors. Firstly, an attention selection module is designed to replace traditional full convolution method to achieve more parameter reduction; Secondly, in order to extract more key ship feature information and accelerate the network extraction, we construct a lightweight multi-scale feature pyramid network; Thirdly, a novel loss function is introduced to enhance the detection of rotated bounding box by addressing the problem of bounding box discontinuities and lack of scale invariance. The proposed method achieves state-of-the-art detection accuracy on two the high-resolution SAR image datasets for arbitrary ship detection including HRSID (85.6% AP50) and RSDD (90.8% AP50), while demonstrating significant accuracy improvements in the complex inshore scenes.

EVALUATION OF ADAPTATION METHODS FOR THE GPT-4O LANGUAGE MODEL IN EDUCATIONAL ENVIRONMENTS

The adaptation of large language models, such as GPT-4O, for educational settings is becoming increasingly relevant, given their potential to improve automated support in learning environments. This study focuses on evaluating five distinct adaptation methods: the standard model (without adaptation), Fine-Tuning, Prompt Engineering, Retrieval-Augmented Generation (RAG), and agent-based systems. These methods were compared to find the most effective for enhancing GPT-4O’s performance in educational tasks. The analysis was conducted using key evaluation metrics such as precision, accuracy, F1-Score, average tokens per response, hallucination rate, and a final comprehensive score. The experimental results indicate that Fine-Tuning, which involves additional training on domain-specific educational data, offers the most significant improvements in terms of accuracy and reduction of erroneous responses (hallucinations). Fine-Tuning is especially effective in educational tasks requiring high accuracy and contextual understanding. Retrieval-Augmented Generation showed promising results by leveraging external data to enhance accuracy and lower hallucinations, making it suitable for tasks needing up-to-date information. Prompt Engineering provided faster response times but had more inaccuracies due to reliance on optimal query formulation without retraining. Agent-based systems excelled in handling complex tasks, though they showed a slight increase in hallucination rates due to their dynamic nature. The baseline performance of the standard GPT-4O model highlighted its limitations, like reduced accuracy and higher hallucination rates, especially in educational contexts. These findings underscore that Fine-Tuning is the most effective adaptation method for educational tasks, offering substantial improvements in accuracy and reliability. Overall, this research highlights the necessity of selecting the appropriate adaptation method based on the specific requirements of educational tasks. The study contributes to the ongoing optimization of large language models for use in educational environments, ensuring that responses are both reliable and contextually relevant.

Facial Detection and Recognition Analysis using Ontology-Driven Machine Learning Model

Facial detection and recognition technologies have achieved remarkable advancements through the integration of ontology-driven machine learning (ML). This study examines how ontology structured framework for representing domain-specific knowledge to enhance the robustness, accuracy, and interpretability of ML models, particularly Convolutional Neural Networks (CNNs), in the field of facial detection and recognition. Acknowledging the limitations of traditional ML methods, such as data inefficiency, inadequate generalization, and insufficient interpretability, this research addresses critical challenges, including semantic discrepancies, variations in environmental conditions, and ethical considerations. By leveraging ontology, the study aims to provide semantic enrichment, contextual adaptation, and data augmentation for ML models. A hybrid methodology is introduced that effectively integrates ontology with CNNs, in conjunction with the Viola-Jones and Eigenfaces algorithms, to improve performance in facial recognition tasks. The study utilizes comprehensive datasets such as CelebA and MegaFace, employing rigorous preprocessing and training processes that incorporate ontology-based feedback mechanisms. The results demonstrated significant improvements in accuracy, robustness, and explainability. Ontology-CNN models outperform traditional approaches in managing variations in facial attributes and environmental conditions. This study concludes that ontology-based frameworks present a promising avenue for developing efficient and ethically responsible facial recognition systems. Future research should focus on exploring the scalability of these models across diverse demographic contexts and further addressing ethical considerations related to privacy and fairness.

Safety and Accuracy of Guided Interradicular Miniscrew Insertion: A Systematic Review and Meta-Analysis.

Background: Achieving ideal anchorage is crucial in orthodontics for controlled tooth movement. Miniscrews (MSs) have improved skeletal anchorage, but freehand placement poses risks like root damage and limited precision. Guided techniques, including radiographic guides and computer-assisted methods (static [sCAS] and dynamic [dCAS]), were developed to enhance accuracy and safety. Objective: This systematic review and meta-analysis aimed to evaluate the safety and accuracy of MS placement using different guidance approaches. Materials: A systematic search up to March 2024 identified studies on guided MS insertion, assessing safety (root contact/damage) and accuracy (angular, coronal, and apical deviations) of guided vs. freehand placement. Two reviewers assessed the risk of bias and study quality using RoB 2 for RCTs, NOS for cohort studies, and an adapted tool for pre-clinical studies. Random-effects meta-analysis was performed for studies with common parameters, and safety outcomes were pooled using logit-transformed proportions. Heterogeneity was evaluated with I² and χ² tests. Results: Eleven studies (652 MSs) were included, though no dCAS studies were analyzed. The only RCT had "some concerns" regarding risk of bias, cohort studies ranged from medium to low quality, and most pre-clinical studies had high bias risk. sCAS significantly reduced root damage compared to freehand methods (OR = 0.11; 95% CI: 0.04-0.36; p < 0.001; I² = 1%) and reduced angular and linear deviations. Due to heterogeneity, no quantitative synthesis of accuracy outcomes was performed. Conclusions: sCAS improves the safety and accuracy of MS insertion compared to freehand and radiographic guide methods. These results highlight the clinical benefits of sCAS in orthodontics. Future studies should refine protocols and explore dCAS for further accuracy improvements.

Advanced Monocular Outdoor Pose Estimation in Autonomous Systems: Leveraging Optical Flow, Depth Estimation, and Semantic Segmentation with Dynamic Object Removal.

Autonomous technologies have revolutionized transportation, military operations, and space exploration, necessitating precise localization in environments where traditional GPS-based systems are unreliable or unavailable. While widespread for outdoor localization, GPS systems face limitations in obstructed environments such as dense urban areas, forests, and indoor spaces. Moreover, GPS reliance introduces vulnerabilities to signal disruptions, which can lead to significant operational failures. Hence, developing alternative localization techniques that do not depend on external signals is essential, showing a critical need for robust, GPS-independent localization solutions adaptable to different applications, ranging from Earth-based autonomous vehicles to robotic missions on Mars. This paper addresses these challenges using Visual odometry (VO) to estimate a camera's pose by analyzing captured image sequences in GPS-denied areas tailored for autonomous vehicles (AVs), where safety and real-time decision-making are paramount. Extensive research has been dedicated to pose estimation using LiDAR or stereo cameras, which, despite their accuracy, are constrained by weight, cost, and complexity. In contrast, monocular vision is practical and cost-effective, making it a popular choice for drones, cars, and autonomous vehicles. However, robust and reliable monocular pose estimation models remain underexplored. This research aims to fill this gap by developing a novel adaptive framework for outdoor pose estimation and safe navigation using enhanced visual odometry systems with monocular cameras, especially for applications where deploying additional sensors is not feasible due to cost or physical constraints. This framework is designed to be adaptable across different vehicles and platforms, ensuring accurate and reliable pose estimation. We integrate advanced control theory to provide safety guarantees for motion control, ensuring that the AV can react safely to the imminent hazards and unknown trajectories of nearby traffic agents. The focus is on creating an AI-driven model(s) that meets the performance standards of multi-sensor systems while leveraging the inherent advantages of monocular vision. This research uses state-of-the-art machine learning techniques to advance visual odometry's technical capabilities and ensure its adaptability across different platforms, cameras, and environments. By merging cutting-edge visual odometry techniques with robust control theory, our approach enhances both the safety and performance of AVs in complex traffic situations, directly addressing the challenge of safe and adaptive navigation. Experimental results on the KITTI odometry dataset demonstrate a significant improvement in pose estimation accuracy, offering a cost-effective and robust solution for real-world applications.

Improving the Performance of Electrotactile Brain-Computer Interface Using Machine Learning Methods on Multi-Channel Features of Somatosensory Event-Related Potentials.

Traditional tactile brain-computer interfaces (BCIs), particularly those based on steady-state somatosensory-evoked potentials, face challenges such as lower accuracy, reduced bit rates, and the need for spatially distant stimulation points. In contrast, using transient electrical stimuli offers a promising alternative for generating tactile BCI control signals: somatosensory event-related potentials (sERPs). This study aimed to optimize the performance of a novel electrotactile BCI by employing advanced feature extraction and machine learning techniques on sERP signals for the classification of users' selective tactile attention. The experimental protocol involved ten healthy subjects performing a tactile attention task, with EEG signals recorded from five EEG channels over the sensory-motor cortex. We employed sequential forward selection (SFS) of features from temporal sERP waveforms of all EEG channels. We systematically tested classification performance using machine learning algorithms, including logistic regression, k-nearest neighbors, support vector machines, random forests, and artificial neural networks. We explored the effects of the number of stimuli required to obtain sERP features for classification and their influence on accuracy and information transfer rate. Our approach indicated significant improvements in classification accuracy compared to previous studies. We demonstrated that the number of stimuli for sERP generation can be reduced while increasing the information transfer rate without a statistically significant decrease in classification accuracy. In the case of the support vector machine classifier, we achieved a mean accuracy over 90% for 10 electrical stimuli, while for 6 stimuli, the accuracy decreased by less than 7%, and the information transfer rate increased by 60%. This research advances methods for tactile BCI control based on event-related potentials. This work is significant since tactile stimulation is an understudied modality for BCI control, and electrically induced sERPs are the least studied control signals in reactive BCIs. Exploring and optimizing the parameters of sERP elicitation, as well as feature extraction and classification methods, is crucial for addressing the accuracy versus speed trade-off in various assistive BCI applications where the tactile modality may have added value.