Specificity Metrics Research Articles

A multi-population-based marine predators algorithm to train artificial neural network

Marine predators algorithm (MPA) is one of the recently proposed metaheuristic algorithms. In the MPA, position update mechanisms are implemented, emphasizing global search in the first part of the search process, balanced search in the middle, and local search in the last part. This may adversely affect the local search capability of the algorithm in the first part of the search process and the global search capability in the last part of the search process. To overcome these issues, an algorithm called MultiPopMPA with a multi-population and multi-search strategy is proposed in this study. Thanks to the proposed algorithm, local, balanced, and global search strategies of the original MPA were utilized from the beginning to the end of the search process. Thus, it is aimed to contribute to a more detailed search of the parameter space. In this study, the proposed algorithm has been applied in training artificial neural networks for 21 different classification datasets. The success of the algorithm has been scored on precision, sensitivity, specificity, and F1-score metrics and compared with eight different metaheuristic algorithms, including the original MPA. In terms of the mean rank of success, the proposed MultiPopMPA has been ranked first in precision, sensitivity, and F1-score metrics and ranked second in the specificity metric. In addition, it has been observed that the proposed algorithm outperforms its competitors in most cases in terms of convergence and stability. Finally, Wilcoxon’s signed-rank test results calculated through the MSE metric showed that the proposed algorithm produced statistically significant results in most cases.

Computational Predictions of MGMT Promoter Methylation in Gliomas: A Mathematical Radiogenomics Approach

In the treatment of glioblastomas, non-invasive methods for determining MGMT gene promoter methylation status are crucial due to their implications for chemotherapy responsiveness. This study utilizes a mathematical and computational framework to extract and analyze radiogenomic data from MRI images to predict the methylation status. The first step in our framework involves extracting radiogenomic data from MRI images. This process requires sophisticated image processing techniques to convert MRI scans into a format suitable for machine learning analysis. The features extracted include textural patterns, intensity distributions, and other relevant radiomic characteristics.To identify the most significant features, we employ a Random Forest (RF) algorithm. Mathematically, RF is an ensemble learning method that operates by constructing multiple decision trees during training and outputting the mode of the classes for classification tasks. The importance of each feature is evaluated based on its contribution to the accuracy of the model, quantified by metrics such as Gini impurity or information gain. Employing advanced machine learning models like VGG19, ResNet50, and Sequential FCNet Artificial Neural Networks, alongside traditional classifiers such as Naive Bayes and Logistic Regression, we analyze features identified by a Random Forest algorithm. Our mathematical approach ensures rigorous evaluation of model accuracy through sensitivity, specificity, and other performance metrics, presenting the Sequential FCNet ANN combined with ResNet50 as the superior model. This research contributes to the field of precision healthcare by enhancing the mathematical methods used in the non-invasive diagnosis of glioblastomas.

PD-DETECTOR: A sustainable and computationally intelligent mobile application model for Parkinson's disease severity assessment

This paper introduces a mobile cloud-based predictive model for assisting Parkinson's disease (PD) patients. PD, a chronic neurodegenerative disorder, impairs motor functions and daily tasks due to the degeneration of dopamine-producing neurons in the brain. The model utilizes smartphones to aid patients in collecting voice samples, which are then sent to a cloud service for storage and processing. A hybrid deep learning model, trained using the UCI Parkinson's Telemonitoring Voice dataset, analyzes this data to estimate the severity of PD symptoms. The model's performance is noteworthy, with accuracy, sensitivity, and specificity metrics of 96.2 %, 94.15 %, and 96.15 %, respectively. Additionally, it boasts a rapid response time of just 13 s. Results are delivered to users via smartphone alert notifications, coupled with a knowledge base feature that educates them about PD. This system provides reliable home-based assessment and monitoring of PD and enables prompt medical intervention, significantly enhancing the quality of life for patients with Parkinson's disease.

Choledochal cancer region detection in hyperspectral images using U-Net based models

Cholangiocarcinoma (CCA) is a type of cancer that forms in the bile duct that carry digestive fluid from the liver. CCA is the primary form of liver cancer that affects population ranging from age 60 to 69 years. CCA is difficult to diagnose at an early stage. Hyperspectral (HS) imaging is an advanced imaging technique that combines spectroscopy with conventional imaging. HS imaging is an emerging field of study which can be used for early CCA detection. HS imaging involves capturing images across various spectral bands, which forms a three-dimensional data cube often called as hyperspectral data cube. In this study, we have utilized U-Net based models, namely U-Net and DenseUNet were used to perform semantic segmentation on the HS images of CCA tissues. A band selective approach was employed to derive a subset of meaningful bands based on the spectrum plot from the HS image. The HS images are further preprocessed with Principal Component Analysis (PCA). The models were further evaluated by computing the accuracy, AUC (Area under the ROC curve), sensitivity and specificity metrics. The proposed models, namely, U-Net and DenseUNet reported an overall accuracy of 73.47% and 77.09% respectively. The DenseUNet models outperforms the U-Net model on every evaluation metric. The proposed models were also compared with other state-of-the-art (SOTA) models trained on various HS dataset. This study explores the application of HS imaging in carcinoma detection. The findings of this study could be used for further enhancement of the approach.

Patterns in parasite diversity and interactions with anurans from the Atlantic Forest

Knowledge of parasite-host interactions is essential for understanding factors associated with the ecology and evolution of both groups. Some aspects, such as host size and phylogeny, as well as parasite specificity, are significant predictors that help unveil the parasite-host relationship. Thus, the goals of this study were: (1) to describe parasite diversity in regions of the Atlantic Forest; (2) to analyze which host characteristics can influence parasite richness of anuran's parasite component community; and (3) to investigate if the prevalence of parasite infection is related to specificity metrics (ecological and phylogenetic), number of infected hosts and parasite's abundance. We identified 49 parasite taxa, classified into three phyla: Nematoda, Acanthocephala, and Platyhelminthes. Supporting the existing literature, our findings corroborate the positive relationship between host size and parasite richness, further emphasizing the significance of this predictor. Parasite prevalence in the host community is related to the number of infected host species and parasite abundance, but not to phylogenetic and ecological specificity indices. This shows that parasite prevalence is strongly associated with infection opportunity, host sampling effort, and high parasite abundance.

Development of an intrusion detection system using mayfly feature selection and artificial neural network algorithms

Protecting the privacy and confidentiality of information and devices in computer networks requires reliable methods of intrusion detection. However, effective intrusion detection is made more difficult by the enormous dimensions of data available in computer networks. To boost intrusion detection classification performance in computer networks, this study developed a feature selection mode for the classification task. The proposed model utilized the Mayfly feature selection algorithm and ANN as the classifiers. The model was also tested without a mayfly algorithm. The model's efficacy was determined through a comparison of its accuracy, specificity, precision, sensitivity, and F1 score. The experimental outcomes revealed that the proposed model is more efficient than existing models based on the performance evaluation and the CIC-IDS 2017 dataset employed in this research. Accuracy scores of 99.94% (using Data+mayfly+ANN) and 90.17% (using Data+ANN) were attained after experimentation. In comparison to existing models, the proposed model yielded better results in terms of accuracy, sensitivity, specificity, and F1-score metrics. The model's sturdiness can be attributed to the use of mayfly techniques, which harness the strength in PSO, GA and FA for selecting optimal feature subsets. The results of this research provide a reliable dimensionality reduction model that may be used in the field of computer networks for intrusion detection and enhancement of security in computer networking environments.

Deep Learning Based System for of Grading Cashew Kernels

Cashew nuts are cultivated widely across the globe and are categorized into various classes according to their size and quality. Currently, cashew kernels are sorted and graded manually, which is time-consuming and labor-intensive. This research proposes hybrid models to automate and accelerate the cashew classification process by integrating classifiers with deep convolutional neural networks (DCNNs). The proposed hybrid models categorize cashew kernels into five groups: W180, W210, W300, W400, and W500. Various hybrid models, including RCNN + OpenCV, were implemented and evaluated based on specificity, accuracy, and other metrics. The results demonstrated that the RCNN model paired with OpenCV achieved a max- imum accuracy of 90%. This study’s findings highlight that automating cashew grading can be significantly enhanced by combining DCNNs with classifiers in hybrid models. These hybrid models improve efficiency by enabling automatic, effective, and precise classification of cashews. By leveraging advanced machine learning techniques, the process becomes faster, more accurate, and less dependent on manual labor. Overall, the integration of deep learning technologies with traditional classifiers offers a promising approach to modernizing and optimizing the cashew grading industry. Future research could further refine these hybrid models to achieve even higher accuracy and efficiency, potentially revolutionizing agricultural sorting processes beyond cashew nuts. Index Terms—DCNN, OpenCV, RCNN, Hybrid Models

Utilization of artificial intelligence in minimally invasive right adrenalectomy: recognition of anatomical landmarks with deep learning

Background The primary surgical approach for removing adrenal masses is minimally invasive adrenalectomy. Recognition of anatomical landmarks during surgery is critical for minimizing complications. Artificial intelligence-based tools can be utilized to create real-time navigation systems during laparoscopic and robotic right adrenalectomy. In this study, we aimed to develop deep learning models that can identify critical anatomical structures during minimally invasive right adrenalectomy. Methods In this experimental feasibility study, intraoperative videos of 20 patients who underwent minimally invasive right adrenalectomy in a tertiary care center between 2011 and 2023 were analyzed and used to develop an artificial intelligence-based anatomical landmark recognition system. Semantic segmentation of the liver, the inferior vena cava (IVC), and the right adrenal gland were performed. Fifty random images per patient during the dissection phase were extracted from videos. The experiments on the annotated images were performed on two state-of-the-art segmentation models named SwinUNETR and MedNeXt, which are transformer and convolutional neural network (CNN)-based segmentation architectures, respectively. Two loss function combinations, Dice-Cross Entropy and Dice-Focal Loss were experimented with for both of the models. The dataset was split into training and validation subsets with an 80:20 distribution on a patient basis in a 5-fold cross-validation approach. To introduce a sample variability to the dataset, strong-augmentation techniques were performed using intensity modifications and perspective transformations to represent different surgery environment scenarios. The models were evaluated by Dice Similarity Coefficient (DSC) and Intersection over Union (IoU) which are widely used segmentation metrics. For pixelwise classification performance, accuracy, sensitivity and specificity metrics were calculated on the validation subset. Results Out of 20 videos, 1000 images were extracted, and the anatomical landmarks (liver, IVC, and right adrenal gland) were annotated. Randomly distributed 800 images and 200 images were selected for the training and validation subsets, respectively. Our benchmark results show that the utilization of Dice-Cross Entropy Loss with the transformer-based SwinUNETR model achieved 78.37%, whereas the CNN-based MedNeXt model reached a 77.09% mDSC score. Conversely, MedNeXt reaches a higher mIoU score of 63.71% than SwinUNETR by 62.10% on a three-region prediction task. Conclusion Artificial intelligence-based systems can predict anatomical landmarks with high performance in minimally invasive right adrenalectomy. Such tools can later be used to create real-time navigation systems during surgery in the near future.

Tumor origin identification through machine learning and gene expression profiling.

e13597 Background: Accurate identification of tumor origin is crucial for effective diagnosis and treatment, particularly in cases of metastatic tumors. Interpretable machine learning models have displayed significant potential in addressing this problem when imaging and immunohistochemistry (IHC) examinations are ineffective. In this study, we developed a panel using gene expression profiles from various tumors and constructed a robust machine learning model for precise identification of tumor origin. Methods: RNA sequencing (RNA-seq) data of 9462 tumor samples originating from 21 different organs were collected from The Cancer Genome Atlas (TCGA). We conducted feature engineering through unsupervised clustering and differential gene expression analysis, selecting a refined panel of 164 genes from a pool of over 60,000 identifiers. Subsequently, a machine learning classifier grounded in Logistic regression (LR) was trained on 9462 samples with the constructed 164-gene panel. To enhance the adaptability of our model to this task, 10-fold cross-validation was employed in the multi-class mode. Two independent test sets, the Primary tumor set (PT, n=3420, including 19 tumor types) and the Metastatic tumor set (MTP, n=100, all originating from the prostate and spread to bone, liver, etc.), were established with samples from Gene Expression Omnibus (GEO) and other published studies. Notably, all samples underwent FPKM normalization and there was no overlap between training and testing samples. Model performance was assessed using accuracy, specificity, and sensitivity metrics. Results: The 164-gene expression panel achieved a cross-validation accuracy of 96.73%. In the assessment of the PT test set, our model achieved an overall 99.62% specificity, 92.31% accuracy and 89.79% sensitivity, exhibiting performance comparable to similar models in other studies. Additionally, our model accurately traced 91% of metastatic tumors in the MTP test set to the prostate, surpassing previous lines of work by a large margin. Conclusions: Gene expression patterns reveal organ-specific characteristics that could be used to identify tumor origin. The combination of a condensed yet comprehensive gene expression panel with a robust machine learning model serves as a promising tool for tumor diagnosis. Ongoing correlative studies aim to extend predictions from tissue organs to cancer subtypes.

An Ensemble Classification Model to Predict Alzheimer’s Incidence as Multiple Classes

This study introduces an ensemble classification model designed to categorize Alzheimer’s disease (AD) into four distinct classes—mild dementia, no dementia, moderate dementia, and very mild dementia—using Magnetic Resonance Imaging (MRI). The proposed model entitled the Ensemble Classification Model to Predict Alzheimer's Incidence as Multiple Classes (PAIMC) that integrates a six-dimensional analysis of MR images, encompassing entropies, Fractal Dimensions, Gray Level Run Length Matrix (GLRLM), Gray Level Co-occurrence Matrix (GLCM), morphological features, and Local Binary Patterns. A four-fold multi-label cross-validation approach was employed on a benchmark dataset to evaluate the model's performance. Quantitative analysis reveals that PAIMC consistently achieves superior Decision Accuracy, F-Score, Specificity, Sensitivity Recall, and Precision metrics compared to existing state-of-the-art models. For instance, PAIMC's Decision Accuracy and Precision outperform the second-best model by a notable margin across all folds. The model also demonstrates a significant improvement in Sensitivity Recall and Specificity, reinforcing its efficacy in the multi-class classification of AD stages. A novel data diversity assessment measure was developed and utilized, further confirming the robustness of the PAIMC model. The results underscore the potential of PAIMC as a highly accurate tool for AD classification in clinical settings.

Accuracy of artificial intelligence-assisted endoscopy in the diagnosis of gastric intestinal metaplasia: A systematic review and meta-analysis.

Gastric intestinal metaplasia is a precancerous disease, and a timely diagnosis is essential to delay or halt cancer progression. Artificial intelligence (AI) has found widespread application in the field of disease diagnosis. This study aimed to conduct a comprehensive evaluation of AI's diagnostic accuracy in detecting gastric intestinal metaplasia in endoscopy, compare it to endoscopists' ability, and explore the main factors affecting AI's performance. The study followed the PRISMA-DTA guidelines, and the PubMed, Embase, Web of Science, Cochrane, and IEEE Xplore databases were searched to include relevant studies published by October 2023. We extracted the key features and experimental data of each study and combined the sensitivity and specificity metrics by meta-analysis. We then compared the diagnostic ability of the AI versus the endoscopists using the same test data. Twelve studies with 11,173 patients were included, demonstrating AI models' efficacy in diagnosing gastric intestinal metaplasia. The meta-analysis yielded a pooled sensitivity of 94% (95% confidence interval: 0.92-0.96) and specificity of 93% (95% confidence interval: 0.89-0.95). The combined area under the receiver operating characteristics curve was 0.97. The results of meta-regression and subgroup analysis showed that factors such as study design, endoscopy type, number of training images, and algorithm had a significant effect on the diagnostic performance of AI. The AI exhibited a higher diagnostic capacity than endoscopists (sensitivity: 95% vs. 79%). AI-aided diagnosis of gastric intestinal metaplasia using endoscopy showed high performance and clinical diagnostic value. However, further prospective studies are required to validate these findings.

Comparison of different 2D muscle indexes measured at the level of the 3rd lumbar vertebra in survival prediction in patients with renal cell carcinoma.

Low computed tomography (CT)-determined muscle mass, commonly determined with height-adjusted muscle indexes (MIs), predicts worse survival in several cancers and has been suggested as a prognostic assessment tool. Although several MIs measured at the level of the 3rd lumbar vertebra (L3) are commonly used, it remains unestablished how different L3-determined MIs perform in survival prognostication compared to each other. The objective of this study was to investigate the performance of different MIs for survival prognostication in renal cell carcinoma (RCC). We retrospectively enrolled 214 consecutive patients with RCC. We determined three L3-MIs (psoas muscle index (PMI), psoas muscle index and erector spinae index (PMI+ESI), and whole skeletal muscle index (SMI)) from preoperative CT scans. Categorization of those with low and normal muscle mass was based on the Youden Index sex-specific MI cut-offs. We determined sensitivity, specificity, and accuracy metrics for predicting 1-year, 5-year, and overall survival (OS) using Cox regression models. Low PMI, PMI+ESI, and SMI significantly predicted decreased 1-year, 5-year, and OS in uni- and multivariate models. PMI+ESI and SMI were more accurate than PMI in males, and PMI and PMI+ESI were more accurate than SMI in females in the prediction of 1-year survival. However, there were no differences in accuracies between MIs in 5-year and OS prediction. PMI+ESI performed well overall in short-term prognostication, but there were no differences between the MIs in long-term prognostication. We recommend the use of PMI+ESI for muscle evaluation, particularly when SMI cannot be evaluated.

Integration of single-cell transcriptomic and chromatin accessibility on heterogenicity of human peripheral blood mononuclear cells utilizing microwell-based single-cell partitioning technology

Abstract Single-cell RNA sequencing (scRNA-Seq) deepens our understanding of cellular development and heterogeneity. However, limitations exist in unraveling cell states and gene regulatory programs. Chromatin state profiles assess gene expression potential and offer insights into transcriptional regulation. Integrated with gene expression data, chromatin accessibility region (CAR) profiles establish fundamental gene regulatory logic for cell fate. ATAC-seq (Assay for Transposase-Accessible Chromatin using Sequencing) is a highly potent approach for profiling genome-wide CARs. To investigate the power of the multiomics assay in identifying differentiated gene regulations, we conducted multiomic snATAC-seq+ snRNA-seq on PBMCs from two different donors, by utilizing the gentle and robust microwell-based single-cell partitioning technology. The assay showed high sensitivity and specificity metrics (>10,000 median unique fragments/cell, >0.7 fragments in peak score). Integrative analysis across donors revealed enriched transcription factor motifs and fragment coverage tracks in distinct cell types, correlating significantly with gene expression data. These findings highlight mRNA's intricate connections with CARs in immune cell development. Our study underscores the power of multiomic analysis in analyzing heterogeneity of PBMC cell populations and offers a toolkit to identify gene regulation specific to diverse cell types, enhancing our comprehension of epigenetic diversity.

Machine Learning-driven Lung Cancer Detection

Machine Learning-driven Lung Cancer Detection

P150 Diagnostic utility of anti-cN1A autoantibody testing in sporadic inclusion body myositis

Abstract Background/Aims Sporadic inclusion body myositis (sIBM) is the most common acquired myopathy in patients aged over 50 years. Diagnostic criteria encompass clinical, biochemical, and myopathological features. Certain myopathological features possess high specificity but low sensitivity; reliance on such findings may lead to delayed or misdiagnoses. The discovery of autoantibodies against cytosolic 5'-Nucleotidase 1A (cN1A) in sIBM patients suggests an autoimmune component in pathogenesis, but the inclusion of this autoantibody in diagnostic criteria remains uncertain. Additionally, optimal testing methodology has not been established, with varying techniques (e.g., ELISA/line-blot) in use. We assessed antibody profiles in an sIBM cohort and reviewed diagnoses in anti-cN1A positive patients. Methods This study was based at the Northern Care Alliance NHS Foundation Trust which hosts a tertiary neuromuscular referral service for the North-West of England. Data sources included myositis clinic and immunology results databases. All patients tested using a validated myositis line-blot immunoassay (EUROLINE 16 Antigen et cN1A [EUROIMMUN]) November 2021-April 2023 were included. Diagnoses of all tested patients were identified through case-note review. Sensitivity and specificity metrics were calculated based on diagnosis and anti-cN1A antibody status. Results The total number of patients tested for anti-cN1A was 495. Of these, 10 patients had a diagnosis of sIBM. The true-positive and false-negative rates were both 50% (n = 5 each). Of the 485 patients who did not have sIBM, true-negativity was 97.52% (n = 473) and false-positivity 2.47% (n = 12). Overall, anti-cN1A had a sensitivity of 50%, specificity of 97.52%, positive predictive value of 29.41% and negative predictive value of 98.95% for sIBM (Table 1). Patients with a false-positive anti-cN1A antibody had diagnoses of: primary Raynaud’s (n = 5); undifferentiated CTD (n = 3); systemic sclerosis (n = 2); non-sIBM myositis (n = 2). Conclusion Anti-cN1A detected using the EUROLINE 16 Antigen assay has high specificity but lacks sensitivity for sIBM, consistent with previous literature. These findings suggest that this assay performs similarly to other cN1A detection methodologies and has utility in confirming the diagnosis of sIBM where there is clinical suspicion. Further real-world analysis of antibody data in larger cohorts is required to clarify the role of serological testing in sIBM diagnostic criteria, to improve diagnostic delays and misclassifications. Disclosure M. Al-Attar: None. T. Khoo: None. J.B. Lilleker: None. H. Chinoy: None.

An Overview on the Advancements of Support Vector Machine Models in Healthcare Applications: A Review

Support vector machines (SVMs) are well-known machine learning algorithms for classification and regression applications. In the healthcare domain, they have been used for a variety of tasks including diagnosis, prognosis, and prediction of disease outcomes. This review is an extensive survey on the current state-of-the-art of SVMs developed and applied in the medical field over the years. Many variants of SVM-based approaches have been developed to enhance their generalisation capabilities. We illustrate the most interesting SVM-based models that have been developed and applied in healthcare to improve performance metrics on benchmark datasets, including hybrid classification methods that combine, for instance, optimization algorithms with SVMs. We even report interesting results found in medical applications related to real-world data. Several issues around SVMs, such as selection of hyperparameters and learning from data of questionable quality, are discussed as well. The several variants developed and introduced over the years could be useful in designing new methods to improve performance in critical fields such as healthcare, where accuracy, specificity, and other metrics are crucial. Finally, current research trends and future directions are underlined.

Utility of Global Registry of Acute Coronary Event Score (GRACE) and Thrombolysis in Myocardial Infarction Risk Score (TIMI Score) in predicting 30-day Mortality in patients undergoing primary angioplasty at a Tertiary care center – A record-based study

Background This research work aims to assess the effectiveness of Global Registry Of Coronary Events (GRACE) and Thrombolysis In Myocardial Infarction (TIMI) risk scores in predicting outcomes for patients who underwent primary Percutaneous Coronary Intervention (PCI). The study included a cohort from a tertiary care centre. Methods This record-based study enrolled 94 patients who underwent primary PCI after a diagnosis of ST segment elevation myocardial infarction (STEMI). The data, for this study, was extracted from sources including, medical records. These records contained the computation and documentation of both GRACE and TIMI scores. The study also conducted Receiver Operating Curve (ROC) curve analysis, subsequently yielding the area under the curve alongside a 95% confidence interval, which was duly reported. Results The current study attempts a comparative analysis of the capacities of TIMI and GRACE scores. Accordingly, the study highlights that both are equally effective in predicting 30-day mortality for patients with STEMI. The TIMI Risk Score, with 85% sensitivity and 99% specificity metrics (also p-value = 0.008), and the GRACE Score, with 85% sensitivity and 99% specificity metrics (also p-value = 0.006), exhibit a strong similarity in the prognostic performance. Also, the sensitivity and specificity metrics were obtained after a thorough analysis of the receiver operating characteristic curve (ROCC). Conclusions The study utilized the TIMI score which demonstrated a sensitivity of 85% and specificity of 99% and the GRACE Score with 85% sensitivity and 99% specificity to predict 30-day mortality in STEMI patients. This included more than one month of observation period. Post observations, the results of the experiment suggested an improvement in the left ventricular functionwhich helped to infer that recovery after primary PCI is gradual rather than an immediate and pronounced recovery.

Dynamic Object Detection in Surveillance Videos using Temporal Convolutional Networks and Federated Learning in Edge Computing Environments

This research addresses the importance of advancing dynamic object detection in surveillance videos by introducing a novel framework that integrates Temporal Convolutional Networks (TCNs) and Federated Learning (FL) within edge computing environments. This research is motivated by the critical need for real-time threat response, enhanced security measures, and privacy preservation in dynamic surveillance scenarios. Leveraging TCNs, the system captures temporal dependencies, providing a comprehensive understanding of object movements. FL ensures decentralized model training, mitigating privacy concerns associated with centralized approaches. Current challenges in real-time processing, privacy preservation, and adaptability to dynamic environments are addressed through innovative solutions. Model optimization techniques optimize TCN efficiency, ensuring real-time processing. Advanced privacy-preserving mechanisms secure FL, addressing privacy concerns. Transfer learning and data augmentation enhance adaptability to dynamic scenarios. The proposed system not only addresses existing challenges but also contributes to the evolution of surveillance technology. Comprehensive metrics, including accuracy, sensitivity, specificity, and real-time processing metrics, provide a thorough evaluation of the system's performance. This research introduces an approach to dynamic object detection, combining TCN and FL in edge computing environments. Results show accuracy exceeding 97%, sensitivity and specificity at 97% and 98%, and F1 score reaching 96%.

Deep learning-based cervical lesion segmentation in colposcopic images

Artificial intelligence assisted cancer detection has changed the ream of diagnosis precision. This study aims to propose a segmentation network using artificial intelligence for accurately segmenting the cervix region and acetowhite lesions in cervigram images, addressing the shortage of skilled colposcopists and streamlining the training process. A computational approach is employed to develop and train a deep learning model specifically tailored for cervix region and acetowhite lesion segmentation in cervigram images. A dataset acquired in collaboration with KIDWAI memorial cancer research institute is used for building the model. Cervigram images are collected for training and validation, and a deep learning architecture is constructed and trained using annotated datasets. The segmentation network based on efficientnet architecture and atrous spatial pyramid pooling is designed to accurately identify and delineate the target regions, with performance evaluation conducted using precision, accuracy, recall, dice score, and specificity metrics. The proposed segmentation network achieves a precision of 0.7387±0.1541, accuracy of 0.9291, recall of 0.7912±0.1439, dice score of 0.7431±0.1506, and specificity of 0.9589±0.0131, indicating its reliability and robustness in segmenting cervix regions and acetowhite lesions in cervigram images. This research demonstrates the feasibility and effectiveness of using artificial intelligence-based computational models for cervix region and acetowhite lesion segmentation in cervigram images. It provides a foundation for further investigations into classifying cervix malignancy using AI techniques, potentially enhancing early detection and treatment of cervical cancer while addressing the shortage of skilled professionals in the field

Hyperparameter optimization approaches to improve the performance of machine learning models for cardiovascular risk prediction

Machine learning algorithms have been used in diverse areas among applications, including healthcare. However, to fit an effective and optimal machine learning model, the hyperparameters need to be tuned. This process is commonly referred to as Hyperparameter Optimization and comprises several approaches. We combined three Hyperparameter Optimization techniques (Bayesian Optimization, Particle Swarm Optimization, and Genetic Algorithm) with three classifiers (Random Forest, Support Vector Machine, and XGBoost) to identify the best combination of hyperparameters that maximize model performance. We use the Framingham dataset to test the proposal. For classifier performance, the Support Vector Machine obtained the best result in recall (96.40%) and F-score (93.86%), while XGBoost obtained the best result in precision (96.30%) and specificity (96.36%). In the accuracy metric, both classifiers achieved 95%. Bayesian optimization had the best results in terms of accuracy, precision, specificity, and F-score metrics. Both Particle Swarm Optimization and Genetic Algorithm obtained the best result in the recall metric.