Learning-based Methodology Research Articles

Performance evaluation of a machine learning-based methodology using dynamical features to detect nonwear intervals in actigraphy data in a free-living setting.

One challenge using wearable sensors is nonwear time. Without a nonwear (e.g., capacitive) sensor, actigraphy data quality can be biased by subjective determinations confounding sleep/wake classification. We developed and evaluated a machine learning algorithm supplemented by dynamic features to discern wear/nonwear episodes. Actigraphy data from wrist actigraph (Spectrum, Philips-Respironics). The built-in nonwear sensor as "ground truth" to classify nonwear periods using other data, mimicking features of Actiwatch 2. Data were collected over 1week from employed adults (n=853). Extreme gradient boosting (XGBoost), a tree-based classifier algorithm, was used to classify wear/nonwear, supplemented by dynamic features calculated over various time windows. The performance of the proposed algorithm was tested over 30-second epochs. Additional analytics and exploratory analyses: Evaluation of the SHapley Additive exPlanations (SHAP) values to find the effectiveness of the dynamic features. The XGBoost classifier yielded substantial improvements in balanced accuracy, sensitivity, and specificity, including dynamic features and comparison to default actiwatch classification algorithms. The proposed classifier effectively distinguished between valid and invalid days, and the duration of contiguous periods of nonwear correctly identified. Our findings highlight the potential of XGBoost using dynamic features of varying activity levels across the time series to provide insights on wear/nonwear classification using a large dataset. The methodology provides an alternative to laborious manual benchmarking of the data for similar devices that do not have a nonwear sensor.

Logical reasoning for human activity recognition based on multisource data from wearable device

Smart wearable devices detection and recording of people’s everyday activities is critical for health monitoring, helping persons with disabilities, and providing care for the elderly. Most of the research that is being conducted uses a machine learning-based methodology; however, these approaches frequently have issues with high computing resource consumption, burdensome training data gathering, and restricted scalability across many contexts. This research suggests a behaviour detection technology based on multi-source sensing and logical reasoning to address these problems. In order to realize the natural fusion of signal processing and logical reasoning in behavior recognition research, this work designs a lightweight behavior recognition solution using the pertinent theories of ontology reasoning in classical artificial intelligence. Machine learning technology is also employed for behavior recognition using the same data set. Once the best model has been chosen, the cross-person recognition results after testing and modification of parameters are 90.8% and 92.1%, respectively. This technology was used to create a behaviour recognition system, and several tests were run to assess how well it worked. The findings demonstrate that the suggested strategy achieves over 90% recognition accuracy for 11 different daily activities, including jogging, walking, and stair climbing. Additionally, the suggested strategy dramatically minimises the quantity of user-provided training data needed in comparison to machine learning-based behaviour identification techniques.

Classification and analysis of chronic health conditions influencing dementia progression using machine learning methods

Dementia, a progressive neurodegenerative condition, poses significant challenges to global healthcare systems due to its rising prevalence and complexity in diagnosis. This study employs a machine learning-based methodology to categorize dementia-related data and forecast disease progression, enabling early diagnosis and personalized treatment planning. Using a structured dataset comprising 24 features and 2230 entries, data preprocessing was conducted to address missing values, encode categorical variables, and normalize numerical features. Exploratory data analysis revealed key insights, including a strong inverse correlation between cognitive test scores and dementia severity. A Random Forest Classifier was trained to predict dementia presence, achieving an accuracy of 91.40%. Feature importance analysis identified "Cognitive_Test_Scores," "Prescription_Memantine," and "Dosage in mg" as the most critical predictors, aligning with clinical evidence. The confusion matrix highlighted high predictive accuracy but revealed misclassifications, particularly false negatives, which are critical in medical contexts. This study underscores the potential of machine learning in dementia diagnosis and management. By identifying significant predictors and achieving robust classification performance, the research provides a foundation for integrating artificial intelligence into healthcare systems to enhance diagnostic accuracy, optimize treatment plans, and improve patient outcomes. Further refinement and validation using larger datasets are recommended for broader applicability in clinical settings.

Developing a machine learning-based methodology for optimal hyperparameter determination—A mathematical modeling of high-pressure and high-temperature drilling fluid behavior

Drilling fluids exhibit complex rheological behavior due to a non-linear response to shear rate variations and high sensitivity to changes in temperature, time, and pressure conditions. The prediction of drilling fluid rheological behavior is crucial for the success of oil well drilling, and it directly impacts the fluid's performance. The dataset used in this study was obtained from extensive rheometric tests of water-based and olefin-based drilling fluids in steady-state flow curves. The optimal hyperparameters were guided by performance metrics and compared with alternative models such as Power-law and Herschel-Bulkley rheological models. Different configurations with different hidden layers, using neuron sequences of 16, 32, and 64, learning rates of 0.001 and 0.01, and the ReLU activation function were used to improve the model's performance. Additionally, the paper delved into the impact of the number of training epochs on the accuracy of shear stress predictions. Finding this equilibrium was identified as a crucial factor in achieving precise results. The neural network model demonstrated remarkable accuracy when using the ML-C3 configuration, with MAE values of 0.535 and R2 of 0.987 in predicting the steady-state flow curves of drilling fluids, establishing itself as a powerful tool for forecasting the rheological behavior of these fluids under diverse operational conditions. The present research significantly contributes to the field of drilling fluid rheology and provides valuable insights for optimizing drilling operations in HPHT environments.

Trusted microelectronics: reverse engineering chip die using U-Net convolutional network

In the field of integrated circuits (IC) reverse engineering, accurate IC die image segmentation is critical for ensuring trust and detecting counterfeits. This study introduces a deep learning-based methodology utilizing a U-Net Convolutional Neural Network (CNN) tailored for IC image segmentation. Our model excels in processing complex IC images with noise, achieving superior segmentation accuracy. The model was evaluated on a dataset of 512 × 512 pixel IC die images, achieving a mean Intersection over Union (IoU) of 81.2%, which is a significant improvement over traditional image processing techniques that achieve an IoU around 65.3%. During training, the model’s Dice Loss showed a sharp decrease, as depicted in the provided graph, highlighting the model’s ability to effectively learn and refine segmentation boundaries. Simultaneously, the training accuracy, as illustrated in the accompanying accuracy graph, improved steadily, reaching approximately 60% but still rising. This convergence of Dice Loss and the upward trend in accuracy demonstrate the model’s robust performance across varying noise levels and its effectiveness in producing precise segmentation outputs. This work underscores the effectiveness of CNNs, particularly the U-Net architecture, in enhancing the accuracy and reliability of IC die image analysis, paving the way for improved IC manufacturing quality assurance.

Video source identification using machine learning: A case study of 16 instant messaging applications

In recent years, there has been a notable increase in the prevalence of cybercrimes related to video content, including the distribution of illegal videos and the sharing of copyrighted material. This has led to the growing importance of identifying the source of video files to trace the owner of the files involved in the incident or identify the distributor. Previous research has concentrated on revealing the device (brand and/or model) that “originally” created a video file. This has been achieved by analysing the pattern noise generated by the image sensor in the camera, the storage structural features of the file, and the metadata patterns. However, due to the widespread use of mobile environments, instant messaging applications (IMAs) such as Telegram and Wire have been utilized to share illegal videos, which can result in the loss of information from the original file due to re-encoding at the application level, depending on the transmission settings. Consequently, it is necessary to extend the scope of existing research to identify the various applications that are capable of re-encoding video files in transit. Furthermore, it is essential to determine whether there are features that can be leveraged to distinguish them from the source identification perspective. In this paper, we propose a machine learning-based methodology for classifying the source application by extracting various features stored in the storage format and internal metadata of video files. To conduct this study, we analyzed 16 IMAs that are widely used in mobile environments and generated a total of 1974 sample videos, taking into account both the transmission options and encoding settings offered by each IMA. The training and testing results on this dataset indicate that the ExtraTrees model achieved an identification accuracy of approximately 99.96 %. Furthermore, we developed a proof-of-concept tool based on the proposed method, which extracts the suggested features from videos and queries a pre-trained model. This tool is released as open-source software for the community.

Evaluation of Hand-Crafted Feature Extraction for Fault Diagnosis in Rotating Machinery: A Survey.

This article presents a comprehensive collection of formulas and calculations for hand-crafted feature extraction of condition monitoring signals. The documented features include 123 for the time domain and 46 for the frequency domain. Furthermore, a machine learning-based methodology is presented to evaluate the performance of features in fault classification tasks using seven data sets of different rotating machines. The evaluation methodology involves using seven ranking methods to select the best ten hand-crafted features per method for each database, to be subsequently evaluated by three types of classifiers. This process is applied exhaustively by evaluation groups, combining our databases with an external benchmark. A summary table of the performance results of the classifiers is also presented, including the percentage of classification and the number of features required to achieve that value. Through graphic resources, it has been possible to show the prevalence of certain features over others, how they are associated with the database, and the order of importance assigned by the ranking methods. In the same way, finding which features have the highest appearance percentages for each database in all experiments has been possible. The results suggest that hand-crafted feature extraction is an effective technique with low computational cost and high interpretability for fault identification and diagnosis.

Penetration Testing and Attack Automation Simulation: Deep Reinforcement Learning Approach

In this research, we propose a revolutionary deep reinforcement learning-based methodology for automated penetration testing. The suggested method uses a deep Q-learning network to develop attack sequences that effectively exploit weaknesses in a target system. The method is tested in a virtual environment, and the findings indicate that it can identify vulnerabilities that manual penetration testing is unable to. A variety of tools, including Deep Q-learning network, MulVAL, Nmap, VirtualBox, Docker, National Vulnerability Database (NVD), and Common Vulnerability Scoring System (CVSS), are used in this work. The suggested method significantly outperforms current automated penetration testing methods. Our proposed methodology can detect flaws that manual penetration testing misses and can be modified (in terms of penalty values) to adapt to the updates of the target system (network) changes. Additionally, it has the potential to greatly enhance penetration testing's effectiveness and efficiency and could contribute to the increased security of computer systems. Experimental tests conducted in this work reveal the effectiveness of DQN automated penetration testing by utilizing the most effective attack vectors in the attack automation process

Prediction of spatter-related defects in metal laser powder bed fusion by analytical and machine learning modelling applied to off-axis long-exposure monitoring

Laser powder bed fusion of metals is increasingly used for fabricating complex parts requiring good mechanical properties. Simultaneously, researchers in the field are intensifying the efforts to reduce defects, such as internal porosities, which hinder a wider industrial adoption of this technology, urging process monitoring to a pivotal role in defect identification and mitigation. Therefore, understanding the correlation between in-process monitoring signals and post-process actual defects is fundamental to taking informed decisions and potential corrective actions during the process. This work focuses on developing models to predict spatter-related defects from specific process signatures detected through off-axis long-exposure imaging. Layer-wise images were properly aligned with corresponding cross-sections from tomographic reconstructions to investigate the relationship between spatter-related signatures and actual defects measured by X-ray computed tomography. This relationship was used as a knowledge basis to develop an analytical image-processing approach and a machine learning-based methodology, which were then compared in terms of their correlation performances. The advantages and limitations of both methods are discussed in the paper. Both approaches led to promising results in the prediction of lack-of-fusion defects caused by spatters, with the machine learning approach showing a prediction accuracy in the order of 90 % for defects with equivalent diameter above 90 µm, while the analytical model needed equivalent diameters larger than 130 µm to reach a prediction accuracy in the order of 80 %. Furthermore, the machine learning method led to strong results regarding early defect detection, with most of the investigated defects properly predicted by analysing two consecutive layers after the signature detection.

Applying Machine Learning Techniques: Uncertainty Quantification in Nonlinear Dynamics Characters Predictions via Gated Recurrent Unit-Based Reduced-Order Models

The development of reduced-order models has been a pivotal advancement in the computational analysis of fluid dynamics, substantially simplifying the complexity and boosting the efficiency of simulations. The accuracy and practicality of these models largely depend on the reduction techniques applied and the inherent characteristics of the fluid dynamics systems they represent. In this paper, we introduce an innovative machine-learning framework for assessing model uncertainty in computationally intensive reduced-order models. By combining subspace construction methods with advanced Bayesian inference techniques, our approach effectively captures the posterior distribution of model parameters, thereby providing an accurate representation of uncertainty in aerodynamic performance predictions. We employ the NACA0012 airfoil as a case study to validate our method’s ability to enhance the efficiency of reduced-order models and precisely measure the uncertainty inherent in predictions made by recurrent neural networks. It is important to note that our approach is influenced by specific constraints and variables that significantly impact the mean and variability of the predicted final lift coefficient distribution. Our findings indicate that setting the goodness of fit (R2) threshold above 0.985 markedly improves the correlation between Computational Fluid Dynamics (CFD) outcomes and model predictions, increasing from 72.2% to 97.9% as the interval amplification factor adjusts from 1.5 to 3. However, this adjustment causes a considerable expansion of the confidence interval, from 0.0737 to 0.1282, an increase of over 70%. Despite these challenges, our machine learning-based methodology provides essential insights into the further development of reduced-order modeling and uncertainty quantification in fluid dynamics. This highlights the need for ongoing research into the model parameters, especially in applications concerning aircraft control systems, to meet design specifications and ensure system reliability.

Setting the morphologic quality limits enabling accurate classification of charred archaeological grape seeds

This study investigates the morphological changes in grape pips resulting from various charring conditions. Employing high-resolution scanning combined with morphometric measurements for morphological analysis, we aimed to understand the effects of charring on grape pips. Our morphometric analysis demonstrated significant alterations in seed shape above 250 °C. The length–width ratio and the occurrence of cracks notably changed, providing a basis for assessing charring conditions. In addition, applying a machine learning classification method, we determined that accurate classification of grape varieties by the morphometric analysis method is feasible for seeds charred at up to 250 °C and 8 h. Integrating the morphometric changes and temperature ranges suitable for classification, we developed a sorting model for archaeological seeds. By projecting length–width ratios onto a curve calculated from controlled conditions, we estimated charring temperatures. Approximately 50% of archaeological seeds deviated from the model, indicating drastic charring conditions. This sorting model facilitates a stringent selection of seeds fit for classification, enhancing the accuracy of our machine learning-based methodology. In conclusion, combining machine learning with morphometric sorting enables the identification of charred grape seeds suitable for identification by the morphometric method. This comprehensive approach provides a valuable tool for future research for the identification of charred grape seeds found in archaeological contexts, enhancing our understanding of ancient viticulture practices and grape cultivation.

Corporate network anomaly detection methodology utilizing machine learning algorithms

ABSTRACT This study addresses the critical need for securing corporate networks against anomalies, a pressing concern in ensuring the comprehensive security of these networks. It aims to develop and validate a new machine learning-based methodology for anomaly detection that is adaptable across various corporate network environments, highlighting the method’s potential practical applications. Employing a systematic approach, the research integrates system analysis of anomaly detection methodologies with an analytical review of machine learning techniques tailored for high-security measures and attack prevention in corporate networks. This dual approach ensures a robust framework for identifying and addressing network anomalies efficiently. The methodology demonstrated notable efficacy, with the proposed machine learning-based anomaly detection techniques achieving an efficiency rate upwards of 90% in identifying and categorizing network traffic types. This high level of precision allows for the effective tracking of network anomalies across diverse corporate networks and their respective devices and equipment. The findings underscore the substantial practical value of the developed methodology, offering a promising avenue for enhancing corporate network security. The implementation of this machine learning-based approach not only facilitates the timely detection of anomalies but also significantly contributes to the improvement of machine learning applications within the realm of network security. Future research could further refine these techniques, exploring scalability and real-time data analysis enhancements to bolster their effectiveness across various network configurations.

Seismic fragility analysis of RC frame structures based on IDA analysis and machine learning

This study introduces a machine learning-based methodology designed to rapidly predict the seismic responses of reinforced concrete (RC) frame structures. The research focuses on three types of RC frame structures: low-rise, multi-story, and small high-rise buildings. Ground shaking records are selected according to the conditional mean spectrum (CMS). A sample database, constructed via Incremental Dynamic Analysis (IDA), facilitates the prediction of structural responses using ground shaking intensity and structural details as inputs. Concurrently, the study performs a feature importance analysis of the model. Machine learning algorithms, including integrated learning and neural networks, are utilized to predict the seismic responses of the RC frame structures. This methodology also assists in evaluating the seismic fragility of these structures. The results show that the discrepancy between the neural network-based seismic fragility assessments and the IDA results is minimal, indicating a high degree of accuracy in the proposed methodology. Among the characteristic parameters, Average Spectral Acceleration (AvgSa) is identified as the most significant. This methodology serves as a valuable tool for the rapid prediction of seismic responses in RC frame structures, demonstrating substantial practical value.

AI-based heart monitoring: the FIFPRO player pilot study

Abstract Background The implementation of remote global heart screening and monitoring services in professional players is essential as it allows: (1) efficient and rapid identification of cardiac abnormalities ensuring their safety and health; (2) standardizing the quality of medical heart checkups in low-resource regions to promote health equity; (3) prioritizing diversity in data collection to aid in the research and prevention of sudden cardiac death in sports. Purpose To demonstrate the feasibility of a global standardized heart screening program for professional footballers by leveraging an artificial intelligence platform to efficiently identify cardiac patterns from Holter’s records. In this study, we presented the most prevalent cardiac patterns, the patterns that prompted further assessment by the cardiologist, and an example of a typical research question exploring the effect of age on the prevalence of cardiac patterns. Methods The pilot study included 100 players from 9 different countries: Australia, Botswana, Bulgaria, Ghana, Greece, Indonesia, Panama, Spain and the UK. Holter data (Bittium Faros 180) were acquired in real-life monitoring from 24h to 48h including training sessions. Electrocardiograms were analyzed through an AI platform with a deep learning-based methodology to detect over 20 cardiac patterns. Medical team reviewed labels of detected cardiac patterns to validate and further train the AI algorithm. Thereafter cardiologists reviewed the cardiac patterns to determine whether further cardiac assessment was required. Results 145 sessions were recorded, with 100 players having one session and 45 players having two sessions. The mean age was 27.5 years (minimum 15, maximum 41), with 31 female players. The median duration of the Holter session was 25.1 hours. Table 1 presents the most frequently detected cardiac patterns: low density premature atrial complex, mild sinus bradycardia and low density premature ventricular complex in over 50% of the players. 18 players were identified with abnormalities requiring further study. 14 presented altered cardiac patterns (6 with negative T-waves, 5 with intraventricular conduction delay, 2 with QRS low voltage and 1 with P wave enlargement) and 3 presented premature ventricular complexes during exercise or high density. 1 was excluded because of inadequate clinical information. Regarding the effect of age, players with mild sinus bradycardia, premature ventricular complex, pause (2-3 seconds) or 2-degree atrioventricular block were found to be significantly older than players without those patterns (Figure 1). Conclusion This is the first global AI-based platform to obtain standardized heart screening program in professional athletes remotely. This study has been validated using diverse data from groups on 5 continents. This approach is beneficial to promote health equity in screening and monitoring of possible cardiomyopathies and channelopathies in order to prevent sudden cardiac death.Table 1Figure 1

Transmission Line Fault Classification Using Conformer Convolution-Augmented Transformer Model

Ensuring a consistently reliable power supply is paramount in power systems. Researchers are engaged in the pursuit of categorizing transmission line failures to design countermeasures for mitigating the associated financial losses. Our study employs a machine learning-based methodology, specifically the Conformer Convolution-Augmented Transformer model, to classify transmission line fault types. This model processes time series input data directly, eliminating the need for expert feature extraction. The training and validation datasets are generated through simulations conducted on a two-terminal transmission line, while testing is conducted on historical data consisting of 108 events that occurred in the Taiwan power system. Due to the limited availability of historical data, they are utilized solely for inference purposes. Our simulations are meticulously designed to encompass potential faults based on an analysis of historical data. A significant aspect of our investigation focuses on the impact of the sampling rate on input data, establishing that a rate of four samples per cycle is sufficient. This suggests that, for our specific classification tasks, relying on lower frequency data might be adequate, thereby challenging the conventional emphasis on high-frequency analysis. Eventually, our methodology achieves a validation accuracy of 100%, although the testing accuracy is lower at 88.88%. The discrepancy in testing accuracy can be attributed to the limited information and the small number of historical events, which pose challenges in bridging the gap between simulated data and real-world measurements. Furthermore, we benchmarked our method against the ELM model proposed in 2023, demonstrating significant improvements in testing accuracy.

Exploring the optimal design space of transparent perovskite solar cells for four-terminal tandem applications through Pareto front optimization

Machine learning algorithms can enhance the design and experimental processing of solar cells, resulting in increased conversion efficiency. In this study, we introduce a novel machine learning-based methodology for optimizing the Pareto front of four-terminal (4T) perovskite-copper indium selenide (CIS) tandem solar cells (TSCs). By training a neural network using the Bayesian regularization-backpropagation algorithm via Hammersley sampling, we achieve high prediction accuracy when testing with unseen data through random sampling. This surrogate model not only reduces computational costs but also potentially enhances device performance, increasing from 29.4% to 30.4% while simultaneously reducing material costs for fabrication by 50%. Comparing experimentally fabricated cells with the predicted optimal cells, the latter show a thinner front contact electrode, charge-carrier transport layer, and back contact electrode. Highly efficient perovskite cells identified from the Pareto front have a perovskite layer thickness ranging from 420 to 580 nm. Further analysis reveals the front contact electrode needs to be thin, while the back contact electrode can have a thickness ranging from 100 to 145 nm and still achieve high efficiency. The charge-carrier transport layers play a crucial role in minimizing interface recombination and ensuring unidirectional current flow. The optimal design space suggests thinner electron and hole transport layer thicknesses of 7 nm, down from 23 to 10 nm, respectively. It indicates a balanced charge-carrier extraction is crucial for an optimized perovskite cell. Overall, the presented methodology and optimized design parameters have the potential to enhance the performance of 4T perovskite/CIS TSC while reducing material fabrication costs.

Cutout as augmentation in contrastive learning for detecting burn marks in plastic granules

Abstract. Plastic granules are a common delivery form for creating products in industries such as the plastic manufacturing, construction and automotive ones. In the corresponding sorting process of plastic granules, diverse defect types could appear. Burn marks, which potentially lead to weakened structural integrity of the plastic, are one of the most common types. Thus, plastic granules with burn marks should be filtered out during the sorting process. Artificial intelligence (AI)-based anomaly detection approaches are widely used in the field of visual-based sorting due to the higher accuracy and lower requirement of expert knowledge compared with classic rule-based algorithms (Chandola et al., 2009). In this contribution, a simple data augmentation strategy, cutout, is implemented as a way of simulating defects when combined with a contrastive learning-based methodology and is proven to improve the accuracy of the anomaly detection of burn marks. Different variants of cutout are also evaluated. Specifically, synthetic image data are used due to the lack of real data.

Insights of effectivity analysis of learning-based approaches towards software defect prediction

Software defect prediction is one of the essential sets of operation towards mitigating issues of risk management in software development known to contribute towards enhancing the quality of software. There is evolution of various methodologies towards resolving this issue while learning-based methodology is witnessed to be the most dominant contributor. The problem identified is that there are yet many unsolved queries associated with practical viability of such learning-based approach adoption in software quality management. Proposed approaches discussed in this paper contributes towards mitigating this challenge by introducing a simplified, compact, and crisp analysis of effectiveness associated with learning-based schemes. The paper presents its major findings of effectivity analysis of machine learning, deep learning, hybrid, and other miscellaneous approaches deployed for fault prediction followed by highlighting research trend. The major findings infer that feature selection, data imbalance, interpretability, and in adequate involvement of context are prime gaps in existing methods. The paper also contributes towards research gap as well as essential learning outcomes of present review work.

Machine-learning Algorithm-based Risk Prediction and Screening-detected Prostate Cancer in A Benign Prostate Hyperplasia Cohort.

Prostate cancer (PCa) is lethal. Our aim in this retrospective cohort study was to use machine learning-based methodology to predict PCa risk in patients with benign prostate hyperplasia (BPH), identify potential risk factors, and optimize predictive performance. The dataset was extracted from a clinical information database of patients at a single institute from January 2000 to December 2020. Patients newly diagnosed with BPH and prescribed alpha blockers/5-alpha-reductase inhibitors were enrolled. Patients were excluded if they had a previous diagnosis of any cancer or were diagnosed with PCa within 1 month of enrolment. The study endpoint was PCa diagnosis. The study utilized the extreme gradient boosting (XGB), support vector machine (SVM) and K-nearest neighbors (KNN) machine-learning algorithms for analysis. The dataset used in this study included 5,122 medical records of patients with and without PCa, with 19 patient characteristics. The SVM and XGB models performed better than the KNN model in terms of accuracy and area under curve. Local interpretable model-agnostic explanation and Shapley additive explanations analysis showed that body mass index (BMI) and late prostate-specific antigen (PSA) were important features for the SVM model, while PSA velocity, late PSA, and BMI were important features for the XGB model. Use of 5-alpha-reductase inhibitor was associated with a higher incidence of PCa, with similar survival outcomes compared to non-users. Machine learning can enhance personalized PCa risk assessments for patients with BPH but more research is necessary to refine these models and address data biases. Clinicians should use them as supplementary tools alongside traditional screening methods.

Classification of Medicinal Plant Leaves for Types and Diseases with Hybrid Deep Learning Methods

Leaf images are often used to detect plant diseases because most disease symptoms appear on the leaves. Analyzes performed by experts in the laboratory environment are expensive and time consuming. Therefore, there is a need for automated plant disease detection systems that are both economical and can help diagnose early symptoms more accurately. In this study, a deep learning-based methodology is presented for the classification of leaf diseases of plants, which are very similar in color, texture, vein and shape and cannot be noticed by non-experts, which are important for traditional medicine and pharmaceutical industry. In the model development process, 7 pre-learning deep learning algorithms and an image data set created from plant leaves in 10 categories were preferred. The proposed model classifies the plant type and diseased condition in the dataset. In the first step of training the model, different learning rates were tested with optimum hyperparameters. In the second part, a test accuracy rate of 98.69% was achieved with the DenseNet121 model, with increased data. At the last stage, after the edge detection processes, the test accuracy value of 67.92% was reached with the DenseNet 121 model.