Performance Of Machine Learning Algorithms Research Articles

Robust Human Activity Recognition for Intelligent Transportation Systems Using Smartphone Sensors: A Position-Independent Approach

This study explores Human Activity Recognition (HAR) using smartphone sensors to address the challenges posed by position-dependent datasets. We propose a position-independent system that leverages data from accelerometers, gyroscopes, linear accelerometers, and gravity sensors collected from smartphones placed either on the chest or in the left/right leg pocket. The performance of traditional machine learning algorithms (Decision Trees (DT), K-Nearest Neighbors (KNN), Random Forest (RF), Support Vector Classifier (SVC), and XGBoost) is compared against deep learning models (Gated Recurrent Unit (GRU), Long Short-Term Memory (LSTM), Temporal Convolutional Networks (TCN), and Transformer models) under two sensor configurations. Our findings highlight that the Temporal Convolutional Network (TCN) model consistently outperforms other models, particularly in the four-sensor non-overlapping configuration, achieving the highest accuracy of 97.70%. Deep learning models such as LSTM, GRU, and Transformer also demonstrate strong performance, showcasing their effectiveness in capturing temporal dependencies in HAR tasks. Traditional machine learning models, including RF and XGBoost, provide reasonable performance but do not match the accuracy of deep learning models. Additionally, incorporating data from linear accelerometers and gravity sensors led to slight improvements over using accelerometer and gyroscope data alone. This research enhances the recognition of passenger behaviors for intelligent transportation systems, contributing to more efficient congestion management and emergency response strategies.

Heart Failure Prediction Using Machine Learning

This study explores the use of XGBoost and Shapley values for heart failure prediction, in addition to processing models such as neural network, CNN, and SVM. By comparing the performance of these different machine learning algorithms, we aim to develop a robust and accurate predictive model for identifying individuals at risk of heart failure. Our findings highlight the strengths and limitations of each approach, providing valuable insights for future research in this area. Ultimately, this research aims to improve the early detection and management of heart failure, leading to better patient outcomes and more efficient healthcare practices. Key Words: heart failure prediction, XGBoost, SHAPley values, neural network, CNN, SVM, machine learning algorithms, early detection, death events.

Optimized Machine Learning Algorithms for Predicting the Punching Shear Resistance of Flat Slabs with Transverse Reinforcement

In the calculation of reinforced concrete (RC) flat slabs with transverse reinforcement, punching shear resistance is one of the most critical factors. It is true that design provisions may be implemented, but they often result in significant biases and deviations from expectations. This study aims to present an optimized machine learning (ML) algorithm for estimating the punching shear resistance. Four machine learning (ML) algorithms (SVR, DT, RF, and XGBoost) with Bayesian optimization (BO) are presented in this paper to provide accurate predictions for flat slabs. The adoptability and optimization of the models are achieved through the analysis of a database of 337 test specimens with nine design parameters. Machine learning (ML) techniques are used to estimate punching shear resistance, which is compared with design provisions and equations relating to critical shear crack theory (CSCT). According to this study, Bayesian optimization is still capable of improving the performance of conventional machine learning algorithms, while the XGBoost-based model offers advanced capabilities. Predictions based on BO-XGBoost are in good agreement with actual values (MAE, RMSE, and R2 are 0.09 MN, 0.14 MN, and 0.92, respectively) in test set. Following a detailed explanation using Shapley additive explanation (SHAP), a high-performance ML approach is used to investigate the predictive results. With the proposed optimized algorithms, it is possible to determine the punching shear resistance of flat slabs with transverse reinforcement during the preliminary stages of the construction.

Sparse discriminant manifold projections for automatic depression recognition

In recent years, depression has become an increasingly serious problem globally. Previous research have shown that EEG-based depression recognition is a promising technique to serve as auxiliary diagnosis methods that provide assistance to clinicians. Typically, in clinical studies, due to the multichannel nature of EEG, the extracted features usually are high-dimensional and contain many redundant information. Therefore, it is necessary to perform dimensionality reduction before classification to improve the performance of machine learning algorithms. However,existing dimensionality reduction techniques do not design the objective function based on the characteristics of EEG signal and the goal of depression recognition, so they are less suitable for dimensionality reduction of EEG features. To solve this problem, in this paper we propose a novel dimensionality reduction technique called sparse discriminant manifold projections(SDMP) for depression recognition. Specifically, the use of the ℓ2-norm instead of the squared ℓ2-norm as a similarity measure in the objective function reduces sensitivity to noise and outliers. Moreover, the local geometric structure and global discriminative properties of data are integrated, which makes the extracted features more discriminative. Finally, the ℓ2,1-norm regularization is introduced to achieve feature selection. Furthermore, The formulation is extended to the ℓ2,p-norm regularization case, which is more likely to offer better sparsity when 0<p<1. Extensive experiments on EEG data show that the SDMP achieves the competitive performance compared with other state-of-the-art dimensionality reduction methods. It also shows the practical application value of our method in detecting depression.

Forecasting Patient Early Readmission from Irish Hospital Discharge Records Using Conventional Machine Learning Models

Background/Objectives: Predicting patient readmission is an important task for healthcare risk management, as it can help prevent adverse events, reduce costs, and improve patient outcomes. In this paper, we compare various conventional machine learning models and deep learning models on a multimodal dataset of electronic discharge records from an Irish acute hospital. Methods: We evaluate the effectiveness of several widely used machine learning models that leverage patient demographics, historical hospitalization records, and clinical diagnosis codes to forecast future clinical risks. Our work focuses on addressing two key challenges in the medical fields, data imbalance and the variety of data types, in order to boost the performance of machine learning algorithms. Furthermore, we also employ SHapley Additive Explanations (SHAP) value visualization to interpret the model predictions and identify both the key data features and disease codes associated with readmission risks, identifying a specific set of diagnosis codes that are significant predictors of readmission within 30 days. Results: Through extensive benchmarking and the application of a variety of feature engineering techniques, we successfully improved the area under the curve (AUROC) score from 0.628 to 0.7 across our models on the test dataset. We also revealed that specific diagnoses, including cancer, COPD, and certain social factors, are significant predictors of 30-day readmission risk. Conversely, bacterial carrier status appeared to have minimal impact due to lower case frequencies. Conclusions: Our study demonstrates how we effectively utilize routinely collected hospital data to forecast patient readmission through the use of conventional machine learning while applying explainable AI techniques to explore the correlation between data features and patient readmission rate.

Classification of Vaginal Cleanliness Grades through Surface‐Enhanced Raman Spectral Analysis via The Deep‐Learning Variational Autoencoder–Long Short‐Term Memory Model

In this study, it is aimed to establish a novel method based on a deep‐learning‐guided surface‐enhanced Raman spectroscopy (SERS) technique to achieve rapid and accurate classification of vaginal cleanliness levels. We proposed a variational autoencoder (VAE) approach to enhance spectral quality, coupled with a deep learning algorithm long short‐term memory (LSTM) neural network to analyze SERS spectra produced by vaginal secretions. The performance of various machine learning (ML) algorithms is assessed using multiple evaluation metrics. Finally, the reliability of the optimal model is tested using blind test data (N = 10/group for each cleanliness level). The data quality of the SERS fingerprints of four types of vaginal secretions is significantly improved after VAE decoding and reconstruction. The signal‐to‐noise ratio of the generated spectra increased from the original 2.58–11.13. Among all algorithms, the VAE–LSTM algorithm demonstrates the best prediction ability and time efficiency. Additionally, blind test datasets yielded an overall accuracy of 85%. In this study, it is concluded that the deep‐learning‐guided SERS technique holds significant potential in rapidly distinguishing between different levels of vaginal cleanliness through human vaginal secretion samples. This contributes to the efficient diagnosis of vaginal cleanliness levels in clinical settings.

Interpretable machine learning model for predicting long-term cardiovascular mortality risk in chronic kidney disease patients using SHAP explainers

Abstract Background Patients with chronic kidney disease (CKD) exhibited a pronounced burden of cardiac mortality. However, early prediction of cardiovascular mortality risk in CKD patients remains an unmet medical need. Purpose This study aimed to evaluate the performance of machine learning (ML) algorithms and identify the most influential factors in predicting long-term cardiac mortality among CKD patients. Methods Six ML models were developed, with the most effective one chosen for predicting and categorising patients into high-risk and low-risk groups based on the maximal Youden’s index. Differences in survival rates were assessed using the log-rank test on Kaplan-Meier curves. Cox regression analysis was employed to investigate the relationship between ML-predicted risk scores and mortality. Furthermore, the SHapley Additive exPlanations (SHAP) method was implemented to provide personalised explanations for model decisions. Results The auto-encoder (AE) model achieved the highest AUC of 0.94, with a sensitivity of 80.95% and specificity of 96.34%. Compared to individuals in low-risk group, those in high-risk group exhibited a significantly elevated risk of cardiovascular mortality (adjusted hazard ratio [HR]: 48.24; P &amp;lt;0.001). According to the SHAP model, age, C-reactive protein, blood urea nitrogen and hypertension were identified as the four most influential factors in the AE model. Conclusions Our study successfully developed an AE machine learning model for predicting 10-year cardiovascular mortality among CKD patients. The combination of ML model and SHAP aids doctors in understanding key features in the model, improving their insight into decision-making related to disease severity assessment, thus facilitating early preventive strategies in clinical practice.

Data-Driven Approaches for State-of-Charge Estimation in Battery Electric Vehicles Using Machine and Deep Learning Techniques

One of the most important functions of the battery management system (BMS) in battery electric vehicle (BEV) applications is to estimate the state of charge (SOC). In this study, several machine and deep learning techniques, such as linear regression, support vector regressors (SVRs), k-nearest neighbor, random forest, extra trees regressor, extreme gradient boosting, random forest combined with gradient boosting, artificial neural networks (ANNs), convolutional neural networks, and long short-term memory (LSTM) networks, are investigated to develop a modeling framework for SOC estimation. The purpose of this study is to improve overall battery performance by examining how BEV operation affects battery deterioration. By using dynamic response simulation of lithium battery electric vehicles (BEVs) and lithium battery packs (LIBs), the proposed research provides realistic training data, enabling more accurate prediction of SOC using data-driven methods, which will have a crucial and effective impact on the safe operation of electric vehicles. The paper evaluates the performance of machine and deep learning algorithms using various metrics, including the R2 Score, median absolute error, mean square error, mean absolute error, and max error. All the simulation tests were performed using MATLAB 2023, Anaconda platform, and COMSOL Multiphysics.

Development and evaluation of a novel framework to enhance k-NN algorithm’s accuracy in data sparsity contexts

This paper presents a novel framework for implementing the k-NN algorithm, designed to enhance its accuracy in contexts with sparse data. The framework addresses limitations in the algorithm’s training process by optimizing data structures. It employs composite datasets generated from the initial data using a data-driven fuzzy Analytic Hierarchy Process weighting scheme. This approach is designed to enhance the informational content in the initial datasets, thus reducing the entropy and implementation uncertainty. The framework was evaluated using 75 publicly available datasets and 3 generated datasets, demonstrating significant accuracy improvements across various k-parameter values. The findings were rigorously generalized using non-parametric hypothesis tests; while the resulting sensitivity was assessed by applying different distance metrics. By enhancing informational content, the composite data structures contribute to both accuracy improvements and scalability, particularly in data-sparse contexts. This relationship underscores the critical role of entropy in enhancing the performance of explainable machine learning algorithms, providing a valuable and interpretable tool for transforming data structures in sparse data environments.

Analysis of Machine Learning-Based NLOS Signal Identification Algorithm for UWB Indoor Localization Using CIR Waveform Features

Abstract. Ultra-wideband (UWB) technology stands out among numerous indoor positioning techniques due to its high operating frequency, low interception capability, resistance to multipath effects, and strong penetration. The UWB uses the time-of-arrival (TOA) to estimate the distance between the transmitter and receiver anchors in centimeter accuracy. However, in complex indoor positioning environments, obstacles such as walls, glass windows, metal plates, and wooden doors may block and reflect signals, inevitably causing non-line-of-sight (NLOS) errors that significantly affect positioning accuracy. The NLOS signal has lower signal energy due to the reflections. Thus, the channel impulse responses (CIR) from NLOS and LOS are different. To address the NLOS signal identification issue in UWB positioning, we utilize UWB CIR data collected from various positioning scenarios as the data source. CIR waveform input features are provided for the NLOS signal recognition model, and four machine learning models—Support Vector Machine (SVM), Multi-Layer Perceptron (MLP), K-Nearest Neighbors (KNN), and XGBoost—are trained and optimized for NLOS signal recognition. The aim of the study is to analyze the performance of different machine learning algorithms for NLOS signal recognition in UWB indoor localization using these features. Experimental results indicate that machine learning-based NLOS signal recognition algorithms can achieve an accuracy of approximately 77.46%, precision of 80.46%, and an F1 score of 0.81. Among the four models, the XGBoost model demonstrates generally better recognition performance.

Pattern Recognition in the Processing of Electromyographic Signals for Selected Expressions of Polish Sign Language.

Gesture recognition has become a significant part of human-machine interaction, particularly when verbal interaction is not feasible. The rapid development of biomedical sensing and machine learning algorithms, including electromyography (EMG) and convolutional neural networks (CNNs), has enabled the interpretation of sign languages, including the Polish Sign Language, based on EMG signals. The objective was to classify the game control gestures and Polish Sign Language gestures recorded specifically for this study using two different data acquisition systems: BIOPAC MP36 and MyoWare 2.0. We compared the classification performance of various machine learning algorithms, with a particular emphasis on CNNs on the dataset of EMG signals representing 24 gestures, recorded using both types of EMG sensors. The results (98.324% versus ≤7.8571% and 95.5307% versus ≤10.2697% of accuracy for CNNs and other classifiers in data recorded with BIOPAC MP36 and MyoWare, respectively) indicate that CNNs demonstrate superior accuracy. These results suggest the feasibility of using lower-cost sensors for effective gesture classification and the viability of integrating affordable EMG-based technologies into broader gesture recognition frameworks, providing a cost-effective solution for real-world applications. The dataset created during the study offers a basis for future studies on EMG-based recognition of Polish Sign Language.

Sitetack: a deep learning model that improves PTM prediction by using known PTMs.

Post-translational modifications (PTMs) increase the diversity of the proteome and are vital to organismal life and therapeutic strategies. Deep learning has been used to predict PTM locations. Still, limitations in datasets and their analyses compromise success. We evaluated the use of known PTM sites in prediction via sequence-based deep learning algorithms. For each PTM, known locations of that PTM were encoded as a separate amino acid before sequences were encoded via word embedding and passed into a convolutional neural network that predicts the probability of that PTM at a given site. Without labeling known PTMs, our models are on par with others. With labeling, however, we improved significantly upon extant models. Moreover, knowing PTM locations can increase the predictability of a different PTM. Our findings highlight the importance of PTMs for the installation of additional PTMs. We anticipate that including known PTM locations will enhance the performance of other proteomic machine learning algorithms. Sitetack is available as a web tool at https://sitetack.net; the source code, representative datasets, instructions for local use, and select models are available at https://github.com/clair-gutierrez/sitetack.

IMOABC: An efficient multi-objective filter–wrapper hybrid approach for high-dimensional feature selection

With the development of data science, the challenge of high-dimensional data has become increasingly prevalent. High-dimensional data contains a significant amount of redundant information, which can adversely affect the performance and effectiveness of machine learning algorithms. Therefore, it is necessary to select the most relevant features from the raw data and perform feature selection on high-dimensional data. In this paper, a novel filter–wrapper feature selection method based on an improved multi-objective artificial bee colony algorithm (IMOABC) is proposed to address the feature selection problem in high-dimensional data. This method simultaneously considers three objectives: feature error rate, feature subset ratio, and distance, effectively improving the efficiency of obtaining the optimal feature subset on high-dimensional data. Additionally, a novel Fisher Score-based initialization strategy is introduced, significantly enhancing the quality of solutions. Furthermore, a new dynamic adaptive strategy is designed, effectively improving the algorithm’s convergence speed and enhancing its global search capability. Comparative experimental results on microarray cancer datasets demonstrate that the proposed method significantly improves classification accuracy and effectively reduces the size of the feature subset when compared to various traditional and state-of-the-art multi-objective feature selection algorithms. IMOABC improves the classification accuracy by 2.27% on average compared to various multi-objective feature selection methods, while reducing the number of selected features by 88.76% on average. Meanwhile, IMOABC shows an average improvement of 4.24% in classification accuracy across all datasets, with an average reduction of 76.73% in the number of selected features compared to various traditional methods.

Advanced Landslide Detection Using Machine Learning and Remote Sensing Data

Landslides can cause severe damage to infrastructure and human life, making early detection and warning systems critical for mitigating their impact. In this study, we propose a machine learning approach for landslide detection using remote sensing data and topographical features. We evaluate the performance of several machine learning algorithms, including Tree, Random Forest, Gradient Boosting, Logistic Regression, Naïve Bayes, AdaBoost, Neural Network, SGD, kNN, and SVM, on a dataset of remote sensing images and topographical features from the Sikkim region in Malaysia. The results show that the SVM algorithm outperforms the other algorithms with an accuracy of 96.7% and a F1 score of 0.97. The study demonstrates the potential of machine learning algorithms for landslide detection, which can help improve early warning systems and reduce the impact of landslides.

Racial and Ethnic Disparities in Predictive Accuracy of Machine Learning Algorithms Developed Using a National Database for 30-Day Complications Following Total Joint Arthroplasty

BackgroundWhile predictive capabilities of machine learning (ML) algorithms for hip and knee total joint arthroplasty (TJA) have been demonstrated in previous studies, their performance in racial and ethnic minority patients has not been investigated. This study aimed to assess the performance of ML algorithms in predicting 30-day complications following TJA in racial and ethnic minority patients. MethodsA total of 267,194 patients undergoing primary TJA between 2013 and 2020 were identified from a national outcomes database. The patient cohort was stratified according to race, with further sub-stratification into Hispanic or non-Hispanic ethnicity. There were two ML algorithms, histogram-based gradient boosting (HGB), and random forest (RF), that were modeled to predict 30-day complications following primary TJA in the overall population. They were subsequently assessed in each racial and ethnic subcohort using discrimination, calibration, accuracy, and potential clinical usefulness. ResultsBoth models achieved excellent (Area under the curve (AUC) > 0.8) discrimination (AUCHGB = AUCRF = 0.86), calibration, and accuracy (HGB: slope = 1.00, intercept = -0.03, Brier score = 0.12; RF: slope = 0.97, intercept = 0.02, Brier score = 0.12) in the non-Hispanic White population (N = 224,073). Discrimination decreased in the White Hispanic (N = 10,429; AUC = 0.75 to 0.76), Black (N = 25,116; AUC = 0.77), Black Hispanic (N = 240; AUC = 0.78), Asian non-Hispanic (N = 4,809; AUC = 0.78 to 0.79), and overall (N = 267,194; AUC = 0.75 to 0.76) cohorts, but remained well-calibrated. We noted the poorest model discrimination (N = 1,870; AUC = 0.67 to 0.68) and calibration in the American-Indian cohort. ConclusionThe ML algorithms demonstrate an inferior predictive ability for 30-day complications following primary TJA in racial and ethnic minorities when trained on existing healthcare big data. This may be attributed to the disproportionate underrepresentation of minority groups within these databases, as demonstrated by the smaller sample sizes available to train the machine learning models. The ML models developed using smaller datasets (e.g., in racial and ethnic minorities) may not be as accurate as larger datasets, highlighting the need for equity-conscious model development.

Predicting paediatric asthma exacerbations with machine learning: a systematic review with meta-analysis

BackgroundAsthma exacerbations in children pose a significant burden on healthcare systems and families. While traditional risk assessment tools exist, artificial intelligence (AI) offers the potential for enhanced prediction models.ObjectiveThis study aims to systematically evaluate and quantify the performance of machine learning (ML) algorithms in predicting the risk of hospitalisation and emergency department (ED) admission for acute asthma exacerbations in children.MethodsWe performed a systematic review with meta-analysis, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The risk of bias and applicability for eligible studies was assessed according to the prediction model study risk of bias assessment tool (PROBAST). The protocol of our systematic review was registered in the International Prospective Register of Systematic Reviews.ResultsOur meta-analysis included seven articles encompassing a total of 17 ML-based prediction models. We found a pooled area under the curve (AUC) of 0.67 (95% CI 0.61–0.73; I2=99%; p&lt;0.0001 for heterogeneity) for models predicting ED admission, indicating moderate accuracy. Notably, models predicting child hospitalisation demonstrated a higher pooled AUC of 0.79 (95% CI 0.76–0.82; I2=95%; p&lt;0.0001 for heterogeneity), suggesting good discriminatory power.ConclusionThis study provides the most comprehensive assessment of AI-based algorithms in predicting paediatric asthma exacerbations to date. While these models show promise and ML-based hospitalisation prediction models, in particular, demonstrate good accuracy, further external validation is needed before these models can be reliably implemented in real-life clinical practice.

Comparative Analysis of Machine Learning Models and Hybrid Ensemble Approach’s for Landslide Prediction

The NH-44 Jammu Srinagar National Highway in India is susceptible to landslides, rock falls and shooting stones due to its geological characteristics and steep slopes. This study aims to compare the performance of various Machine Learning (ML) algorithms and hybrid models in predicting landslides using historical data. Seven optimized ML approaches Support Vector Classifier (SVC), Logistic Regression (LR), Decision Tree (DT), K Nearest Classifier (KNC), Random Forest (RF), GaussianNB (GNB) and AdaBoost Classifier (ABC) are used. Additionally, two Hybrid Ensemble methods, the Voting Hybrid Model (VHM) and Stacking Hybrid Model (SHM), are introduced. The performance of each model is evaluated using accuracy, precision, recall, F1-score and AUC metrics. Results indicate that all the models performed well, with hybrid ensemble models surpassing all individual algorithms. The Stacking Hybrid Model (SHM) excels achieving 99.4% accuracy and 98.5% AUC, outperforming the Voting Hybrid Model (VHM). Hybrid models consistently outperform individual models in accuracy and AUC. These proposed methods exhibit robustness and enhanced results in addressing this issue. This framework can help to predict the landslides with high accuracy which can save the lives through timely evacuations from high-risk areas.

Comparative analysis of machine learning techniques for detecting potability of water

This research aims to evaluate the effectiveness of machine learning algorithms in determining the potability of water. In the study, a total of 3276 water samples were analyzed for 10 different features that determine the potability of water. Besides that, the study's consideration is to evaluate the impact of trimming, IQR, and percentile methods on the performance of machine learning algorithms. The models were built using nine different classification algorithms (Logistic Regression, Decision Trees, Random Forest, XGBoost, Naive Bayes, K-Nearest Neighbors, Support Vector Machine, AdaBoost, and Bagging Classifier). According to the results, filling the missing data with the population mean and handling outliers with Trimming and IQR methods improved the performance of the models. Random Forest and Decision Tree algorithms were the most accurate in determining the potability of water. The findings of this research are of high importance to sustainable water resource management and serve as a crucial input for the decision-making process on the quality of water. The study also offers an example for researchers working on datasets that contain missing values and outliers.

Enhancement of The Performance of Machine Learning Algorithms to Rival Deep Learning Algorithms in Predicting Stock Prices.

This paper objectives to improve stock market prediction accuracy by training data on sentiment analysis of tweets, overcoming volatility and complexity. Utilizing the Use of natural language processing (NLP) algorithms, the tweet's sentiments were classified into (negative - neutral - and positive). The stock value price was predicted by implementing Machine learning algorithms (KNN, SVM, GBM, LR, DT, RF, EL). Among the techniques of ML, (GBM) achieved the greatest results in terms of accuracy (96%). Its results were compared with the results of a deep learning algorithm that uses the same data where GBM got better results, and other algorithms showed results (KNN = 55%, SVM=90%, LR=82%, DT=90%, RF=90%, EL = 88%). The results obtained were superior to previous studies.

Comparative Performance of Autoencoders and Traditional Machine Learning Algorithms in Clinical Data Analysis for Predicting Post-Staged GKRS Tumor Dynamics.

Introduction: Accurate prediction of tumor dynamics following Gamma Knife radiosurgery (GKRS) is critical for optimizing treatment strategies for patients with brain metastases (BMs). Traditional machine learning (ML) algorithms have been widely used for this purpose; however, recent advancements in deep learning, such as autoencoders, offer the potential to enhance predictive accuracy. This study aims to evaluate the efficacy of autoencoders compared to traditional ML models in predicting tumor progression or regression after GKRS. Objectives: The primary objective of this study is to assess whether integrating autoencoder-derived features into traditional ML models can improve their performance in predicting tumor dynamics three months post-GKRS in patients with brain metastases. Methods: This retrospective analysis utilized clinical data from 77 patients treated at the "Prof. Dr. Nicolae Oblu" Emergency Clinic Hospital-Iasi. Twelve variables, including socio-demographic, clinical, treatment, and radiosurgery-related factors, were considered. Tumor progression or regression within three months post-GKRS was the primary outcome, with 71 cases of regression and 6 cases of progression. Traditional ML models, such as Logistic Regression, Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Extra Trees, Random Forest, and XGBoost, were trained and evaluated. The study further explored the impact of incorporating features derived from autoencoders, particularly focusing on the effect of compression in the bottleneck layer on model performance. Results: Traditional ML models achieved accuracy rates ranging from 0.91 (KNN) to 1.00 (Extra Trees). Integrating autoencoder-derived features generally enhanced model performance. Logistic Regression saw an accuracy increase from 0.91 to 0.94, and SVM improved from 0.85 to 0.96. XGBoost maintained consistent performance with an accuracy of 0.94 and an AUC of 0.98, regardless of the feature set used. These results demonstrate that hybrid models combining deep learning and traditional ML techniques can improve predictive accuracy. Conclusion: The study highlights the potential of hybrid models incorporating autoencoder-derived features to enhance the predictive accuracy and robustness of traditional ML models in forecasting tumor dynamics post-GKRS. These advancements could significantly contribute to personalized medicine, enabling more precise and individualized treatment planning based on refined predictive insights, ultimately improving patient outcomes.