Bisphosphonate-related osteonecrosis of the jaw (BRONJ) is a rare yet serious condition that affects patients undergoing bisphosphonate therapy for osteoporosis and malignant bone diseases. Accurate classification of BRONJ stages is essential for early diagnosis and optimal treatment planning. Traditional diagnostic methods rely on subjective clinical assessments and radiographic analysis, which can lead to inconsistencies and delays in treatment. This study proposes an AI-driven classification model for BRONJ to address these challenges using machine learning (ML) and deep learning (DL) techniques. The research evaluates the performance of Support Vector Machines (SVM), Random Forest (RF), and Multilayer Perceptron (MLP) on a dataset containing demographic information, clinical symptoms, laboratory results, and patient medical history. To mitigate class imbalance, Synthetic Minority Over-sampling Technique (SMOTE) and class merging strategies were applied. The optimized MLP model achieved an accuracy of 88.24%, improving generalization through regularization, dropout layers, and hyperparameter tuning. The results demonstrate the feasibility of ML and DL models for BRONJ classification, offering a more objective and automated approach to disease staging. The proposed model has the potential to reduce diagnostic variability, improve risk stratification, and assist clinicians in decision-making. By leveraging AI techniques, this study paves the way for more efficient and standardized BRONJ diagnosis, ultimately contributing to better patient outcomes and enhanced clinical workflows.

In the context of the ageing of the global population and the increasing prevalence of Alzheimer's disease (AD), early and accurate diagnosis is crucial for effective management and treatment. Using Exploratory Data Analysis (EDA) we dissect the complex relationships between various risk factors and disease progression, establishing a basis for our predictive modelling. Uncovering critical insights, and emphasizing the importance of adopting a multidimensional approach to analyze diverse datasets effectively, our study highlights the critical role of data quality and diversity in improving model performance. The fundamental aspect of our analysis focuses on the predictive power of combining different data types, which traditionally include clinical parameters, genetic markers, and demographic and lifestyle data.The research highlights the application of machine learning (ML) techniques for early detection and predictive analysis of Alzheimer's disease, demonstrating the enormous potential of artificial inelegance in transforming healthcare diagnostics. The study conducted a comparative analysis of various ML algorithms and evaluated their efficiency in disease detection.This research contributes to the academic discourse on the diagnosis of Alzheimer's disease and provides practical insights for the application of artificial intelligence and machine learning in clinical practice.

Nowadays, huge amounts of data are generated experimentally in systems biology as well in clinics and other healthcare and medical institutions. This has resulted in the emergence of new concepts: big data and NoSQL databases that are becoming more popular and promising especially for analyzing complex interactions that exist in biological networks. These heterogeneous and voluminous data, which are usually semi-structured or unstructured, highly connected and unpredictable, need to be integrated and stored properly. With the growth of data size and data complexity, NoSQL databases have outperformed traditional relational databases for analysis, access and querying. Particularly, to represent various complex relationships among entities in physics, in biological, social and computer networks, graph-based databases are very suitable. These databases are appropriate to represent, store and query heavily interconnected data, especially for large-scale network data that exist in biology. This paper describes the biological and other networks in biomedicine and surveys the most widely used graph databases and software tools and packages and their properties. Additionally, the portraying, analyzing and querying of biological networks are described. The analysis of the biological networks results in gaining very significant insights into the relevant information for the biological processes, such as diseases, interactions and regulatory mechanisms that occur in biological networks, as well as studying their properties.

Riboswitch, a part of regulatory mRNA (50-250nt in length), has two main classes: aptamer and expression platform. One of the main challenges raised during the classification of riboswitch is imbalanced data. That is a circumstance in which the records of a sequences of one group are very small compared to the others. Such circumstances lead classifier to ignore minority group and emphasize on majority ones, which results in a skewed classification. We considered sixteen riboswitch families, to be in accord with recent riboswitch classification work, that contain imbalanced sequences. The sequences were split into training and test set using a newly developed pipeline. From 5460 k-mers (k value 1 to 6) produced, 156 features were calculated based on CfsSubsetEval and BestFirst function found in WEKA 3.8. Statistically tested result was significantly difference between balanced and imbalanced sequences (p < 0.05). Besides, each algorithm also showed a significant difference in sensitivity, specificity, accuracy, and macro F-score when used in both groups (p < 0.05). Several k-mers clustered from heat map were discovered to have biological functions and motifs at the different positions like interior loops, terminal loops and helices. They were validated to have a biological function and some are riboswitch motifs. The analysis has discovered the importance of solving the challenges of majority bias analysis and overfitting. Presented results were generalized evaluation of both balanced and imbalanced models, which implies their ability of classifying, to classify novel riboswitches. The Python source code is available at https://github.com/Seasonsling/riboswitch.