Multi-target drug discovery has become an essential strategy for treating complex diseases involving multiple molecular pathways. Traditional single-target approaches often fall short in addressing the multifactorial nature of conditions such as cancer and neurodegenerative disorders. With the rise in large-scale biological data and algorithmic advances, machine learning (ML) has emerged as a powerful tool to accelerate and optimize multi-target drug development. This review presents a comprehensive overview of ML techniques, including advanced deep learning (DL) approaches like attention-based models, and highlights their application in multi-target prediction, from traditional supervised learning to modern graph-based and multi-task learning frameworks. We highlight real-world applications in oncology, central nervous system disorders, and drug repurposing, showcasing the translational potential of ML in systems pharmacology. Major challenges are discussed, such as data sparsity, lack of interpretability, limited generalizability, and integration into experimental workflows. We also address ethical and regulatory considerations surrounding model transparency, fairness, and reproducibility. Looking forward, we explore promising directions such as generative modeling, federated learning, and patient-specific therapy design. Together, these advances point toward a future of precision polypharmacology driven by biologically informed and interpretable ML models. This review aims to provide researchers and practitioners with a roadmap for leveraging ML in the development of safer and more effective multi-target therapeutics.

Cancer’s heterogeneity presents significant challenges in accurate diagnosis and effective treatment, including the complexity of identifying tumor subtypes and their diverse biological behaviors. This review examines how feature selection techniques address these challenges by improving the interpretability and performance of machine learning (ML) models in high-dimensional datasets. Feature selection methods—such as filter, wrapper, and embedded techniques—play a critical role in enhancing the precision of cancer diagnostics by identifying relevant biomarkers. The integration of multi-omics data and ML algorithms facilitates a more comprehensive understanding of tumor heterogeneity, advancing both diagnostics and personalized therapies. However, challenges such as ensuring data quality, mitigating overfitting, and addressing scalability remain critical limitations of these methods. Artificial intelligence (AI)-powered feature selection offers promising solutions to these issues by automating and refining the feature extraction process. This review highlights the transformative potential of these approaches while emphasizing future directions, including the incorporation of deep learning (DL) models and integrative multi-omics strategies for more robust and reproducible findings.

Osteoporosis is a systemic bone disorder characterized by decreased bone mass and deteriorated microarchitecture, leading to an increased risk of fractures. Recent studies have revealed that its pathogenesis involves complex biological processes beyond bone remodeling, including oxidative stress, chronic inflammation, cellular senescence, osteoimmunology, gut microbiota alterations, and epigenetic modifications. Oxidative stress disrupts bone homeostasis by promoting excessive free radical production and osteoclast activity. Chronic inflammation and the accumulation of senescent cells impair skeletal repair mechanisms. Advances in osteoimmunology have highlighted the critical role of immune–bone crosstalk in regulating bone resorption and formation. Moreover, the gut–bone axis, mediated by microbial metabolites, influences bone metabolism through immune and endocrine pathways. Epigenetic changes, such as DNA methylation and histone modification, contribute to gene–environment interactions, affecting disease progression. Multi-omics approaches (genomics, proteomics, and metabolomics) systematically identify molecular networks and comorbid links with diabetes/cardiovascular diseases, revealing pathological feedback loops that exacerbate bone loss. In conclusion, osteoporosis pathogenesis extends beyond bone remodeling to encompass systemic inflammation, immunometabolic dysregulation, and gut microbiota–host interactions. Future research should focus on integrating multi-omics biomarkers with targeted therapies to advance precision medicine strategies for osteoporosis prevention and treatment.

Gastrointestinal (GI) diseases, including gastric and colorectal cancers, significantly impact global health, necessitating accurate and efficient diagnostic methods. Endoscopic examination is the primary diagnostic tool; however, its accuracy is limited by operator dependency and interobserver variability. Advancements in deep learning, particularly convolutional neural networks (CNNs), show great potential for enhancing GI disease detection and classification. This review explores the application of CNNs in endoscopic imaging, focusing on polyp and tumor detection, disease classification, endoscopic ultrasound, and capsule endoscopy analysis. We discuss the performance of CNN models with traditional diagnostic methods, highlighting their advantages in accuracy and real-time decision support. Despite promising results, challenges remain, including data availability, model interpretability, and clinical integration. Future directions include improving model generalization, enhancing explainability, and conducting large-scale clinical trials. With continued advancements, CNN-powered artificial intelligence systems could revolutionize GI endoscopy by enhancing early disease detection, reducing diagnostic errors, and improving patient outcomes.

Precise drug target discovery is pivotal to mitigating the escalating costs and high attrition rates that characterize pharmaceutical research and development. Given that traditional single-omics methods often fail to elucidate the systemic complexity of human diseases, deep learning (DL)-enabled multi-omics integration has emerged as a transformative frontier. This review systematically summarizes the advancements in DL-driven multi-omics integration for drug target discovery. First, the multi-omics data foundation and integration strategies are delineated, followed by an exploration of the DL architectures utilized for processing such data. Subsequently, the efficacy of DL-driven multi-omics integration is examined regarding the identification of novel disease drivers, prediction of synthetic lethality interactions, and prioritization of therapeutic targets. Finally, addressing persistent challenges related to data sparsity, model interpretability, and target druggability and validation hurdles, emerging opportunities driven by Generative AI, Large Multimodal Models (LMMs), Explainable AI (XAI), and multidimensional feasibility assessment frameworks are discussed in the context of advancing precision medicine.

Advances in artificial intelligence (AI) and intelligent computing have significantly accelerated drug discovery by enabling accurate modeling of complex biomedical relationships. Among these efforts, drug repositioning—identifying novel therapeutic uses for approved or investigational drugs—offers a cost-effective and time-efficient alternative to de novo drug development. While non-negative matrix factorization (NMF) has been widely adopted for uncovering latent drug–disease associations, conventional implementations often neglect the biological context that underpins these relationships. In this work, we propose a novel NMF-based drug repositioning model that incorporates biological context (NMFIBC), which integrates drug and disease similarity networks through graph-regularized optimization to enhance predictive performance. This design enhances both the robustness and interpretability of association prediction. Extensive benchmarking on multiple gold-standard datasets demonstrates that NMFIBC outperforms existing methods across a range of metrics, including AUC, precision, and F1-score. Moreover, case studies involving clinically relevant drugs validate the biological plausibility of the predicted associations using public databases such as DrugBank, CTD, and KEGG. The proposed framework provides a powerful, context-aware AI strategy for discovering actionable insights in drug repositioning research.