Unveiling Breast Cancer Risk Profiles: A Comprehensive Survival Clustering Analysis Empowered by an Online Web Application for Personalized Medicine

Abstract

Medical doctors frequently rely on assistance tools during the decision-making process or when determining suitable chemotherapy options. These tools can take the form of recommendation systems, online test calculators, or web-based applications. They provide support not only in making recommendations but also in conducting thorough profile investigations of patients. Previous researchers have developed web-based survival analysis tools in the cancer survival field. However, many of these tools provide only basic functionality and rely on simplistic models, offering only a superficial understanding of the data. In this study, we undertake a comprehensive analysis of risk profiles using survival clustering techniques applied to a real-world dataset and developed an accessible online Shiny application to facilitate easier utilization of our findings. By leveraging survival clustering, we aim to uncover distinct subgroups based on survival patterns and identify unique risk profiles associated with breast cancer patients. Our online app provides a user-friendly interface for researchers and clinicians to explore the results, enabling them to gain valuable insights into the complex landscape of breast cancer risk profiles. This interactive tool offers a more accessible means of understanding and utilizing the implications of our research in personalized medicine and clinical decision-making.

Longitudinal Analysis of Step Counts in Parkinson’s Disease Patients: Insights from a Web-Based Application

“…Therefore, it is imperative to conduct a comprehensive longitudinal analysis of step counts, accounting for potential risk factors, using extensive and well-sourced long-term follow-up data, such as the PPMI study. On the other hand, although a couple of Shiny apps have been published online in our previous studies [16, 17], but they focus on different diseases, such as cancer or kidney failure, there is no accessible online platform for describing the baseline characteristics of PD patients or for modeling the time-varying trends in step counts between PD patients and healthy participants. This study addresses this gap by constructing a longitudinal model based on PPMI data and offering an online tool for clinical investigators to gain a more detailed visualization of these trends.…”

Comprehensive Swelling Analysis of Alginate Hydrogel

“…[28,29] Statistical methods were employed to analyze and interpret experimental data, providing a quantitative understanding of the effects of different variables on the hydrogel properties. [30][31][32][33] Furthermore, machine learning algorithms were applied to predict and optimize hydrogel characteristics based on the experimental input parameters. This advanced analytical approach not only enhances the accuracy of predictions but also opens avenues for designing alginate hydrogels with tailored properties for specific biomedical applications.…”

Analysis of polyurethane/gelatin complex hydrogel system for protein imprinting

“…[31,32] The evolving field of hydrogel-based imprinting techniques also extends to advanced methodologies such as surface imprinting, where statistical and machine learning algorithms can enhance design precision. [33][34][35][36][37][38] Techniques involving the imprinting of molecules on the surface of inorganic or organic carriers have gained prominence due to the resulting materials exhibiting well-defined morphology, thermal stability, and mechanical strength. [39,40] This growing area of research opens new possibilities for designing hydrogel-based materials with enhanced imprinting efficiency and stability through the integration of statistical and machine learning methods.…”