NORMA eResearch @NCI Library

Hybrid Machine Learning and Mathematical Modeling for Tumor Dynamics Prediction: Comparing SPIONs against mNP-FDG

Chattopadhyay, Amit K., Unkundiye, Aimee Pascaline N. and Pearce, Gillian (2025) Hybrid Machine Learning and Mathematical Modeling for Tumor Dynamics Prediction: Comparing SPIONs against mNP-FDG. Annals of Biostatistics & Biometric Applications, 6 (3). ISSN 2641-6336

Full text not available from this repository.
Official URL: https://irispublishers.com/abba/pdf/ABBA.MS.ID.000...

Abstract

Outline: Since the late eighties, Superparamagnetic Iron Oxide Nanoparticles (SPIONs) have been extensively studied for their exceptional ability in targeted drug delivery, typically coerced by magnetic fields and delivered at chemotherapy sites. Both iron-coated (SPIONs) and fluorodeoxyglucose-coated (mNP-FDG) magnetic nanoparticles are known to be highly potent in their ability to deliver drugs with uncanny precision to cancerous cells while minimizing damage to healthy cells. Of late, though, questions have been raised about the potential increase in cytotoxicity, particularly with SPIONs. Combining Machine Learning (Extreme Gradient Boosting) with continuum modeling (exponential and logistic growth), we find that while mNP-FDG can control tumor progression within 2 days compared to 18 days by SPIONs, for complete termination of the tumor, SPIONS (20 days) are superior compared to mNP-FDG (more than 40 days). We also provide an interactive graphical user interface (GUI) developed with Tkinter/Python that allows users to input relevant data, such as treatment type and time, to receive real-time tumor volume predictions. Our ML-guided prediction indicates joint therapy as the optimum choice, where mNP-FDG is largely responsible for controlling the initial spread of tumor the tumor spread, followed by SPIONs for complete eradication, facilitating personalized cancer treatment in clinical practice.

Main Limitations: This modeling study uses trendline data from multiple published sources that may have been conducted under varying experimental conditions. Thus, data optimization and normalization are potential challenges. Also, different forms of cancer may have been addressed in these separate studies. However, our robust modeling infrastructure ensures genericity through multiple ensembles averaging of results that consistently converge.

Objective: The present study aims at a comparative analysis of two groups of inorganic molecules, Superparamagnetic Iron Oxide Nanoparticles (SPIONs) and fluorodeoxyglucose-coated (mNP- FDG) magnetic nanoparticles, to understand the relative merits and demerits in optimizing specificity and cytotoxicity of chemotherapy treatment when administered through these agents. We also analyze the possibility of combinatorial drug administration using both agents.

Item Type: Article
Uncontrolled Keywords: Tumor dynamics; superparamagnetic iron oxide nanoparticles; mNP-FDG; machine learning; artificial intelligence; graphical user interface
Subjects: Q Science > QH Natural history > QH301 Biology > Methods of research. Technique. Experimental biology > Data processing. Bioinformatics > Artificial intelligence
Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Artificial intelligence
Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Q Science > Life sciences > Medical sciences > Pathology > Tumors
Divisions: School of Business > Staff Research and Publications
Depositing User: Tamara Malone
Date Deposited: 30 May 2025 13:43
Last Modified: 30 May 2025 13:43
URI: https://norma.ncirl.ie/id/eprint/7693

Actions (login required)

View Item View Item