View in EDS

A Hybrid Machine Learning and Fractional-Order Dynamical Framework for Multi-Scale Prediction of Breast Cancer Progression.

Saved in:
Bibliographic Details
Title: A Hybrid Machine Learning and Fractional-Order Dynamical Framework for Multi-Scale Prediction of Breast Cancer Progression.
Authors: Amilo, David, Sadri, Khadijeh, Hincal, Evren, Hafez, Mohamed
Source: Computer Modeling in Engineering & Sciences (CMES); 2025, Vol. 145 Issue 2, p2189-2222, 34p
Subject Terms: BREAST cancer, MACHINE learning, ARTIFICIAL neural networks, DIAGNOSIS, RISK assessment, FRACTIONAL differential equations
Abstract: Breast cancer's heterogeneous progression demands innovative tools for accurate prediction. We present a hybrid framework that integrates machine learning (ML) and fractional-order dynamics to predict tumor growth across diagnostic and temporal scales. On the Wisconsin Diagnostic Breast Cancer dataset, seven ML algorithms were evaluated, with deep neural networks (DNNs) achieving the highest accuracy (97.72%). Key morphological features (area, radius, texture, and concavity) were identified as top malignancy predictors, aligning with clinical intuition. Beyond static classification, we developed a fractional-order dynamical model using Caputo derivatives to capture memory-driven tumor progression. The model revealed clinically interpretable patterns: lower fractional orders correlated with prolonged aggressive growth, while higher orders indicated rapid stabilization, mimicking indolent subtypes. Theoretical analyses were rigorously proven, and numerical simulations closely fit clinical data. The framework's clinical utility is demonstrated through an interactive graphics user interface (GUI) that integrates real-time risk assessment with growth trajectory simulations. [ABSTRACT FROM AUTHOR]
Copyright of Computer Modeling in Engineering & Sciences (CMES) is the property of Tech Science Press and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Complementary Index
Description
Abstract:Breast cancer's heterogeneous progression demands innovative tools for accurate prediction. We present a hybrid framework that integrates machine learning (ML) and fractional-order dynamics to predict tumor growth across diagnostic and temporal scales. On the Wisconsin Diagnostic Breast Cancer dataset, seven ML algorithms were evaluated, with deep neural networks (DNNs) achieving the highest accuracy (97.72%). Key morphological features (area, radius, texture, and concavity) were identified as top malignancy predictors, aligning with clinical intuition. Beyond static classification, we developed a fractional-order dynamical model using Caputo derivatives to capture memory-driven tumor progression. The model revealed clinically interpretable patterns: lower fractional orders correlated with prolonged aggressive growth, while higher orders indicated rapid stabilization, mimicking indolent subtypes. Theoretical analyses were rigorously proven, and numerical simulations closely fit clinical data. The framework's clinical utility is demonstrated through an interactive graphics user interface (GUI) that integrates real-time risk assessment with growth trajectory simulations. [ABSTRACT FROM AUTHOR]
ISSN:15261492
DOI:10.32604/cmes.2025.070298