Abstract:
Accurate cancer classification using gene expression data is essential for early diagnosis and effective treatment planning. Gene expression datasets are typically high-dimensional, containing thousands of gene features but relatively few patient samples, which pose challenges for traditional machine learning approaches, making them sensitive to noise and prone to over fitting. In this study, we introduce a novel preprocessing framework based on fractional calculus to enhance the analysis of gene expression profiles. By applying fractional derivatives to the gene expression vectors, subtle variations and long range dependencies among genes can be captured, revealing complex correlations that conventional integer-order methods often miss. This transformation effectively enhances the reparability of malignant and benign profiles, providing more informative and robust features for subsequent classification models. Preliminary evaluations indicate that the fractional feature transformation improves overall classification accuracy while mitigating the dimensionality burden and noise sensitivity inherent in high-dimensional genomic data. Beyond improving predictive performance, this approach bridges mathematical innovation with practical biomedical applications, demonstrating the potential of fractional calculus to uncover hidden patterns in complex biological datasets. By integrating advanced mathematical techniques with real- world cancer diagnostics, the proposed methodology offers a promising avenue for precision oncology, enabling more reliable and interpretable gene expression-based classification. This work highlights the value of combining interdisciplinary approach mathematics, computer science, and biomedical science-to address pressing challenges in modern healthcare and underscores the potential of fractional methods in transforming biological data analysis.
Page(s):
184-184
DOI:
DOI not available
Published:
Journal: 4th International Conference of Sciences “Revamped Scientific Outlook of 21st Century, 2025” , November 12,2025, Volume: 1, Issue: 1, Year: 2025
Keywords:
Data analysis
,
gene expression
,
Fractional Derivative
,
cancer classification
,
feature preprocessing