ADNF-Clustering: An Adaptive and Dynamic Neuro-Fuzzy Clustering for Leukemia Prediction

2025-06-23

AI Summary

To address the challenge of dynamically evolving cellular patterns and the difficulty of real-time uncertainty quantification in leukemia microscopic images, this paper proposes an adaptive neuro-fuzzy clustering framework. The method introduces a Fuzzy Temporal Index (FTI) to dynamically regulate clustering parameters and integrates density-weighted merging with entropy-guided splitting to suppress oversegmentation/undersegmentation and enable online uncertainty modeling. It combines CNN-based feature extraction, fuzzy c-means soft initialization, online micro-clustering updates, and topological optimization into an end-to-end streaming clustering pipeline. Evaluated on the C-NMC dataset, the framework achieves a silhouette coefficient of 0.51—significantly outperforming static baselines—and demonstrates strong potential for clinical deployment, supporting the continuous evolution required for personalized leukemia diagnosis and treatment.

Technology Category

• Machine Learning: Clustering
• Reasoning under Uncertainty: Uncertainty Representations
• Computer Vision: Segmentation

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Leukemia diagnosis and monitoring rely increasingly on high-throughput image data, yet conventional clustering methods lack the flexibility to accommodate evolving cellular patterns and quantify uncertainty in real time. We introduce Adaptive and Dynamic Neuro-Fuzzy Clustering, a novel streaming-capable framework that combines Convolutional Neural Network-based feature extraction with an online fuzzy clustering engine. ADNF initializes soft partitions via Fuzzy C-Means, then continuously updates micro-cluster centers, densities, and fuzziness parameters using a Fuzzy Temporal Index (FTI) that measures entropy evolution. A topology refinement stage performs density-weighted merging and entropy-guided splitting to guard against over- and under-segmentation. On the C-NMC leukemia microscopy dataset, our tool achieves a silhouette score of 0.51, demonstrating superior cohesion and separation over static baselines. The method's adaptive uncertainty modeling and label-free operation hold immediate potential for integration within the INFANT pediatric oncology network, enabling scalable, up-to-date support for personalized leukemia management.

Problem

• Adaptive clustering for evolving leukemia cellular patterns
• Real-time uncertainty quantification in high-throughput image data
• Dynamic neuro-fuzzy framework for label-free leukemia prediction

Innovation

• Combines CNN feature extraction with online fuzzy clustering
• Updates cluster parameters using Fuzzy Temporal Index
• Performs density-weighted merging and entropy-guided splitting