Wavelet‐Enhanced Deformable Convolutional Network for Breast Cancer Classification in High‐Resolution Histopathology Images
- 29 May 2025
- journal article
- Published by Wiley in Applied Research
- Vol. 4 (3)
Abstract
The limitations of deep learning methods in processing high‐resolution inputs can impact the accuracy and efficiency of their results. This study presents a new architectural framework that combines wavelet‐based preprocessing with deformable convolutional networks to classify high‐resolution histopathological images. Our methodology utilizes multi‐resolution wavelet decomposition for efficient feature extraction which maintains diagnostically significant information. This improvement is augmented by deformable convolutions, which improve robustness against geometric transformations of the inputs. Empirical evaluation on the BreaKHis data set shows an image‐level accuracy of 96.47% and a patient‐level accuracy of 96.55% at 200× magnification. The architecture consistently performs well across different magnification levels, with particular efficiency at higher resolutions where detailed morphological features are essential for accurate diagnosis. Ablation studies support our key architectural contributions, including reduced computational complexity through wavelet‐based feature extraction, improved geometric invariance via deformable convolutions, and better classification performance than conventional methods. These findings suggest significant potential for improving diagnostic workflows in clinical settings where pathological expertise may be limited.Keywords
This publication has 40 references indexed in Scilit:
- Wavelet‐Based Feature Extraction for Efficient High‐Resolution Image ClassificationEngineering Reports, 2025
- High-resolution AI image dataset for diagnosing oral submucous fibrosis and squamous cell carcinomaScientific Data, 2024
- DFEDC: Dual fusion with enhanced deformable convolution for medical image segmentationImage and Vision Computing, 2024
- Multi-layer pseudo-supervision for histopathology tissue semantic segmentation using patch-level classification labelsMedical Image Analysis, 2022
- Breast histopathological image analysis using image processing techniques for diagnostic purposes: A methodological reviewJournal of Medical Systems, 2021
- Residual learning based CNN for breast cancer histopathological image classificationInternational Journal of Imaging Systems and Technology, 2020
- Pre‐trained convolutional neural networks as feature extractors for diagnosis of breast cancer using histopathologyInternational Journal of Imaging Systems and Technology, 2020
- Global trends in population-based survival for 303,169 adults diagnosed with glioblastoma in 44 countries during 2000-2014 (CONCORD-3)Annals of Oncology, 2019
- Multiple instance learning for histopathological breast cancer image classificationExpert Systems with Applications, 2019
- Breast cancer in sub‐Saharan Africa: How does it relate to breast cancer in African‐American women?Cancer, 2005