Abstract
Automated breast cancer classification from mammography remains a significant challenge due to subtle distinctions between benign and malignant tissue. In this work, we present a hybrid framework combining deep convolutional features from a ResNet-50 backbone with handcrafted descriptors and transformer-based embeddings. Using the CBIS-DDSM dataset, we benchmark our ResNet-50 baseline (AUC: 78.1%) and demonstrate that fusing handcrafted features with deep ResNet-50 and DINOv2 features improves AUC to 79.6% (setup d1), with a peak recall of 80.5% (setup d1) and highest F1 score of 67.4% (setup d1). Our experiments show that handcrafted features not only complement deep representations but also enhance performance beyond transformer-based embeddings. This hybrid fusion approach achieves results comparable to state-of-the-art methods while maintaining architectural simplicity and computational efficiency, making it a practical and effective solution for clinical decision support.
| Original language | English |
|---|---|
| Publication status | Accepted/In press - 26 Jul 2025 |
| Event | International Conference on Image Processing Theory, Tools and Applications: IPTA 2025 - Istanbul, Türkiye, Istanbul, Turkey Duration: 13 Oct 2025 → 16 Oct 2025 Conference number: 14 https://ipta-conference.com/ipta25/index.php |
Conference
| Conference | International Conference on Image Processing Theory, Tools and Applications |
|---|---|
| Abbreviated title | IPTA 2025 |
| Country/Territory | Turkey |
| City | Istanbul |
| Period | 13/10/25 → 16/10/25 |
| Internet address |
UN SDGs
This output contributes to the following UN Sustainable Development Goals (SDGs)
-
SDG 3 Good Health and Well-being
-
SDG 9 Industry, Innovation, and Infrastructure
ASJC Scopus subject areas
- Computer Vision and Pattern Recognition
Classification according to Österreichische Systematik der Wissenschaftszweige (ÖFOS 2012)
- 102019 Machine learning
Applied Research Level (ARL)
- ARL Level 3 - Proof of the functionality of a principle
Research focus/foci
- Industrial Informatics
- Applied Health Innovation
Fingerprint
Dive into the research topics of 'Hybrid Deep Learning and Handcrafted Feature Fusion for Mammographic Breast Cancer Classification'. Together they form a unique fingerprint.Projects
- 1 Finished
-
AIBIA: Research and Transfer Junior Lab on AI in Biomedical Image Analysis
Wegenkittl, S. (CoPI), Gadermayr, M. (PI) & Löwenstein, K. (CoI)
1/04/23 → 31/03/26
Project: Funded research
Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver