Salvato in:
Dettagli Bibliografici
Autori principali: Cieślak, Daniel, Reca, Miriam, Onyshchenko, Olena, Rumiński, Jacek
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2507.05314
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908445679026176
author Cieślak, Daniel
Reca, Miriam
Onyshchenko, Olena
Rumiński, Jacek
author_facet Cieślak, Daniel
Reca, Miriam
Onyshchenko, Olena
Rumiński, Jacek
contents Accurate segmentation of wounds and scale markers in clinical images remainsa significant challenge, crucial for effective wound management and automatedassessment. In this study, we propose a novel dual-attention U-Net++ archi-tecture, integrating channel-wise (SCSE) and spatial attention mechanisms toaddress severe class imbalance and variability in medical images effectively.Initially, extensive benchmarking across diverse architectures and encoders via 5-fold cross-validation identified EfficientNet-B7 as the optimal encoder backbone.Subsequently, we independently trained two class-specific models with tailoredpreprocessing, extensive data augmentation, and Bayesian hyperparameter tun-ing (WandB sweeps). The final model ensemble utilized Test Time Augmentationto further enhance prediction reliability. Our approach was evaluated on a bench-mark dataset from the NBC 2025 & PCBBE 2025 competition. Segmentationperformance was quantified using a weighted F1-score (75% wounds, 25% scalemarkers), calculated externally by competition organizers on undisclosed hard-ware. The proposed approach achieved an F1-score of 0.8640, underscoring itseffectiveness for complex medical segmentation tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2507_05314
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Dual-Attention U-Net++ with Class-Specific Ensembles and Bayesian Hyperparameter Optimization for Precise Wound and Scale Marker Segmentation
Cieślak, Daniel
Reca, Miriam
Onyshchenko, Olena
Rumiński, Jacek
Image and Video Processing
Artificial Intelligence
Computer Vision and Pattern Recognition
Machine Learning
I.4.6; I.5.1; I.2.6; J.3
Accurate segmentation of wounds and scale markers in clinical images remainsa significant challenge, crucial for effective wound management and automatedassessment. In this study, we propose a novel dual-attention U-Net++ archi-tecture, integrating channel-wise (SCSE) and spatial attention mechanisms toaddress severe class imbalance and variability in medical images effectively.Initially, extensive benchmarking across diverse architectures and encoders via 5-fold cross-validation identified EfficientNet-B7 as the optimal encoder backbone.Subsequently, we independently trained two class-specific models with tailoredpreprocessing, extensive data augmentation, and Bayesian hyperparameter tun-ing (WandB sweeps). The final model ensemble utilized Test Time Augmentationto further enhance prediction reliability. Our approach was evaluated on a bench-mark dataset from the NBC 2025 & PCBBE 2025 competition. Segmentationperformance was quantified using a weighted F1-score (75% wounds, 25% scalemarkers), calculated externally by competition organizers on undisclosed hard-ware. The proposed approach achieved an F1-score of 0.8640, underscoring itseffectiveness for complex medical segmentation tasks.
title Dual-Attention U-Net++ with Class-Specific Ensembles and Bayesian Hyperparameter Optimization for Precise Wound and Scale Marker Segmentation
topic Image and Video Processing
Artificial Intelligence
Computer Vision and Pattern Recognition
Machine Learning
I.4.6; I.5.1; I.2.6; J.3
url https://arxiv.org/abs/2507.05314