Enhancing U-Net for Wrist Fracture Segmentation in X-ray Images using Adaptive Callbacks and Weighted Loss Functions

Teuku Radillah, Sarjon Defit, Gunadi Widi Nurcahyo
Pages 2623-2635
Abstract Views 689
PDF Downloads 532
Crossref Citations -
DOI https://doi.org/10.47738/jads.v6i4.952

Abstract


The detection of wrist fracture through medical imaging is causing considerable challenges due to the subtle and variable manifestation of such ruptures, necessitating precise and reliable segmentation methods. Therefore, this research aimed to propose an improved U-Net model for detecting wrist fracture. The model incorporated two innovations, namely adaptive callback training and weighted loss combination. The adaptive callback mechanism could be performed by dynamically adjusting the training parameters based on the model performance to prevent overfitting and accelerate convergence. At the same time, the loss function combined Dice Loss and Binary Cross-Entropy (BCE) Loss with linear as well as non-linear exponential weighting strategies, ensuring balanced optimization between region-based accuracy and pixel classification. During this analysis, a series of experiments were conducted on a curated wrist X-ray image dataset, and the results showed that the proposed method expressed superior performance in terms of segmentation accuracy when compared with previous U-Net and other state-of-the-art procedures. The proposed method achieved 91% accuracy, 87% precision, 86% recall, and 87% F1 score. Following this discussion, the findings showed the efficacy of the adaptive training design and loss function in improving the strength and sensitivity of the model in detecting wrist fracture

Keywords


Wrist Fracrure Segmentation, U-Net Enhancement, Adaptive Callback Training, Wighted Loss Function, X-ray Image Segmentation

Full Text:

PDF

References


· Giuseppe John R. Zech, Carotenuto, · Zenas Igbinoba, · Clement Vinh Tran, · Elena Insley, · Alyssa Baccarella, and T. T. Wong, “Detecting pediatric wrist fractures using deep learning based object detection,” Pediatr Radiol, vol. https://do, no. 1125–1134, 2023.

J. R. Zech et al., “Detecting pediatric wrist fractures using deep-learning-based object detection,” Pediatr Radiol, vol. 53, no. 6, pp. 1125–1134, May 2023, doi: 10.1007/s00247-023-05588-8.

S. Turk, O. Bingol, A. Coskuncay, and T. Aydin, “The impact of implementing backbone architectures on fracture segmentation in X-ray images,” Engineering Science and Technology, an International Journal, vol. 59, Nov. 2024, doi: 10.1016/j.jestch.2024.101883.

M. Widiasri, N. Suciati, C. Fatichah, E. R. Astuti, R. H. Putra, and A. Z. Arifin, “Alveolar Bone and Mandibular Canal Segmentation on Cone Beam Computed Tomography Images Using U-Net,” Proceedings of the 2023 International Conference on Instrumentation, Control, and Automation, ICA 2023, no. 008, pp. 36–41, 2023, doi: 10.1109/ICA58538.2023.10273108.

J. Du, K. Guan, P. Liu, Y. Li, and T. Wang, “Boundary-Sensitive Loss Function With Location Constraint for Hard Region Segmentation,” IEEE J Biomed Health Inform, vol. 27, no. 2, pp. 992–1003, 2023, doi: 10.1109/JBHI.2022.3222390.

S. E. Roshan, J. Tanha, M. Zarrin, A. F. Babaei, H. Nikkhah, and Z. Jafari, “A deep ensemble medical image segmentation with novel sampling method and loss function,” Comput Biol Med, vol. 172, Apr. 2024, doi: 10.1016/j.compbiomed.2024.108305.

S. H. Choi, “Optimizing Deep Learning Models with Hybrid Nonlinear Loss Functions: Integrating Heterogeneous Nonlinear Properties for Enhanced Performance,” International Journal of Fuzzy Logic and Intelligent Systems, vol. 24, no. 4, pp. 317–332, Dec. 2024, doi: 10.5391/IJFIS.2024.24.4.317.

A. R. Hareendranathan, A. Tripathi, M. R. Panicker, J. Zhang, N. Boora, and J. Jaremko, “Deep Learning Approach for Automatic Wrist Fracture Detection Using Ultrasound Bone Probability Maps,” SN Compr Clin Med, vol. 5, no. 1, Nov. 2023, doi: 10.1007/s42399-023-01608-8.

A. Khanal, R. Rizk, and K. C. Santosh, “Ensemble Deep Convolutional Neural Network to Identify Fractured Limbs using CT Scans,” in Proceedings - 2023 IEEE Conference on Artificial Intelligence, CAI 2023, Institute of Electrical and Electronics Engineers Inc., 2023, pp. 156–157. doi: 10.1109/CAI54212.2023.00075.

R. Dhanalakshmi, M. Thenmozhi, S. Saxena, and H. Mahalingam, “Convolutional Neural Network Model based Deep Learning Approach for Osteoporosis Fracture Detection,” in Proceedings of the 1st IEEE International Conference on Networking and Communications 2023, ICNWC 2023, Institute of Electrical and Electronics Engineers Inc., 2023. doi: 10.1109/ICNWC57852.2023.10127367.

M. R, Vadhera, Sharma, “A novel hybrid loss based Encoder–Decoder model for accurate Pulmonary Embolism segmentation,” Int. j. inf. tecnol, 2025, doi: https://doi.org/10.1007/s41870-024-02398-5.

M. Ali et al., “Segmentation of MRI tumors and pelvic anatomy via cGAN-synthesized data and attention-enhanced U-Net,” Pattern Recognit Lett, vol. 187, no. November 2024, pp. 100–106, 2025, doi: 10.1016/j.patrec.2024.11.003.

J. Sun, Y. Li, X. Wu, C. Tang, S. Wang, and Y. Zhang, “HAD-Net: An attention U-based network with hyper-scale shifted aggregating and max-diagonal sampling for medical image segmentation,” Computer Vision and Image Understanding, vol. 249, no. September, p. 104151, 2024, doi: 10.1016/j.cviu.2024.104151.

Q. Ming and X. Xiao, “Towards Accurate Medical Image Segmentation with Gradient-Optimized Dice Loss,” IEEE Signal Process Lett, vol. 31, pp. 191–195, 2024, doi: 10.1109/LSP.2023.3329437.

E. Bosco, G. Magenes, and G. Matrone, “Echocardiographic Image Segmentation with Vision Transformers: A Comparative Analysis of Different Loss Functions,” 2024 IEEE International Symposium on Medical Measurements and Applications, MeMeA 2024 - Proceedings, 2024, doi: 10.1109/MeMeA60663.2024.10596874.

R. Maurya, P. Thirwarni, T. Gopalakrishnan, and M. Karnati, “Combining Focal loss with Cross-entropy loss for Pneumonia Classification with a Weighted Sampling Approach,” 2024. doi: 10.1109/IATMSI60426.2024.10502684.

F. Piri, N. Karimi, and S. Samavi, “Enhanced Segmentation in Abdominal CT Images: Leveraging Hybrid CNN-Transformer Architectures and Compound Loss Function,” 2024 IEEE 5th World AI IoT Congress, AIIoT 2024, pp. 363–369, 2024, doi: 10.1109/AIIoT61789.2024.10579036.

L. Zeng and S. Guo, “Automatic Liver Tumor Segmentation from CT Images with Dual Attention Mechanism,” 2024 9th International Conference on Electronic Technology and Information Science, ICETIS 2024, pp. 477–480, 2024, doi: 10.1109/ICETIS61828.2024.10593773.

Z. Song and L. Wang, “Robust Image Segmentation Algorithm Based on Local Correntropy-Based Generalized Loss Function,” 2024 7th International Conference on Advanced Algorithms and Control Engineering, ICAACE 2024, pp. 12–16, 2024, doi: 10.1109/ICAACE61206.2024.10548610.

Y. Song, J. Y. C. Teoh, K. S. Choi, and J. Qin, “Dynamic Loss Weighting for Multiorgan Segmentation in Medical Images,” IEEE Trans Neural Netw Learn Syst, 2023, doi: 10.1109/TNNLS.2023.3243241.

T. H. Nguyen et al., “Systematic Evaluation of Loss Functions for Ovarian Tumors Segmentation from Ultrasound Images,” 2023 1st International Conference on Health Science and Technology, ICHST 2023, 2023, doi: 10.1109/ICHST59286.2023.10565336.

M. Geijer, E. Gunnlaugsson, L. Arvidsson, E. Österhed, and M. Tägil, “Outcome of follow up computed tomography of suspected occult scaphoid fracture after normal radiography,” Emerg Radiol, no. 0123456789, 2024, doi: 10.1007/s10140-024-02307-0.

J. Sawant and A. D. Vibhute, “U-Net-based blood vessel segmentation using an improved data augmentation with green channel images,” Procedia Comput Sci, vol. 260, pp. 1071–1079, 2025, doi: 10.1016/j.procs.2025.03.292.

Y. Zheng et al., “Adaptive boundary-enhanced Dice loss for image segmentation,” Biomed Signal Process Control, vol. 106, Aug. 2025, doi: 10.1016/j.bspc.2025.107741.

Y. Xu et al., “Automatic segmentation of prostate cancer metastases in PSMA PET/CT images using deep neural networks with weighted batch-wise dice loss,” Comput Biol Med, vol. 158, May 2023, doi: 10.1016/j.compbiomed.2023.106882.

Z. Zhou, L. Cai, P. Yin, X. Qian, Y. Dai, and Z. Zhou, “Narrow-band loss - A novel loss function focused on target boundary,” Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS, pp. 1–4, 2023, doi: 10.1109/EMBC40787.2023.10340038.

X. X. Ming Qi, “Toward Accurate Medical Image Segmentation With Gradien-Optimized Dice Loss,” IEEE Signal Process Lett, vol. 31, pp. 191–195, 2024, doi: DOI: 10.1109/LSP.2023.3329437.

S. Kato and K. Hotta, “Adaptive t-vMF dice loss: An effective expansion of dice loss for medical image segmentation,” Comput Biol Med, vol. 168, Jan. 2024, doi: 10.1016/j.compbiomed.2023.107695.

S. Koles, S. Karakas, A. P. Ndigande, and S. Ozer, “Using Different Loss Functions with YOLACT++ for Real-Time Instance Segmentation,” 2023 46th International Conference on Telecommunications and Signal Processing, TSP 2023, pp. 264–267, 2023, doi: 10.1109/TSP59544.2023.10197832.

R. R, S. S, N. V. Kumar, R. S, and P. B V, “Integrated pixel-level crack detection and quantification using an ensemble of advanced U-Net architectures,” Results in Engineering, vol. 25, Mar. 2025, doi: 10.1016/j.rineng.2024.103726.

M. Attari, N. P. Nguyen, K. Palaniappan, and F. Bunyak, “Multi-Loss Topology-Aware Deep Learning Network for Segmentation of Vessels in Microscopy Images,” Proceedings - Applied Imagery Pattern Recognition Workshop, 2023, doi: 10.1109/AIPR60534.2023.10440665.

B. Gao, R. Yao, and Y. Li, “Physics-informed neural networks with adaptive loss weighting algorithm for solving partial differential equations,” Computers and Mathematics with Applications, vol. 181, pp. 216–227, Mar. 2025, doi: 10.1016/j.camwa.2025.01.007.

A. Akagic, M. Kapo, E. Kandić, M. Bećirović, and N. Kadrić, “Brain Tumor Segmentation of MRI Images with U-Net and DeepLabV3+,” pp. 1–6, 2024, doi: 10.1109/icmi60790.2024.10585749.

M. Obayya et al., “A novel U-net model for brain tumor segmentation from MRI images,” Alexandria Engineering Journal, vol. 126, pp. 220–230, Jul. 2025, doi: 10.1016/j.aej.2025.04.051.

R. Joshi, “Segmentation of Teeth in Panoramic X-ray Image Using U-net Algorithm,” 2024 3rd International Conference on Artificial Intelligence for Internet of Things, AIIoT 2024, no. AIIoT, pp. 1–6, 2024, doi: 10.1109/AIIoT58432.2024.10574786.

R. Karthick Manoj, S. Aasha Nandhini, and M. Batumalay, “Automated Brain Tumor Analysis with Multimodal Fusion and Augmented Intelligence,” Journal of Applied Data Sciences, vol. 6, no. 2, pp. 1277–1290, May 2025, doi: 10.47738/jads.v6i2.719.

B. Garg, D. C. Lohani, and B. Rana, “A Comparative Study of U-Net-based Segmentation Architectures for Crack Detection,” in Procedia Computer Science, Elsevier B.V., 2025, pp. 3305–3314. doi: 10.1016/j.procs.2025.04.588.

P. Belwal and S. Singh, “Multiple Sclerosis Lesion Segmentation using MRI: U-Net Approach,” Procedia Comput Sci, vol. 260, pp. 513–522, 2025, doi: 10.1016/j.procs.2025.03.229.




DOI: https://doi.org/10.47738/jads.v6i4.952

Refbacks

  • There are currently no refbacks.


Barcode

Journal of Applied Data Sciences

ISSN:2723-6471 (Online)
Publisher:Bright Publisher
Website:http://bright-journal.org/JADS
Email:taqwa@amikompurwokerto.ac.id (principal contact)
  support@bright-journal.org (technical issues)

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0