ABSTRACT
Cone Beam Computed Tomography (CBCT) has emerged as a valuable imaging technique in dental diagnosis, which enables comprehensive evaluation of dental pathologies, aiding in the diagnosis and treatment planning of complex cases such as impacted teeth, dental implants, and orthodontic treatment. With its lower radiation dose compared to medical CT, CBCT provides a safe and efficient imaging modality. Recent advancements in unsupervised learning-based denoising algorithms have demonstrated their effectiveness in enhancing the quality of low-dose and noisy images. However, existing methods typically apply uniform denoising to the entire image, neglecting the need for targeted processing of specific regions. In this paper, we propose a collaborative denoising approach that combines image segmentation with the noise2sim algorithm, based on unsupervised learning. By leveraging the results of image segmentation, our method improves the denoising capabilities of the model. Specifically, different regions of the CBCT image undergo varying degrees of denoising, guided by the segmentation results. Fine-tuning is subsequently employed to optimize the model’s denoising performance. Experimental results demonstrate the superiority of our proposed method over state-of-the-art unsupervised learning denoising techniques.
- Joshua Batson and Loic Royer. 2019. Noise2self: Blind denoising by self-supervision. In International Conference on Machine Learning. PMLR, 524–533.Google Scholar
- Jianan Cui, Kuang Gong, Ning Guo, Chenxi Wu, Xiaxia Meng, Kyungsang Kim, Kun Zheng, Zhifang Wu, Liping Fu, Baixuan Xu, 2019. PET image denoising using unsupervised deep learning. European journal of nuclear medicine and molecular imaging 46 (2019), 2780–2789.Google Scholar
- Weiwei Cui, Liaoyuan Zeng, Bunsan Chong, and Qianni Zhang. 2021. Toothpix: Pixel-Level Tooth Segmentation in Panoramic X-Ray Images based on Generative Adversarial Networks. 2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI) (2021), 1346–1350.Google ScholarCross Ref
- Ross Girshick. 2015. Fast r-cnn. In Proceedings of the IEEE international conference on computer vision. 1440–1448.Google ScholarDigital Library
- Zhengmin Kong, Feng Xiong, Chenggang Zhang, Zhuolin Fu, Maoqi Zhang, Jingxin Weng, and Mingzhe Fan. 2020. Automated Maxillofacial Segmentation in Panoramic Dental X-Ray Images Using an Efficient Encoder-Decoder Network. IEEE Access 8 (2020), 207822–207833.Google ScholarCross Ref
- Alexander Krull, Tim-Oliver Buchholz, and Florian Jug. 2019. Noise2void-learning denoising from single noisy images. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2129–2137.Google ScholarCross Ref
- Jaakko Lehtinen, Jacob Munkberg, Jon Hasselgren, Samuli Laine, Tero Karras, Miika Aittala, and Timo Aila. 2018. Noise2Noise: Learning image restoration without clean data. arXiv preprint arXiv:1803.04189 (2018).Google Scholar
- Xiaomeng Li, Hao Chen, Xiaojuan Qi, Qi Dou, Chi-Wing Fu, and Pheng-Ann Heng. 2018. H-DenseUNet: hybrid densely connected UNet for liver and tumor segmentation from CT volumes. IEEE transactions on medical imaging 37, 12 (2018), 2663–2674.Google Scholar
- Yucheng Lu, Zhixin Xu, Moon Hyung Choi, Jimin Kim, and Seung-Won Jung. 2023. Multi-frame-based Cross-domain Image Denoising for Low-dose Computed Tomography. arXiv preprint arXiv:2304.10839 (2023).Google Scholar
- Chuang Niu, Mengzhou Li, Fenglei Fan, Weiwen Wu, Xiaodong Guo, Qing Lyu, and Ge Wang. 2020. Suppression of correlated noise with similarity-based unsupervised deep learning. arXiv e-prints (2020), arXiv–2011.Google Scholar
- Ayse Betul Oktay. 2017. Tooth detection with Convolutional Neural Networks. 2017 Medical Technologies National Congress (TIPTEKNO) (2017), 1–4.Google ScholarCross Ref
- Olaf Ronneberger, Philipp Fischer, and Thomas Brox. 2015. U-net: Convolutional networks for biomedical image segmentation. In Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part III 18. Springer, 234–241.Google Scholar
- Joonyoung Song, Jae-Heon Jeong, Dae-Soon Park, Hyun-Ho Kim, Doo-Chun Seo, and Jong Chul Ye. 2020. Unsupervised denoising for satellite imagery using wavelet directional CycleGAN. IEEE Transactions on Geoscience and Remote Sensing 59, 8 (2020), 6823–6839.Google ScholarCross Ref
- Chunwei Tian, Lunke Fei, Wenxian Zheng, Yong Xu, Wangmeng Zuo, and Chia-Wen Lin. 2020. Deep learning on image denoising: An overview. Neural Networks 131 (2020), 251–275.Google ScholarCross Ref
- Feng Wang, Trond R Henninen, Debora Keller, and Rolf Erni. 2020. Noise2Atom: unsupervised denoising for scanning transmission electron microscopy images. Applied Microscopy 50, 1 (2020), 1–9.Google ScholarCross Ref
- Siwei Yu, Jianwei Ma, and Wenlong Wang. 2019. Deep learning for denoisingdeep learning for denoising. Geophysics 84, 6 (2019), V333–V350.Google ScholarCross Ref
- Young-Jun Yu. 2016. Machine learning for dental image analysis. arXiv preprint arXiv:1611.09958 (2016).Google Scholar
- Yuxing Zhao, Yue Li, Xintong Dong, and Baojun Yang. 2018. Low-frequency noise suppression method based on improved DnCNN in desert seismic data. IEEE Geoscience and Remote Sensing Letters 16, 5 (2018), 811–815.Google ScholarCross Ref
- Long Zhou, Xiaozhuang Wang, Min Hou, Ping Li, Chunlong Fu, Yanjun Ren, Tingting Shao, Xi Hu, Jihong Sun, and Hongwei Ye. 2022. Low-dose CT reconstruction by self-supervised learning in the projection domain. arXiv preprint arXiv:2203.06824 (2022).Google Scholar
Index Terms
- Concrete learning method for segmentation and denoising using CBCT Image
Recommendations
Cone-Beam Computed Tomography (CBCT) Segmentation by Adversarial Learning Domain Adaptation
Medical Image Computing and Computer Assisted Intervention – MICCAI 2019AbstractCone-beam computed tomography (CBCT) is increasingly used in radiotherapy for patient alignment and adaptive therapy where organ segmentation and target delineation are often required. However, due to the poor image quality, low soft tissue ...
A supervised network for fast image-guided radiotherapy (IGRT) registration
3D/3D image registration in IGRT, which aligns planning Computed Tomography (CT) image set with on board Cone Beam CT (CBCT) image set in a short time with high accuracy, is still a challenge due to its high computational cost and complex anatomical ...
Liver segmentation using Turbolift learning for CT and cone-beam C-arm perfusion imaging
AbstractModel-based reconstruction employing the time separation technique (TST) was found to improve dynamic perfusion imaging of the liver using C-arm cone-beam computed tomography (CBCT). To apply TST using prior knowledge extracted from CT ...
Highlights- This research proposes Turbolift learning which trains a modified version of the multi-scale Attention UNet on different liver segmentation tasks serially.
Comments