Noise2Match: Self-Supervised Low-Dose CT Denoising via Adaptive Inter-Slice Matching

Langrui Zhou; Guang Li; Xinyu Huang; Ziteng Zhou; Zuyang Huang; Hongming Shan; Shouhua Luo; Ning Gu

cs.CVOct 13, 2026

Noise2Match: Self-Supervised Low-Dose CT Denoising via Adaptive Inter-Slice Matching

Authors: Langrui Zhou, Guang Li, Xinyu Huang, Ziteng Zhou, Zuyang Huang, Hongming Shan, Shouhua Luo, Ning Gu

Organizations: Jiangsu Key Laboratory for Biomaterials and Devices, School of Biological Science and Medical Engineering, Southeast University, Nanjing 210009, China. · Institute of Science and Technology for Brain-Inspired Intelligence, Fudan University, Shanghai 201203, China · Jiangsu Key Laboratory for Cardiovascular Information and Health Engineering Medicine, Institute of Clinical Medicine, Nanjing Drum Tower Hospital, Medical School, Nanjing University, Nanjing 210093, China

Abstract

With the rapid development of deep learning, supervised convolutional neural network methods have greatly improved the performance of low-dose computed tomography (CT) denoising. However, they require paired noisy-clean training data, which is difficult to obtain in practice. Some classic self-supervised denoising methods rely on the assumption of pixel-wise independent noise, making them ineffective for correlated noise in low-dose CT (LDCT) images. Considering the high similarity and weaker noise correlation between adjacent CT slices, several recently proposed self-supervised methods use adjacent LDCT slices for training. Nevertheless, these approaches fail to effectively account for the non-rigid misalignment between slices caused by anatomical variation, hindering them from satisfying the equal-mean condition required by the classic Noise2Noise framework. In this paper, we propose Noise2Match, a self-supervised LDCT denoising method based on adaptive inter-slice matching. Noise2Match introduces an inter-slice matching network to model the label transition process between adjacent slices, and then applies correction to the denoiser's outputs to better satisfy the equal-mean condition. We theoretically prove that, under idealized conditions, training Noise2Match using adjacent LDCT slices is equivalent to supervised training with paired noisy-clean images. We further find that Noise2Match can also effectively perform low-dose conebeam CT (CBCT) denoising, which could be attributed to its correction operation that reduces inter-slice noise correlation. To the best of our knowledge, Noise2Match is the first effective image-domain self-supervised method for low-dose CBCT. Extensive experiments demonstrate that Noise2Match outperforms existing image-domain selfsupervised methods, offering broader applicability and higher practical value.

Figures & tables

Fig. 1: In self-supervised LDCT denoising based on adjacent slices, (a) directly computing the loss between same-coordinate pixels of adjacent slices does not conform to (b) the non-rigid inter-slice misalignment caused by anatomical variation.

Fig. 2: Overview of the proposed Noise2Match for LDCT denoising. The denoising network $f_{\theta}$ takes noisy LDCT images as input and predicts the corresponding clean images. The inter-slice matching network $M_{\phi}$ estimates the label transition matrix between adjacent slices based on the denoised results. The estimated transition matrix is then used to correct the denoiser’s output.

Fig. 3: Qualitative results of different methods on the Mayo 2016 and Mayo 2020 datasets. (a) NDCT (ground truth), (b) LDCT, (c) BM3D, (d) RED-CNN, (e) Noise2Clean, (f) Noise2Void, (g) Neighbor2Neighbor, (h) SDCNN, (i) MM-Net, (j) Noise2Sim, (k) SVB, (l) Noise2Match ( ours ).

Methods	Mayo 2016			Mayo 2020
Methods	PSNR (dB) $\uparrow$	SSIM $\uparrow$	RMSE (HU) $\downarrow$	PSNR (dB) $\uparrow$	SSIM $\uparrow$	RMSE (HU) $\downarrow$
LDCT	34.75 $\pm$ 0.95	0.8211 $\pm$ 0.0342	36.81 $\pm$ 4.23	36.44 $\pm$ 3.04	0.8682 $\pm$ 0.0662	26.68 $\pm$ 11.29
BM3D	37.25 $\pm$ 0.88	0.8952 $\pm$ 0.0253	27.58 $\pm$ 2.91	39.03 $\pm$ 2.80	0.8977 $\pm$ 0.0291	25.21 $\pm$ 6.00
RED-CNN	40.56 $\pm$ 0.89	0.9435 $\pm$ 0.0123	18.83 $\pm$ 1.99	43.54 $\pm$ 2.85	0.9691 $\pm$ 0.0184	14.16 $\pm$ 5.95
Noise2Clean	40.52 $\pm$ 0.90	0.9437 $\pm$ 0.0127	18.93 $\pm$ 2.01	43.41 $\pm$ 2.95	0.9698 $\pm$ 0.0173	14.44 $\pm$ 6.49
Noise2Void	35.42 $\pm$ 0.91	0.8420 $\pm$ 0.0317	34.06 $\pm$ 3.76	38.35 $\pm$ 2.97	0.8762 $\pm$ 0.0582	26.13 $\pm$ 10.96

TABLE I: Quantitative results (mean

\pm

std) of different methods on the Mayo 2016 and Mayo 2020 datasets.

Fig. 4: Qualitative results of different self-supervised methods on the Mayo 2016 and Piglet datasets. These models are all directly trained on the corresponding single-volume LDCT data. (a) NDCT (ground truth), (b) LDCT, (c) Noise2Void, (d) Neighbor2Neighbor, (e) SDCNN, (f) MM-Net, (g) Noise2Sim, (h) SVB, (i) Noise2Match ( ours ).

	Methods	PSNR (dB) $\uparrow$	SSIM $\uparrow$	RMSE (HU) $\downarrow$
Mayo 2016	LDCT	36.39 $\pm$ 1.86	0.8605 $\pm$ 0.0493	31.06 $\pm$ 7.62
	Noise2Void	36.81 $\pm$ 1.83	0.8747 $\pm$ 0.0467	29.56 $\pm$ 7.14
	Neighbor2Neighbor	37.27 $\pm$ 1.76	0.8919 $\pm$ 0.0444	27.99 $\pm$ 6.47
	SDCNN	38.03 $\pm$ 1.22	0.9085 $\pm$ 0.0299	25.34 $\pm$ 3.90
	MM-Net	39.01 $\pm$ 1.41	0.9246 $\pm$ 0.0309	22.71 $\pm$ 4.09
	Noise2Sim	39.44 $\pm$ 1.44	0.9411 $\pm$ 0.0229	21.53 $\pm$ 3.81

TABLE II: Quantitative results (mean

\pm

std) of different self-supervised methods on the Mayo 2016 and Piglet datasets. These models are all directly trained on the corresponding single-volume LDCT data.

Fig. 5: Qualitative results of different self-supervised methods on the Real-World CBCT and ICASSP CBCT datasets. (a) NDCT (ground truth; unavailable for the Real-World CBCT Dataset), (b) LDCT, (c) Noise2Void, (d) Neighbor2Neighbor, (e) SDCNN, (f) MM-Net, (g) Noise2Sim, (h) SVB, (i) Noise2Match ( ours ).

Fig. 6: Qualitative results of selected ROIs for calculating CNR and Std (HU). The blue and yellow ROIs are used to calculate CNR, and the yellow ROIs are used to compute Std. (a) LDCT, (b) Noise2Void, (c) Neighbor2Neighbor, (d) SDCNN, (e) MM-Net, (f) Noise2Sim, (g) SVB, (h) Noise2Match ( ours ).

Fig. 7: Line-profile comparison across the airway boundary indicated by the green line segment in Fig. 6 (a1).

Methods	Real-World CBCT		ICASSP CBCT
Methods	CNR $\uparrow$	Std (HU) $\downarrow$	PSNR (dB) $\uparrow$	SSIM $\uparrow$	RMSE (HU) $\downarrow$	CNR $\uparrow$	Std (HU) $\downarrow$
LDCT	0.73	94.63	29.30 $\pm$ 0.66	0.7260 $\pm$ 0.0378	67.45 $\pm$ 6.60	0.90	138.06
Noise2Void	0.92	76.30	31.33 $\pm$ 0.97	0.8390 $\pm$ 0.0243	54.59 $\pm$ 5.63	1.46	82.92
Neighbor2Neighbor	0.95	72.61	31.37 $\pm$ 0.90	0.8466 $\pm$ 0.0194	53.03 $\pm$ 5.76	1.79	76.15
SDCNN	2.66	32.64	31.29 $\pm$ 1.16	0.8765 $\pm$ 0.0255	53.97 $\pm$ 7.87	3.87	41.33
MM-Net	1.03	69.53	31.84 $\pm$ 0.94	0.8655 $\pm$ 0.0293	49.50 $\pm$ 6.12	2.17	64.14

TABLE III: Quantitative results of different self-supervised methods on the Real-World CBCT and ICASSP CBCT datasets.

Fig. 8: Quantitative analysis of the inter-slice noise correlation before and after the correction operation on the Mayo 2016, Mayo 2020, Real-World CBCT, and ICASSP CBCT datasets.

Fig. 9: Representative visualization of the deformation fields learned by the inter-slice matching network on the Mayo 2016, Mayo 2020, Real-World CBCT, and ICASSP CBCT datasets.

Dataset	Jacobian determinant	Folding ratio
Mayo 2016	0.9999 $\pm$ 0.0495	0%
Mayo 2020	1.0001 $\pm$ 0.0727	0%
Real-World CBCT	1.0002 $\pm$ 0.0604	0%
ICASSP CBCT	1.0007 $\pm$ 0.0703	0%

TABLE IV: Jacobian statistics of the deformation fields estimated by the inter-slice matching network.

Fig. 10: Qualitative ablation results of different components on the test data from Mayo 2016 Dataset. (a) NDCT, (b) LDCT, (c) w/o $M_{\phi}$ , $\mathcal{L}_{Corr}$ , $\mathcal{L}_{Sm}$ , and $R.W.$ , (d) w/o $\mathcal{L}_{Sm}$ and $R.W.$ , (e) w/o $R.W.$ , (f) Noise2Match.

$M_{\phi}$	$\mathcal{L}_{Corr}$	$\mathcal{L}_{Sm}$	$R.W.$	PSNR (dB) $\uparrow$	SSIM $\uparrow$	RMSE (HU) $\downarrow$
-	-	-	-	37.86 $\pm$ 0.81	0.9229 $\pm$ 0.0211	24.83 $\pm$ 2.13
✓	✓	-	-	35.90 $\pm$ 0.91	0.8552 $\pm$ 0.0299	32.22 $\pm$ 3.53
✓	✓	✓	-	39.68 $\pm$ 0.85	0.9368 $\pm$ 0.0187	20.38 $\pm$ 2.02
✓	✓	✓	✓	40.08 $\pm$ 0.84	0.9410 $\pm$ 0.0167	19.89 $\pm$ 1.96

TABLE V: Quantitative ablation results of different components on the Mayo 2016 Dataset.

Fig. 11: Qualitative ablation results of different parameter k on the Mayo 2016, Mayo 2020, Real-World CBCT and ICASSP CBCT datasets.

	$k$	PSNR (dB) $\uparrow$	SSIM $\uparrow$	RMSE (HU) $\downarrow$
Mayo 2016	1	40.08 $\pm$ 0.84	0.9410 $\pm$ 0.0167	19.89 $\pm$ 1.96
	2	39.37 $\pm$ 0.96	0.9334 $\pm$ 0.0172	21.92 $\pm$ 2.27
	3	38.59 $\pm$ 0.97	0.9261 $\pm$ 0.0156	23.64 $\pm$ 2.64
	4	38.10 $\pm$ 1.18	0.9182 $\pm$ 0.0187	25.10 $\pm$ 3.52
Mayo 2020	1	42.88 $\pm$ 2.48	0.9648 $\pm$ 0.0189	15.32 $\pm$ 6.46
	2	40.93 $\pm$ 1.84	0.9587 $\pm$ 0.0150	19.02 $\pm$ 5.23

TABLE VI: Quantitative ablation results of different parameter

k

on the Mayo 2016, Mayo 2020 and ICASSP CBCT datasets.

Fig. 12: Analysis of the loss-reweighting mask on the Mayo 2016 training set. (a) Representative reweighting masks obtained with different $th$ values. (b) Slice-wise masked-out pixel ratio in a representative training volume using $th=80$ HU.

$th$	PSNR (dB) $\uparrow$	SSIM $\uparrow$	RMSE (HU) $\downarrow$
60	40.08 $\pm$ 0.82	0.9406 $\pm$ 0.0170	19.90 $\pm$ 1.96
70	40.07 $\pm$ 0.84	0.9409 $\pm$ 0.0169	19.92 $\pm$ 1.97
80	40.08 $\pm$ 0.84	0.9410 $\pm$ 0.0167	19.89 $\pm$ 1.96
90	40.05 $\pm$ 0.85	0.9411 $\pm$ 0.0152	19.98 $\pm$ 2.01
100	40.07 $\pm$ 0.83	0.9407 $\pm$ 0.0173	19.91 $\pm$ 1.96

TABLE VII: Quantitative ablation results of different parameter

th

on the Mayo 2016 Dataset.

Fig. 13: Empirical evaluation of the equivariance condition in Theorem 1 using a representative adjacent-slice pair from the Mayo 2016 training set. From left to right: $f_{\theta}(x_{i})\circ\hat{T}_{i,k^{+}}$ , $f_{\theta}(x_{i}\circ\hat{T}_{i,k^{+}})$ , and the absolute difference between the two results.

Self-Supervised Image Denoising

Self-supervised image denoising learns to remove noise from images without requiring clean training targets.

See all 25 papers

Sep 1, 2026

Physics-Driven Independent Pair Generation for Iterative Self-Supervised Low-Dose CT Denoising

Low-dose computed tomography (LDCT) measurements contain mixed Poisson-Gaussian noise. However, most self-supervised methods rely on generic image statistics and do not explicitly model this noise, which may limit their ability to effectively suppress realistic LDCT noise. To address this issue, we propose a physics-driven framework with cross-domain iteration for self-supervised LDCT denoising. The proposed framework proceeds in three main steps. First, a learned sinogram prior and the LDCT noise model guide posterior inference of photon counts, enabling separation of the Poisson and Gaussian components. Second, the separated Poisson and Gaussian components are respectively processed by binomial thinning and Gaussian data thinning to construct two branches, and residual scaling matches each branch's noise level to that of the observation, yielding a training pair with approximately independent noise realizations from one low-dose measurement. Finally, the pair is used to train an image-domain network whose forward-projected outputs update the prior. Through cross-domain iteration, the prior and the training pair are progressively refined while maintaining consistency with CT acquisition physics. Experiments on simulated data from AAPM, LIDC-IDRI, and LoDoPaB-CT and on real LDCT data show consistent gains over the evaluated self-supervised baselines across dose levels, with performance comparable to the evaluated supervised baseline.

Xianlei Han, Shaoyu Wang, Jiancheng Fang +2

School of Mathematics and Computer Sciences, Nanchang University, Nanchang 330031, China · School of Information Engineering, Nanchang University, Nanchang 330031, China · School of Biomedical Engineering, Sun Yat-sen University, Shenzhen 510275, Guangdong, China

Self-Supervised Image Denoising Self-Supervised Learning Computed Tomography LDCT Denoising

Ask PDF

May 11, 2026

FrequencyCT: Frequency Domain Self-supervised Low-dose CT Denoising

Despite extensive research on computed tomography (CT) denoising, few studies exploit projection-domain data characteristics to mitigate noise correlation. To bridge this gap, this work proposes FrequencyCT, the first zero-shot self-supervised method for pseudo-sample generation in the frequency domain for low-dose CT denoising. Specifically, by exploiting the distinct frequency-domain distributions of noise and true signal, a regional low-frequency anchoring technique is proposed. Applying phase-preserving noise and mask perturbations to the high-frequency region generates pseudo-samples for self-supervision. Driven by the exponential correlation between noise variance of noisy projections and the underlying true signal, consistent data truncation is applied to the generated samples to stabilize optimization gradients. Evaluation results on multiple public and real datasets confirm the clinical application potential of this research, which provides an innovative perspective for the field of denoising. The code is available at: https://github.com/yqx7150/FrequencyCT.

Guoquan Wei, Liu Shi, Chong Chen +1

School of Information Engineering, Nanchang University, Nanchang 330031, China · SKLMS, ICMSEC, Academy of Mathematics and Systems Science, Chinese Academy of Sciences, Beijing 100190, China

Self-Supervised Image Denoising Self-Supervised Learning Frequency-Domain Image Processing Computed Tomography LDCT Denoising Image Denoising

Ask PDF

Feb 25, 2026

PatchDenoiser: Parameter-efficient multi-scale patch learning and fusion denoiser for Low-dose CT imaging

Low-dose CT images are essential for reducing radiation exposure in cancer screening, pediatric imaging, and longitudinal monitoring protocols, but their quality is often degraded by noise from low-dose acquisition, patient motion, or scanner limitations, affecting both clinical interpretation and downstream analysis. Traditional filtering approaches often over-smooth and lose fine anatomical details, while deep learning methods, including CNNs, GANs, and transformers, may struggle to preserve such details or require large, computationally expensive models, limiting clinical practicality. We propose PatchDenoiser, a lightweight, energy-efficient multi-scale patch-based denoising framework. It decomposes denoising into local texture extraction and global context aggregation, fused via a spatially aware patch fusion strategy. This design enables effective noise suppression while preserving fine structural and anatomical details. PatchDenoiser is ultra-lightweight, with far fewer parameters and lower computational complexity than CNN, GAN, and transformer based denoisers. On the 2016 Mayo Low-Dose CT dataset, PatchDenoiser consistently outperforms state-of-the-art CNN- and GAN-based methods in PSNR and SSIM. It is robust to variations in slice thickness, reconstruction kernels, and HU windows, generalizes across scanners without fine-tuning, and reduces parameters by ~9x and energy consumption per inference by ~27x compared with conventional CNN denoisers. PatchDenoiser thus provides a practical, scalable, and computationally efficient solution for medical image denoising, balancing performance, robustness, and clinical deployability.

Jitindra Fartiyal, Pedro Freire, Sergei K. Turitsyn +1

Aston Institute of Photonic Technologies, Aston University, Aston Triangle, Birmingham, B4 7ET, West Midlands, United Kingdom

LDCT Denoising Image Denoising Image Restoration

Ask PDF

Abstract

Self-Supervised Image Denoising

Explore similar work

Physics-Driven Independent Pair Generation for Iterative Self-Supervised Low-Dose CT Denoising

FrequencyCT: Frequency Domain Self-supervised Low-dose CT Denoising

PatchDenoiser: Parameter-efficient multi-scale patch learning and fusion denoiser for Low-dose CT imaging