Author name cluster

Fumin Guo

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

4 papers

1 author row

AAAI Conference 2026 Conference Paper

DFMN: A Dual-feet Matching Network with Hybrid Transformer-based Feature Extractor for Unsupervised Deformable Medical Image Registration

Liwen Li
Xinrui Guo
Wentao Guo
Shunqi Yang
Fumin Guo

Deformable medical image registration is essential in medical image analyses. Recent transformer-based registration methods have achieved high registration accuracy. However, these methods often rely on patch embedding at the beginning of encoding, resulting in limited ability to capture detailed anatomical structural information in the images and explore local semantic relationships within individual patches. Here, we proposed a novel Dual-feet Encoder (DFEnc) to asynchronously model semantic information from moving and fixed images at various scales through two separate branches in three steps. For each step, features from adjacent resolution levels were processed by a Single Step Hybrid Extractor (SSHExt), which performed patch convolution to preserve local information, followed by several transformer blocks to capture global context. Dense connections were employed to enhance semantic awareness across adjacent feature resolution levels. Additionally, we introduced a Feature Fusion-based Decoder (FFDec) to progressively fuse features related to the fixed and moving images and to generate intermediate deformation fields at each stage, enabling accurate image alignment through stepwise warping and alignment refinement. Extensive ablation studies demonstrated the effectiveness of the proposed DFEnc, SSHExt, and FFDec. Compared to a state-of-the-art AutoFuse-Trans method, our approach yielded improvements in Dice of 1.14%, 1.77%, and 4.47% on the ACDC, OASIS, and Abdomen CT datasets, respectively, while maintaining relatively low computational cost. These results suggest the utility of the proposed approach for broad research and clinical applications.

PDF Details DOI

JBHI Journal 2025 Journal Article

VWV-SSL: Carotid vessel-wall-volume segmentation via sequence structural similarity and augmentation consistency-based self-supervised learning

Ran Zhou
Furong Wang
Jing Ding
Zhongwei Huang
Haitao Gan
Fumin Guo
Aaron Fenster

Vessel wall volume (VWV) is a critical three dimensional ultrasound metric used to assess the progression and regression of carotid atherosclerosis. Ac curate measurement of VWV requires the segmentation of the media-adventitia boundary (MAB) and the lumen intima boundary (LIB) of the carotid arteries. Although deep learning methods can automatically segment the MAB and LIB and quantify VWV, they rely heavily on a large dataset with annotated images for training, which is time consuming and labor-intensive. Self-supervised learning (SSL) provides a possible solution to this challenge. However, existing SSL methods do not consider the similarities in the image sequences of 3D ultrasound. This paper proposes a novel SSL algorithm, named VWV-SSL, for 3D carotid ultrasound (3DUS) image segmentation to generate VWV measurement. VWV-SSL utilizes the sequence structural similarity and strong-weak augmented feature consistency of carotid ultrasound images to conduct the self-supervised task, which enables the networks to better learn the feature presentations of the vessel in the self-supervised task training. We applied VWV-SSL on the widely used 3D U-Net and evaluated it on 1158 3D US (579 of the common carotid artery and 579 of the bifurcation) from250subjects. Comparedtobaselinenetworks, our SSL method showed a significant improvement in segmentation performance when trained on a small number of labeled images (n = 15, 45 and 75 subjects). Moreover, the performance of VWV-SSL was superior to that of state-of-art SSL algorithms. These results indicate that our method can improve the performance of 3D U-Net when trained on a small number of labeled images, suggesting that VWV SSL could be applied in clinical practice to monitor the progression of atherosclerosis.

Details DOI

JBHI Journal 2021 Journal Article

Deep Learning-Based Measurement of Total Plaque Area in B-Mode Ultrasound Images

Ran Zhou
Fumin Guo
M. Reza Azarpazhooh
Samineh Hashemi
Xinyao Cheng
J. David Spence
Mingyue Ding
Aaron Fenster

Measurement of total-plaque-area (TPA) is important for determining long term risk for stroke and monitoring carotid plaque progression. Since delineation of carotid plaques is required, a deep learning method can provide automatic plaque segmentations and TPA measurements; however, it requires large datasets and manual annotations for training with unknown performance on new datasets. A UNet++ ensemble algorithm was proposed to segment plaques from 2D carotid ultrasound images, trained on three small datasets (n = 33, 33, 34 subjects) and tested on 44 subjects from the SPARC dataset (n = 144, London, Canada). The ensemble was also trained on the entire SPARC dataset and tested with a different dataset (n = 497, Zhongnan Hospital, China). Algorithm and manual segmentations were compared using Dice-similarity-coefficient (DSC), and TPAs were compared using the difference ( $\Delta$ TPA), Pearson correlation coefficient ( r ) and Bland-Altman analyses. Segmentation variability was determined using the intra-class correlation coefficient (ICC) and coefficient-of-variation (CoV). For 44 SPARC subjects, algorithm DSC was 83. 3–85. 7%, and algorithm TPAs were strongly correlated ( r = 0. 985–0. 988; p $^2$ using the three training datasets. Algorithm ICC for TPAs (ICC = 0. 996) was similar to intra- and inter-observer manual results (ICC = 0. 977, 0. 995). Algorithm CoV = 6. 98% for plaque areas was smaller than the inter-observer manual CoV (7. 54%). For the Zhongnan dataset, DSC was 88. 6% algorithm and manual TPAs were strongly correlated ( r = 0. 972, p $\Delta$ TPA = −0. 44 $\pm$ 4. 05 mm $^2$ and ICC = 0. 985. The proposed algorithm trained on small datasets and segmented a different dataset without retraining with accuracy and precision that may be useful clinically and for research.

Details DOI

JBHI Journal 2021 Journal Article

Left Ventricle Quantification Challenge: A Comprehensive Comparison and Evaluation of Segmentation and Regression for Mid-Ventricular Short-Axis Cardiac MR Data

Wufeng Xue
Jiahui Li
Zhiqiang Hu
Eric Kerfoot
James Clough
Ilkay Oksuz
Hao Xu
Vicente Grau

Automatic quantification of the left ventricle (LV) from cardiac magnetic resonance (CMR) images plays an important role in making the diagnosis procedure efficient, reliable, and alleviating the laborious reading work for physicians. Considerable efforts have been devoted to LV quantification using different strategies that include segmentation-based (SG) methods and the recent direct regression (DR) methods. Although both SG and DR methods have obtained great success for the task, a systematic platform to benchmark them remains absent because of differences in label information during model learning. In this paper, we conducted an unbiased evaluation and comparison of cardiac LV quantification methods that were submitted to the Left Ventricle Quantification (LVQuan) challenge, which was held in conjunction with the Statistical Atlases and Computational Modeling of the Heart (STACOM) workshop at the MICCAI 2018. The challenge was targeted at the quantification of 1) areas of LV cavity and myocardium, 2) dimensions of the LV cavity, 3) regional wall thicknesses (RWT), and 4) the cardiac phase, from mid-ventricle short-axis CMR images. First, we constructed a public quantification dataset Cardiac-DIG with ground truth labels for both the myocardium mask and these quantification targets across the entire cardiac cycle. Then, the key techniques employed by each submission were described. Next, quantitative validation of these submissions were conducted with the constructed dataset. The evaluation results revealed that both SG and DR methods can offer good LV quantification performance, even though DR methods do not require densely labeled masks for supervision. Among the 12 submissions, the DR method LDAMT offered the best performance, with a mean estimation error of 301 mm $^2$ for the two areas, 2. 15 mm for the cavity dimensions, 2. 03 mm for RWTs, and a 9. 5% error rate for the cardiac phase classification. Three of the SG methods also delivered comparable performances. Finally, we discussed the advantages and disadvantages of SG and DR methods, as well as the unsolved problems in automatic cardiac quantification for clinical practice applications.

Details DOI