Shuai Zhang

Shuai Zhang

Auto-generated profile, based on publicly available information

Shuai Zhang

UCL Hawkes Institute, University College London, London, UK

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
14 December 2020 in IEEE Transactions on Medical Robotics and Bionics
Verified
Journal Article
Closed access
SZShuai Zhang
Shuai Zhang
LZLiang Zhao
Liang Zhao
SHShoudong Huang
Shoudong Huang
MYMenglong Ye
Menglong Ye
QHQi Hao
Qi Hao
This paper presents a framework for 3D reconstruction of colonic surface using stereo colonoscopic images. Due to the limited overlaps between consecutive frames and the nonexistence of large loop closures during a normal screening colonoscopy, the state-of-art simultaneous localization and mapping (SLAM) algorithms cannot be directly applied to this scenario, thus a colon model segmented from CT scans is used together with the colonosocopic images to achieve the colon 3D reconstruction with high accuracy. The proposed framework includes 3D scan (point cloud with RGB information) reconstruction from stereo images, a visual odometry (VO) based camera pose initialization module, a 3D registration scheme for matching texture scans to the segmented colon model, and a barycentric-based texture rendering module for mapping textures from colonoscopic images to the reconstructed colonic surface. A realistic simulator is developed using Unity to simulate the procedures of colonoscopy and used to provide experimental datasets in different scenarios. Experimental results demonstrate the good performance of the proposed 3D colonic surface reconstruction method in terms of accuracy and robustness. Currently, the framework requires a pre-operative colon model as the template for colon reconstruction and can reconstruct 3D colon maps when the colon has no large deformation and the colon structure is clearly visible. The datasets used in this paper and the developed simulator are made publicly available for other researchers to use.

28 March 2025 in IEEE Transactions on Image Processing
Verified
Journal Article
Closed access
XZXiaoqing Zhang
Xiaoqing Zhang
ZXZunjie Xiao
Zunjie Xiao
JMJingzhe Ma
Jingzhe Ma
XWXiao Wu
Xiao Wu
JZJilu Zhao
Jilu Zhao
JLJiang Liu
Jiang Liu
Salient and small lesions (e.g., microaneurysms on fundus) both play significant roles in real-world disease diagnosis under medical image examinations. Although deep neural networks (DNNs) have achieved promising medical image classification performance, they often have limitations in capturing both salient and small lesion information, restricting performance improvement in imbalanced medical image classification. Recently, with the advent of DNN-based style transfer in medical image generation, the roles of clinical styles have attracted great interest, as they are crucial indicators of lesions. Motivated by this observation, we propose a novel Adaptive Dual-Axis Style-based Recalibration (ADSR) module, leveraging the potential of clinical styles to guide DNNs in effectively learning salient and small lesion information from a dual-axis perspective. ADSR first emphasizes salient lesion information via global style-based adaptation, then captures small lesion information with pixel-wise style-based fusion. We construct an ADSR-Net for imbalanced medical image classification by stacking multiple ADSR modules. Additionally, DNNs typically adopt cross-entropy loss for parameter optimization, which ignores the impacts of class-wise predicted probability distributions. To address this, we introduce a new Class-wise Statistics Loss (CWS) combined with CE to further boost imbalanced medical image classification results. Extensive experiments on five imbalanced medical image datasets demonstrate not only the superiority of ADSR-Net and CWS over state-of-the-art (SOTA) methods but also their improved confidence calibration results. For example, ADSR-Net with the proposed loss significantly outperforms CABNet50 by 21.39% and 27.82% in F1 and B-ACC while reducing 3.31% and 4.57% in ECE and BS on ISIC2018.

3 October 2024
Verified
Book Chapter
Closed access
PDF file
SBSierra Bonilla
Sierra Bonilla
SZShuai Zhang
Shuai Zhang
DPDimitrios Psychogyios
Dimitrios Psychogyios
DSDanail Stoyanov
Danail Stoyanov
FVFrancisco Vasconcelos
Francisco Vasconcelos
SBSophia Bano
Sophia Bano

1 October 2018
Verified
Conference Paper
Closed access
SZShuai Zhang
Shuai Zhang
RHRuihua Han
Ruihua Han
WHWankuan Huang
Wankuan Huang
SWShuaijun Wang
Shuaijun Wang
QHQi Hao
Qi Hao
In this paper, we propose an improved UWB based indoor localization system using Bayesian filtering techniques. The system contains two key components: (1) miniaturized, high updating rate and highly reconfigurable UWB sensors with a linear regression model to calibrate range measurement errors; (2) a set of Bayesian filters which can improve the localization precision by utilizing the spatial correlation between the stationary UWB base stations and the mobile UWB station. Furthermore, a novel measurement transform is proposed to reduce the computational complexity. Experiments are performed in an indoor environment with the ground truth obtained by the motion capture system to validate and evaluate the proposed indoor localization system.

14 August 2024 in Bone & Joint Open
Verified
Journal Article
Open access
PDF file
AFAndreas Fontalis
Andreas Fontalis
BZBaixiang Zhao
Baixiang Zhao
PPPierre Putzeys
Pierre Putzeys
FMFabio Mancino
Fabio Mancino
SZShuai Zhang
Shuai Zhang
FHFares S. Haddad
Fares S. Haddad
Aims: Precise implant positioning, tailored to individual spinopelvic biomechanics and phenotype, is paramount for stability in total hip arthroplasty (THA). Despite a few studies on instability prediction, there is a notable gap in research utilizing artificial intelligence (AI). The objective of our pilot study was to evaluate the feasibility of developing an AI algorithm tailored to individual spinopelvic mechanics and patient phenotype for predicting impingement.Methods: This international, multicentre prospective cohort study across two centres encompassed 157 adults undergoing primary robotic arm-assisted THA. Impingement during specific flexion and extension stances was identified using the virtual range of motion (ROM) tool of the robotic software. The primary AI model, the Light Gradient-Boosting Machine (LGBM), used tabular data to predict impingement presence, direction (flexion or extension), and type. A secondary model integrating tabular data with plain anteroposterior pelvis radiographs was evaluated to assess for any potential enhancement in prediction accuracy.Results: We identified nine predictors from an analysis of baseline spinopelvic characteristics and surgical planning parameters. Using fivefold cross-validation, the LGBM achieved 70.2% impingement prediction accuracy. With impingement data, the LGBM estimated direction with 85% accuracy, while the support vector machine (SVM) determined impingement type with 72.9% accuracy. After integrating imaging data with a multilayer perceptron (tabular) and a convolutional neural network (radiograph), the LGBM’s prediction was 68.1%. Both combined and LGBM-only had similar impingement direction prediction rates (around 84.5%).Conclusion: This study is a pioneering effort in leveraging AI for impingement prediction in THA, utilizing a comprehensive, real-world clinical dataset. Our machine-learning algorithm demonstrated promising accuracy in predicting impingement, its type, and direction. While the addition of imaging data to our deep-learning algorithm did not boost accuracy, the potential for refined annotations, such as landmark markings, offers avenues for future enhancement. Prior to clinical integration, external validation and larger-scale testing of this algorithm are essential.Cite this article: Bone Jt Open 2024;5(8):671–680.

1 September 2025 in IEEE Robotics and Automation Letters
Verified
Journal Article
Closed access
XWXu Wang
Xu Wang
Jialang XuJialang Xu
Jialang Xu
SZShuai Zhang
Shuai Zhang
BHBaoru Huang
Baoru Huang
DSDanail Stoyanov
Danail Stoyanov
EMEvangelos Mazomenos
Evangelos Mazomenos
Stereo disparity estimation is crucial for obtaining depth information in robot-assisted minimally invasive surgery (RAMIS). While current deep learning methods have made significant advancements, challenges remain in achieving an optimal balance between accuracy, robustness, and inference speed. To address these challenges, we propose the StereoMamba architecture, which is specifically designed for stereo disparity estimation in RAMIS. Our approach is based on a novel Feature Extraction Mamba (FE-Mamba) module, which enhances long-range spatial dependencies both within and across stereo images. To effectively integrate multi-scale features from FE-Mamba, we then introduce a novel Multidimensional Feature Fusion (MFF) module. Experiments against the state-of-the-art on the ex-vivo SCARED benchmark demonstrate that StereoMamba achieves superior performance on EPE of 2.64 px and depth MAE of 2.55 mm, the second-best performance on Bad2 of 41.49% and Bad3 of 26.99%, while maintaining an inference speed of 21.28 FPS for a pair of high-resolution images (1280 × 1024), striking the optimum balance between accuracy, robustness, and efficiency. Furthermore, by comparing synthesized right images, generated from warping left images using the generated disparity maps, with the actual right image, StereoMamba achieves the best average SSIM (0.8970) and PSNR (16.0761), exhibiting strong zero-shot generalization on the in-vivo RIS2017 and StereoMIS datasets.