Xiaohuan Cao
Research Subjects
Xiaohuan Cao has a total of 90 co-authors. Here are the top co-authors by number of shared publications:
- Dinggang Shen
ShanghaiTech University
16 shared publications Zhong Xue
Shanghai United Imaging Intelligence Co. Ltd
6 shared publications- Zhong Xue
Shanghai United Imaging Intelligence Company Ltd., Shanghai, China
4 shared publications - Dong Nie
University of North Carolina at Chapel Hill, NC, USA
3 shared publications Jianlong Nie
ShanghaiTech University
3 shared publicationsBoyang Zhang
Shanghai United Imaging Intelligence Co., Ltd., Shanghai, China
2 shared publications
Latest Publications
- 26 September 2026
- Published by Springer Nature
- p. 43-53
Anatomy-Aware Standard Plane Localization in 3D Ultrasound with Geometric Constraints
- 26 September 2026
- Published by Springer Nature
- p. 43-53
Anatomy-Aware Standard Plane Localization in 3D Ultrasound with Geometric Constraints
- Open Access
- 5 June 2026
- Published by Springer Nature in npj Digital Medicine
- Vol. 9 (1)
Learning like a radiologist: a medical vision-language model for radiological image analysis via curriculum learning
Medical vision-language models (MVLMs) offer promise in clinical practice but face limitations in generalizability, data quality, and clinically meaningful evaluation. We propose RadiSim-CL, an MVLM trained via curriculum learning by simulating the three-phase pathway of a radiologist: foundational knowledge understanding, anatomical knowledge, and advanced diagnostic reasoning. To support this, we curate RadiSim, a 12-million image-text pair dataset aligned to these phases. We evaluate the model using a five-stage coarse-to-fine validation framework: (1) modality recognition, (2) anatomical recognition, (3) anatomical localization, (4) abnormality and disease diagnosis, and (5) disease differentiation and grading. This framework spans 24 zero-shot subtasks across MR, CT, and DR imaging. RadiSim-CL achieves comparable performance to state-of-the-art baselines in both foundational and anatomical tasks, and demonstrates superior capabilities in complex reasoning (e.g., an AUC of 0.953 for brain tumor diagnosis and an accuracy of 0.764 for meningioma grading). Ablation studies further confirm the curriculum's effectiveness. RadiSim-CL thus offers a scalable, clinically aligned solution to enhance diagnostic precision.
- 5 June 2026
- Published by Springer Nature in npj Digital Medicine
- Vol. 9 (1)
Open AccessLearning like a radiologist: a medical vision-language model for radiological image analysis via curriculum learning
Medical vision-language models (MVLMs) offer promise in clinical practice but face limitations in generalizability, data quality, and clinically meaningful evaluation. We propose RadiSim-CL, an MVLM trained via curriculum learning by simulating the three-phase pathway of a radiologist: foundational knowledge understanding, anatomical knowledge, and advanced diagnostic reasoning. To support this, we curate RadiSim, a 12-million image-text pair dataset aligned to these phases. We evaluate the model using a five-stage coarse-to-fine validation framework: (1) modality recognition, (2) anatomical recognition, (3) anatomical localization, (4) abnormality and disease diagnosis, and (5) disease differentiation and grading. This framework spans 24 zero-shot subtasks across MR, CT, and DR imaging. RadiSim-CL achieves comparable performance to state-of-the-art baselines in both foundational and anatomical tasks, and demonstrates superior capabilities in complex reasoning (e.g., an AUC of 0.953 for brain tumor diagnosis and an accuracy of 0.764 for meningioma grading). Ablation studies further confirm the curriculum's effectiveness. RadiSim-CL thus offers a scalable, clinically aligned solution to enhance diagnostic precision.
- 23 October 2024
- Published by Springer Nature
- p. 191-200
SDF-Net: A Hybrid Detection Network for Mediastinal Lymph Node Detection on Contrast CT Images
- 23 October 2024
- Published by Springer Nature
- p. 191-200
SDF-Net: A Hybrid Detection Network for Mediastinal Lymph Node Detection on Contrast CT Images
- 4 October 2024
- Published by Springer Nature
- p. 662-672
Hierarchical Symmetric Normalization Registration Using Deformation-Inverse Network
- 4 October 2024
- Published by Springer Nature
- p. 662-672
Hierarchical Symmetric Normalization Registration Using Deformation-Inverse Network
- Open Access
SDF-Net: A Hybrid Detection Network for Mediastinal Lymph Node Detection on Contrast CT Images
Accurate lymph node detection and quantification are crucial for cancer diagnosis and staging on contrast-enhanced CT images, as they impact treatment planning and prognosis. However, detecting lymph nodes in the mediastinal area poses challenges due to their low contrast, irregular shapes and dispersed distribution. In this paper, we propose a Swin-Det Fusion Network (SDF-Net) to effectively detect lymph nodes. SDF-Net integrates features from both segmentation and detection to enhance the detection capability of lymph nodes with various shapes and sizes. Specifically, an auto-fusion module is designed to merge the feature maps of segmentation and detection networks at different levels. To facilitate effective learning without mask annotations, we introduce a shape-adaptive Gaussian kernel to represent lymph node in the training stage and provide more anatomical information for effective learning. Comparative results demonstrate promising performance in addressing the complex lymph node detection problem.
Open AccessSDF-Net: A Hybrid Detection Network for Mediastinal Lymph Node Detection on Contrast CT Images
Accurate lymph node detection and quantification are crucial for cancer diagnosis and staging on contrast-enhanced CT images, as they impact treatment planning and prognosis. However, detecting lymph nodes in the mediastinal area poses challenges due to their low contrast, irregular shapes and dispersed distribution. In this paper, we propose a Swin-Det Fusion Network (SDF-Net) to effectively detect lymph nodes. SDF-Net integrates features from both segmentation and detection to enhance the detection capability of lymph nodes with various shapes and sizes. Specifically, an auto-fusion module is designed to merge the feature maps of segmentation and detection networks at different levels. To facilitate effective learning without mask annotations, we introduce a shape-adaptive Gaussian kernel to represent lymph node in the training stage and provide more anatomical information for effective learning. Comparative results demonstrate promising performance in addressing the complex lymph node detection problem.
- 7 June 2024
- Published by Elsevier in Neural Networks
- Vol. 178, 106426
Detail-preserving image warping by enforcing smooth image sampling
- 7 June 2024
- Published by Elsevier in Neural Networks
- Vol. 178, 106426
Detail-preserving image warping by enforcing smooth image sampling
- Open Access
- 1 February 2024
- Published by AME Publishing Company in Quantitative Imaging in Medicine and Surgery
- Vol. 14 (2) , 1803-1819
Development and validation of a radiomics model based on T2-weighted imaging for predicting the efficacy of high intensity focused ultrasound ablation in uterine fibroids
Background: The heterogeneity of uterine fibroids in magnetic resonance imaging (MRI) is complex for a subjective visual evaluation, therefore it is difficult for an accurate prediction of the efficacy of high intensity focused ultrasound (HIFU) ablation in fibroids before the treatment. The purpose of this study was to set up a radiomics model based on MRI T2-weighted imaging (T2WI) for predicting the efficacy of HIFU ablation in uterine fibroids, and it would be used in preoperative screening of the fibroids for achieving high non-perfused volume ratio (NPVR). Methods: A total of 178 patients with uterine fibroids were consecutively enrolled and treated with ultrasound-guided HIFU under conscious sedation between February 2017 and December 2021. Among them, 96 patients with 108 uterine fibroids with high ablation efficacy (NPVR ≥80%, h_NPVR) and 82 patients with 92 fibroids with lower ablation efficacy (NPVR Results: The logistic regression model showed that the mean area under the curve (AUC) of the training set was 0.817 [95% confidence interval (CI): 0.755–0.882], and the testing set was 0.805 (95% CI: 0.670–0.941), respectively, which indicated a strong classification ability. The Delong test showed that there was no significant difference in the area under the ROC curve between the training set and testing set (P>0.05). Conclusions: The radiomics model based on T2WI is feasible and effective for predicting the efficacy of HIFU ablation in treatment of uterine fibroids.
- 1 February 2024
- Published by AME Publishing Company in Quantitative Imaging in Medicine and Surgery
- Vol. 14 (2) , 1803-1819
Open AccessDevelopment and validation of a radiomics model based on T2-weighted imaging for predicting the efficacy of high intensity focused ultrasound ablation in uterine fibroids
Background: The heterogeneity of uterine fibroids in magnetic resonance imaging (MRI) is complex for a subjective visual evaluation, therefore it is difficult for an accurate prediction of the efficacy of high intensity focused ultrasound (HIFU) ablation in fibroids before the treatment. The purpose of this study was to set up a radiomics model based on MRI T2-weighted imaging (T2WI) for predicting the efficacy of HIFU ablation in uterine fibroids, and it would be used in preoperative screening of the fibroids for achieving high non-perfused volume ratio (NPVR). Methods: A total of 178 patients with uterine fibroids were consecutively enrolled and treated with ultrasound-guided HIFU under conscious sedation between February 2017 and December 2021. Among them, 96 patients with 108 uterine fibroids with high ablation efficacy (NPVR ≥80%, h_NPVR) and 82 patients with 92 fibroids with lower ablation efficacy (NPVR Results: The logistic regression model showed that the mean area under the curve (AUC) of the training set was 0.817 [95% confidence interval (CI): 0.755–0.882], and the testing set was 0.805 (95% CI: 0.670–0.941), respectively, which indicated a strong classification ability. The Delong test showed that there was no significant difference in the area under the ROC curve between the training set and testing set (P>0.05). Conclusions: The radiomics model based on T2WI is feasible and effective for predicting the efficacy of HIFU ablation in treatment of uterine fibroids.
- 5 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 28 (4) , 2163-2174
Structure-Aware Registration Network for Liver DCE-CT Images
Image registration of liver dynamic contrast-enhanced computed tomography (DCE-CT) is crucial for diagnosis and image-guided surgical planning of liver cancer. However, intensity variations due to the flow of contrast agents combined with complex spatial motion induced by respiration brings great challenge to existing intensity-based registration methods. To address these problems, we propose a novel structure-aware registration method by incorporating structural information of related organs with segmentation-guided deep registration network. Existing segmentation-guided registration methods only focus on volumetric registration inside the paired organ segmentations, ignoring the inherent attributes of their anatomical structures. In addition, such paired organ segmentations are not always available in DCE-CT images due to the flow of contrast agents. Different from existing segmentation-guided registration methods, our proposed method extracts structural information in hierarchical geometric perspectives of line and surface. Then, according to the extracted structural information, structure-aware constraints are constructed and imposed on the forward and backward deformation field simultaneously. In this way, all available organ segmentations, including unpaired ones, can be fully utilized to avoid the side effect of contrast agent and preserve the topology of organs during registration. Extensive experiments on an in-house liver DCE-CT dataset and a public LiTS dataset show that our proposed method can achieve higher registration accuracy and preserve anatomical structure more effectively than state-of-the-art methods.
- 5 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 28 (4) , 2163-2174
Structure-Aware Registration Network for Liver DCE-CT Images
Image registration of liver dynamic contrast-enhanced computed tomography (DCE-CT) is crucial for diagnosis and image-guided surgical planning of liver cancer. However, intensity variations due to the flow of contrast agents combined with complex spatial motion induced by respiration brings great challenge to existing intensity-based registration methods. To address these problems, we propose a novel structure-aware registration method by incorporating structural information of related organs with segmentation-guided deep registration network. Existing segmentation-guided registration methods only focus on volumetric registration inside the paired organ segmentations, ignoring the inherent attributes of their anatomical structures. In addition, such paired organ segmentations are not always available in DCE-CT images due to the flow of contrast agents. Different from existing segmentation-guided registration methods, our proposed method extracts structural information in hierarchical geometric perspectives of line and surface. Then, according to the extracted structural information, structure-aware constraints are constructed and imposed on the forward and backward deformation field simultaneously. In this way, all available organ segmentations, including unpaired ones, can be fully utilized to avoid the side effect of contrast agent and preserve the topology of organs during registration. Extensive experiments on an in-house liver DCE-CT dataset and a public LiTS dataset show that our proposed method can achieve higher registration accuracy and preserve anatomical structure more effectively than state-of-the-art methods.
- 15 October 2023
- Published by Springer Nature
- p. 243-252
A Bone Lesion Identification Network (BLIN) in CT Images with Weakly Supervised Learning
- 15 October 2023
- Published by Springer Nature
- p. 243-252
A Bone Lesion Identification Network (BLIN) in CT Images with Weakly Supervised Learning