MXMengsu Xiao

Mengsu Xiao

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Mengsu Xiao has a total of 75 co-authors. Here are the top co-authors by number of shared publications:

  • Qingli Zhu

    Department of Ultrasound, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, China

    7 shared publications
  • Jing Zhang

    Department of Ultrasound, Peking Union Medical College Hospital, Beijing, China

    3 shared publications
  • Hong Yang

    Key Laboratory of Endocrinology of National Health Commission, Department of Endocrinology, Peking Union Medical College Hospital, Chinese Academy of Medical Science and Peking Union Medical College, Beijing 100730, China

    2 shared publications
  • Jing Qin

    Department of Ultrasound, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences, Beijing, China

    2 shared publications
  • Ke Lv

    Department of Ultrasound, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China

    2 shared publications
  • Li Ma

    Department of Ultrasound, Peking Union Medical College Hospital, Beijing 100730, China

    2 shared publications

Latest Publications

  • Open Access

    Multimodal Deep Learning-Based Classification of Breast Non-Mass Lesions Using Gray Scale and Color Doppler Ultrasound

    by Tianjiao Wang, Qingli Zhu, Tianxiang Yu, Denis Leonov, Xinran Shi, Zhuhuang Zhou, Ke Lv, Mengsu Xiao, Jianchu Li

    Objectives: To propose a multimodal deep learning method for the classification of benign and malignant breast non-mass lesions (NMLs) using grayscale and color Doppler ultrasound and to compare the performance of multi-modality and single-modality breast ultrasound (BUS) models. Methods: This retrospective study collected 248 pathologically confirmed NMLs from 241 female patients comprising grayscale and color Doppler BUS images from March 2018 to November 2024. Three types of convolutional neural networks (CNNs), including ResNet50, ResNet18, and VGG16, were evaluated as single-modality (grayscale or color Doppler) models via five-fold cross-validations. The optimal model for each single-modality approach was chosen as the backbone network for multimodal deep learning. Features extracted from grayscale and color Doppler BUS images were then concatenated to predict the probabilities of benignity and malignancy. The diagnostic efficacy of the multi-modality BUS models was comparatively evaluated against single-modality counterparts. Results: The single-modality VGG16 models outperformed the other two CNN types for both grayscale and color Doppler BUS using five-fold cross-validations. Additionally, single-modality grayscale models outperformed single-modality color Doppler models. With a mean accuracy of 91.54%, sensitivity of 94.15%, specificity of 87.30%, F1 score of 0.93, and area under the receiver operating characteristic curve (AUC) of 0.96, the multimodal VGG16 models performed better than single-modality counterparts. Conclusions: VGG 16-based multimodal ultrasound deep learning showed excellent diagnostic efficacy in distinguishing between benign and malignant NMLs, indicating therapeutic potential to help radiologists assess NMLs.


  • Open Access

    The Sonographic Characteristics of Unicentric Castleman Disease - a single-center Retrospective Study

    by Zihan Liu, Zihan Niu, Yuhan Gao, Mengsu Xiao, Ying Wang, Qingli Zhu, Lu Zhang

    Background Unicentric Castleman disease (UCD) is a rare group of non-neoplastic lymphoproliferative disorders. This study aims to summarize the specific ultrasonic manifestations of UCD. Methods This retrospective study included patients who underwent preoperative ultrasound for enlarged lymph nodes and were later diagnosed with UCD between January 2016 and March 2024. Ultrasound features, including lymph node size, cortical characteristics, corticomedullary interface, hyperechoic regions, and Doppler flow signals, were recorded. Pathological types were classified as hyaline vascular (HV), plasma cell (PC), or mixed. The ultrasonic features of each UCD subtype were systematically analyzed. Results A total of 41 patients were enrolled, including 29 patients with HV-type, 4 with PC-type and 8 with mixed type. All patients exhibited enlarged LNs characterized by solitary mass, well-defined margins and increased cortical thickness. Among them, 95.12% (39/41) of cases had an indistinct corticomedullary interface. 41.46% (17/41) showed eccentric or asymmetrical cortical thickening, and 58.54% (24/41) demonstrated complete effacement of the fatty hilum. About 24.39% (10/41) of cases exhibited macrocalcification, and 56.10% (23/41) displayed short linear hyperechoic foci within the LNs. Additionally, HV-type and mixed-type patients had more abundant blood flow signals than PC-type patients (75.86% vs 25% vs 87.50%, P = 0.018). Conclusions Ultrasound features of UCD typically include large, solitary masses with well-defined margins, a thickened cortex, and loss of the fatty hilum. Key imaging findings are micro-calcifications and short linear hyper echoes. Ultrasound is an effective and non-invasive tool for the early detection and diagnosis of UCD. Trial registration retrospectively registered.

  • Open Access

    Determining the Accuracy and Interobserver Agreement of 4 Ultrasound Scores in Crohn's Disease Assessment: Correlations With Endoscopy

    by Jing Qin, Li Ma, Meng-Yuan Zhou, Wen-Bo Li, Meng-Su Xiao, Zi-Han Niu, Hong Yang, Qing-Li Zhu

    INTRODUCTION: Gastrointestinal ultrasound (GIUS) is recommended for monitoring Crohn's disease (CD). GIUS scores are used to quantify CD activity. Among them, International Bowel Ultrasound Segmental Activity Score (IBUS-SAS), Bowel Ultrasound Score (BUSS), Simple Ultrasound Score, and Simple Ultrasound Score for Crohn's Disease are most commonly used. The aim of this study was to compare and correlate the performance of such indicators with endoscopic activity and to calculate interobserver agreement. METHODS: Consecutive patients with CD at our hospital between June 2015 and July 2021 were retrospectively enrolled. All patients underwent ileocolonoscopy after medical treatment. GIUS was performed within 2 weeks, and 4 GIUS scores were independently calculated. Receiver operating characteristic curve analyses were used to determine a cutoff value. Cohen kappa (κ) coefficient was calculated to estimate the agreement between GIUS findings. RESULTS: A total of 106 patients with CD were enrolled. 80.2% (85/106) were endoscopic active (Simple Endoscopic Score for Crohn's disease ≥3), and 8.49% (9/106) were severe cases (Simple Endoscopic Score for Crohn's disease ≥9). All GIUS features (bowel wall thickness, color Doppler signs, bowel wall stratification, inflammatory signals at the mesentery) were statistically significant in assessing CD activity (P < 0.05). IBUS-SAS showed the highest area under the curve (0.98; 95% CI: 0.96–1.00) and specificity (95.2%) for a cutoff value of 46.50. However, IBUS-SAS had only moderate agreement (Cohen κ = 0.427; P < 0.001). BUSS had substantial interobserver agreement (Cohen κ = 0.947; P < 0.001), with a similar diagnostic value (sensitivity, 100.0%; accuracy, 95.3%; area under the curve of 0.96 [95% CI: 0.91–1.00] for a cutoff value of 4.58). DISCUSSION: GIUS score is an efficient and reliable method to assess CD activity. BUSS achieved a high accuracy and excellent interobserver agreement, which is more suitable for treatment assessment.


  • Open Access

    Ultrasound Characteristics Can Predict Response to Biologics Therapy in Stricturing Crohn's Disease

    by Li Ma, Yudi He, Wenbo Li, Mengsu Xiao, Mengyuan Zhou, Jing Qin, Hong Yang, Wei Liu, Qingli Zhu

    INTRODUCTION: Stricture is a common complication in Crohn's disease (CD). Accurate identification of strictures that poorly respond to biologic therapy is essential for making optimal therapeutic decisions. The aim of this study was to determine the association between ultrasound characteristics of strictures and their therapeutic outcomes. METHODS: Consecutive CD patients with symptomatic strictures scheduled for biologic therapy were retrospectively recruited at a tertiary hospital. Baseline intestinal ultrasound was conducted to assess stricture characteristics, including bowel wall thickness, length, stratification, vascularity, and creeping fat wrapping angle. Patients were followed up for a minimum of 1 year, during which long-term outcomes including surgery, steroid-free clinical remission, and mucosal healing were recorded. Statistical analyses were performed. RESULTS: A total of 43 patients were enrolled. Strictures were located in the ileocecal region (39.5%), colon (37.2%), anastomosis (20.9%), and small intestine (2.3%). The median follow-up time was 17 months (interquartile range 7–25), with 27 patients (62.8%) undergoing surgery. On multivariant analysis, creeping fat wrapping angle > 180° (odds ratio: 6.2, 95% confidence interval [CI]: 1.1–41.1) and a high Limberg score (odds ratio: 2.3, 95% CI: 1.4–6.0) were independent predictors of surgery, with an area under the curve of 0.771 (95% CI: 0.602–0.940), accuracy of 83.7%, sensitivity of 96.3%, and specificity of 62.5%. On Cox survival analysis, creeping fat >180° was significantly associated with surgery (hazard ratio, 5.2; 95% CI: 1.2–21.8; P = 0.03). In addition, creeping fat was significantly associated with steroid-free clinical remission (P = 0.015) and mucosal healing (P = 0.06). DISCUSSION: Intestinal ultrasound characteristics can predict outcomes in patients with stricturing CD who undertook biologic therapy.


  • Open Access

    The added value of ultrasound imaging biomarkers to clinicopathological factors for the prediction of high-risk Oncotype DX recurrence scores in patients with breast cancer

    by Yanwen Luo, Yuanjing Gao, Zihan Niu, Jing Zhang, Zhenzhen Liu, Yanna Zhang, Songjie Shen, Yuxin Jiang, Mengsu Xiao, Qingli Zhu

    Background: The Oncotype DX (ODX) recurrence score (RS), a 21-gene assay, has been proven to recognize patients at high risk of recurrence (RS ≥26) who would benefit from chemotherapy. However, it has limited availability and high costs. Our study thus aimed to identify ultrasound (US) imaging biomarkers and develop a prediction model for identifying patients with a high ODX RS. Methods: In this retrospective study, consecutive patients with T1–3N0–1M0 breast cancer who were hormone receptor positive and human epidermal growth factor receptor 2 (HER2) negative who had an available ODX RS were reviewed. Patients treated from May 2012 and December 2015 were placed into a training cohort, and those treated from January 2016 to January 2017 were placed in a validation cohort. Clinicopathologic data were collected, and preoperative US scans were analyzed. Univariable and multivariable regression analyses were performed to evaluate the independent predictors for a high-risk of breast cancer in the training cohort, and a nomogram was developed and evaluated with the area under the receiver operating characteristic curve (AUC), calibration curve, and decision curve analysis (DCA). Results: A total of 363 patients were in the training cohort and 160 in the validation cohort, with the proportion with a high RS (RS 26–100) being 14% and 13.1%, respectively. Echogenic halo, enhanced posterior echo, low level of progesterone receptor (PR), and high Ki-67 index were identified as independent risk factors for high RS (all P values Conclusions: US features may serve as valuable imaging biomarkers for the prediction of high recurrence risk in patients with T1–3N0–1M0 breast cancer and hormone receptor (HR)-positive and HER2-negative status. A nomogram incorporating PR status, Ki-67 index, and US imaging biomarkers showed a good discrimination ability in the early selection of patients at high risk of recurrence, especially in those with stage T1N0M0 disease.


  • Open Access

    Predicting the Efficacy of Neoadjuvant Chemotherapy for Pancreatic Cancer Using Deep Learning of Contrast-Enhanced Ultrasound Videos

    by Yuming Shao, Yingnan Dang, Yuejuan Cheng, Yang Gui, Xueqi Chen, Tianjiao Chen, Yan Zeng, Li Tan, Jing Zhang, Mengsu Xiao, et al.

    Contrast-enhanced ultrasound (CEUS) is a promising imaging modality in predicting the efficacy of neoadjuvant chemotherapy for pancreatic cancer, a tumor with high mortality. In this study, we proposed a deep-learning-based strategy for analyzing CEUS videos to predict the prognosis of pancreatic cancer neoadjuvant chemotherapy. Pre-trained convolutional neural network (CNN) models were used for binary classification of the chemotherapy as effective or ineffective, with CEUS videos collected before chemotherapy as the model input, and with the efficacy after chemotherapy as the reference standard. We proposed two deep learning models. The first CNN model used videos of ultrasound (US) and CEUS (US+CEUS), while the second CNN model only used videos of selected regions of interest (ROIs) within CEUS (CEUS-ROI). A total of 38 patients with strict restriction of clinical factors were enrolled, with 76 original CEUS videos collected. After data augmentation, 760 and 720 videos were included for the two CNN models, respectively. Seventy-six-fold and 72-fold cross-validations were performed to validate the classification performance of the two CNN models. The areas under the curve were 0.892 and 0.908 for the two models. The accuracy, recall, precision and F1 score were 0.829, 0.759, 0.786, and 0.772 for the first model. Those were 0.864, 0.930, 0.866, and 0.897 for the second model. A total of 38.2% and 40.3% of the original videos could be clearly distinguished by the deep learning models when the naked eye made an inaccurate classification. This study is the first to demonstrate the feasibility and potential of deep learning models based on pre-chemotherapy CEUS videos in predicting the efficacy of neoadjuvant chemotherapy for pancreas cancer.


  • Open Access

    Clinical, dermoscopic, and ultrasonic monitoring of the response to biologic treatment in patients with moderate-to-severe plaque psoriasis

    by Juncheng Wang, Qingli Zhu, Feng Li, Mengsu Xiao, Jie Liu

    Background: Assessment of therapeutic response of psoriasis has relied traditionally on clinical observation, and effective non-invasive tools are desirable.Objectives: To investigate the value of dermoscopy and high-frequency ultrasound (HFUS) in the monitoring of psoriatic lesions treated with biologics.Methods: Patients with moderate-to-severe plaque psoriasis treated with biologics were evaluated by clinical, dermoscopic, and ultrasonic scores at weeks 0, 4, 8, and 12. Clinical scores, including Psoriasis Area Severity Index (PASI) and target lesion score (TLS), were evaluated at representative lesions. Dermoscopy was performed to assess the red background, vessels, and scales on a 4-point scale as well as the presence of hyperpigmentation, hemorrhagic spots, and linear vessels. HFUS was performed to measure the thicknesses of the superficial hyperechoic band and subepidermal hypoechoic band (SLEB). The correlation between clinical, dermoscopic, and ultrasonic evaluation was also analyzed.Results: In total, 24 patients were analyzed and achieved 85.3 and 87.5% reduction of PASI and TLS, respectively, after 12 weeks of treatment. The red background, vessels, and scales scores under dermoscopy were reduced by 78.5, 84.1, and 86.5%, respectively. Some patients developed hyperpigmentation and linear vessels after treatment. Hemorrhagic dots slowly subside over the therapeutic course. Ultrasonic scores were significantly improved with an average reduction of 53.9% in superficial hyperechoic band thickness and 89.9% in SLEB thickness. TLS in the clinical variables, scales in dermoscopic variables, and SLEB in ultrasonic variables decreased the most significantly in the early stage of treatment (week 4) with 55.4, 57.7, and 59.1% (P > 0.05), respectively. Most of the variables, including the red background, vessels, scales, and SLEB thickness, were strongly correlated with TLS. High correlations were also found between the SLEB thickness and the red background or vessels scores, and between the superficial hyperechoic band thickness and the scales scores.Conclusion: Both dermoscopy and HFUS were useful in the therapeutic monitoring of moderate-to-severe plaque psoriasis.


  • Open Access

    Deep learning-based system for automatic prediction of triple-negative breast cancer from ultrasound images

    by Alexandre Boulenger, Yanwen Luo, Chenhui Zhang, Chenyang Zhao, Yuanjing Gao, Mengsu Xiao, Qingli Zhu, Jie Tang

    To develop a deep-learning system for the automatic identification of triple-negative breast cancer (TNBC) solely from ultrasound images. A total of 145 patients and 831 images were retrospectively enrolled at Peking Union College Hospital from April 2018 to March 2019. Ultrasound images and clinical information were collected accordingly. Molecular subtypes were determined from immunohistochemical (IHC) results. A CNN with VGG-based architecture was then used to predict TNBC. The model’s performance was evaluated using randomized k-fold stratified cross-validation. A t-SNE analysis and saliency maps were used for model visualization. TNBC was identified in 16 of 145 (11.03%) patients. One hundred fifteen (80%) patients, 15 (10%) patients, and 15 (10%) patients formed the train, validation, and test set respectively. The deep learning system exhibits good efficacy, with an AUC of 0.86 (95% CI: 0.64, 0.95), an accuracy of 85%, a sensitivity of 86%, a specificity of 86%, and an F1-score of 0.74. In addition, the internal representation features learned by the model showed clear differentiation across molecular subtype groups. Such a deep learning system can automatically predict triple-negative breast cancer preoperatively and accurately. It may help to get to more precise and comprehensive management. Graphical Abstract


  • Open Access

    Simultaneous Detection of CNVs and SNVs Improves the Diagnostic Yield of Fetuses with Ultrasound Anomalies and Normal Karyotypes

    by Qingwei Qi, Yulin Jiang, Xiya Zhou, Hua Meng, Na Hao, Jiazhen Chang, Junjie Bai, Chunli Wang, Mingming Wang, Jiangshan Guo, et al.

    The routine assessment to determine the genetic etiology for fetal ultrasound anomalies follows a sequential approach, which usually takes about 6–8 weeks turnaround time (TAT). We evaluated the clinical utility of simultaneous detection of copy number variations (CNVs) and single nucleotide variants (SNVs)/small insertion-deletions (indels) in fetuses with a normal karyotype with ultrasound anomalies. We performed CNV detection by chromosomal microarray analysis (CMA) or low pass CNV-sequencing (CNV-seq), and in parallel SNVs/indels detection by trio-based clinical exome sequencing (CES) or whole exome sequencing (WES). Eight-three singleton pregnancies with a normal fetal karyotype were enrolled in this prospective observational study. Pathogenic or likely pathogenic variations were identified in 30 cases (CNVs in 3 cases, SNVs/indels in 27 cases), indicating an overall molecular diagnostic rate of 36.1% (30/83). Two cases had both a CNV of uncertain significance (VOUS) and likely pathogenic SNV, and one case carried both a VOUS CNV and an SNV. We demonstrated that simultaneous analysis of CNVs and SNVs/indels can improve the diagnostic yield of prenatal diagnosis with shortened reporting time, namely, 2–3 weeks. Due to the relatively long TAT for sequential procedure for prenatal genetic diagnosis, as well as recent sequencing technology advancements, it is clinically necessary to consider the simultaneous evaluation of CNVs and SNVs/indels to enhance the diagnostic yield and timely TAT, especially for cases in the late second trimester or third trimester.


  • Open Access

    Preliminary study of contrast-enhanced ultrasound in combination with blue dye vs. indocyanine green fluorescence, in combination with blue dye for sentinel lymph node biopsy in breast cancer

    by Yidong Zhou, Yan Li, Feng Mao, Jing Zhang, Qingli Zhu, Songjie Shen, Yan Lin, Xiaohui Zhang, He Liu, Mengsu Xiao, et al.

    Background: This preliminary study aimed to examine the feasibility of sentinel lymph node biopsy (SLNB) using contrast-enhanced ultrasound (CEUS) vs. indocyanine green fluorescence (ICG), combined with blue dye in patients with breast cancer. Methods: This was a retrospective study of consecutive female patients with invasive stage I-III (based on pre-operative physical examination and imaging) primary breast cancer at the Peking Union Medical College Hospital between 01/2013 and 01/2015 who underwent preoperative SLNB by ICG + blue dye or CEUS + blue dye. The numbers of detected SLNs, detection rates, and recurrence-free survival (RFS) rates were compared between the two groups. Results: A total of 443 patients were included. The detection rates of SLNs in the CEUS + blue dye and ICG + blue dye groups were 98.4 and 98.1%, respectively (P = 0.814). The average numbers of SLNs detected per patient showed no significant difference between the two groups (3.06 ± 1.33 and 3.12 ± 1.31 in the CEUS + blue dye and ICG + blue dye groups, respectively; P = 0.659). After a median follow-up of 46 months, five patients in the CEUS + blue dye group and 15 in the ICG + blue dye group had recurrence. RFS rates showed no significant difference (P = 0.55). Conclusion: This preliminary study suggests that CEUS + blue dye and ICG + blue dye are both feasible for SLN detection in breast cancer.