Kai Yang

Dr. Kai Yang

radiology, Shanghai Sixth People's Hospital, Shanghai, China
orcid

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

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

Representative Publications

View All Publications
15 March 2023 in Journal of Sport and Health Science
Verified
Journal Article
Open access
JZJieyuan Zhang
Jieyuan Zhang
Kai YangKai Yang
Kai Yang
CWCheng Wang
Cheng Wang
WGWenqi Gu
Wenqi Gu
XLXueqian Li
Xueqian Li
ZSZhongmin Shi
Zhongmin Shi
Chronic ankle instability (CAI) is a common sequela following an acute lateral ankle sprain (LAS). To treat an acute LAS more effectively and efficiently, it is important to identify patients at substantial risk for developing CAI. This study identifies magnetic resonance imaging (MRI) manifestations for predicting CAI development after a first episode of LAS and explores appropriate clinical indications for ordering MRI scans for these patients. All patients with a first-episode LAS who received plain radiograph and MRI scanning within the first 2 weeks after LAS from December 1, 2017 to December 1, 2019 were identified. Data were collected using the Cumberland Ankle Instability Tool at final follow-up. Demographic and other related clinical variables, including age, sex, body mass index, and treatment were also recorded. Univariable and multivariable analyses were performed successively to identify risk factors for CAI after first-episode LAS. A total 131 out of 362 patients with a mean follow-up of 3.0 ± 0.6 years (mean ± SD; 2.0-4.1 years) developed CAI after first-episode LAS. According to multivariable regression, development of CAI after first-episode LAS was associated with 5 prognostic factors: age (odds ratio (OR) = 0.96, 95% confidence interval (95%CI): 0.93-1.00, p = 0.032); body mass index (OR = 1.09, 95%CI: 1.02-1.17, p = 0.009); posterior talofibular ligament injury (OR = 2.17, 95%CI: 1.05-4.48, p = 0.035); large bone marrow lesion of the talus (OR = 2.69, 95%CI: 1.30-5.58, p = 0.008), and Grade 2 effusion of the tibiotalar joint (OR = 2.61, 95%CI: 1.39-4.89, p = 0.003). When patients had at least 1 positive clinical finding in the 10-m walk test, anterior drawer test, or inversion tilt test, they had a 90.2% sensitivity and 77.4% specificity in terms of detecting at least 1 prognostic factor by MRI. MRI scanning is valuable in predicting CAI after first-episode LAS for those patients with at least 1 positive clinical finding in the 10-m walk test, anterior drawer test, and inversion tilt test. Further prospective and large-scale studies are necessary for validation.

15 January 2024 in Frontiers in Bioengineering and Biotechnology
Verified
Journal Article
Open access
MYMengjie Ying
Mengjie Ying
YWYufan Wang
Yufan Wang
Kai YangKai Yang
Kai Yang
HWHaoyuan Wang
Haoyuan Wang
XLXudong Liu
Xudong Liu
Purpose: To construct a deep learning knowledge distillation framework exploring the utilization of MRI alone or combing with distilled Arthroscopy information for meniscus tear detection.Methods: A database of 199 paired knee Arthroscopy-MRI exams was used to develop a multimodal teacher network and an MRI-based student network, which used residual neural networks architectures. A knowledge distillation framework comprising the multimodal teacher network T and the monomodal student network S was proposed. We optimized the loss functions of mean squared error (MSE) and cross-entropy (CE) to enable the student network S to learn arthroscopic information from the teacher network T through our deep learning knowledge distillation framework, ultimately resulting in a distilled student network ST. A coronal proton density (PD)-weighted fat-suppressed MRI sequence was used in this study. Fivefold cross-validation was employed, and the accuracy, sensitivity, specificity, F1-score, receiver operating characteristic (ROC) curves and area under the receiver operating characteristic curve (AUC) were used to evaluate the medial and lateral meniscal tears detection performance of the models, including the undistilled student model S, the distilled student model ST and the teacher model T.Results: The AUCs of the undistilled student model S, the distilled student model ST, the teacher model T for medial meniscus (MM) tear detection and lateral meniscus (LM) tear detection are 0.773/0.672, 0.792/0.751 and 0.834/0.746, respectively. The distilled student model ST had higher AUCs than the undistilled model S. After undergoing knowledge distillation processing, the distilled student model demonstrated promising results, with accuracy (0.764/0.734), sensitivity (0.838/0.661), and F1-score (0.680/0.754) for both medial and lateral tear detection better than the undistilled one with accuracy (0.734/0.648), sensitivity (0.733/0.607), and F1-score (0.620/0.673).Conclusion: Through the knowledge distillation framework, the student model S based on MRI benefited from the multimodal teacher model T and achieved an improved meniscus tear detection performance.

14 October 2022 in Orthopaedic Journal of Sports Medicine
Verified
Journal Article
Open access
SFShaoling Fu
Shaoling Fu
Kai YangKai Yang
Kai Yang
XLXueqian Li
Xueqian Li
CCCheng Chen
Cheng Chen
GMGuohua Mei
Guohua Mei
ZSZhongmin Shi
Zhongmin Shi
Background: Arthroscopic microfracture for osteochondral lesion of the talus (OLT) has shown good functional outcomes in the short and long term. Purpose: To investigate 5-year radiographic and clinical outcomes after arthroscopic microfracture in treatment of OLT and the effectiveness of adjunct therapies including platelet-rich plasma (PRP) and hyaluronic acid (HA). Study Design: Cohort study; Level of evidence, 2. Methods: We prospectively enrolled 432 patients who underwent arthroscopic microfracture for OLT from May 1, 2011, to May 31, 2015. Magnetic resonance imaging (MRI) and weightbearing radiographs were performed annually after the initial surgery. The MOCART (magnetic resonance observation of cartilage repair tissue) score was used to evaluate the structure of the repaired cartilage on MRI, and patient-reported outcomes (American Orthopaedic Foot and Ankle Society ankle-hindfoot scale [AOFAS] and the Foot and Ankle Outcome Score) were collected annually. The primary outcome measure was 5-year AOFAS score. We recorded baseline characteristics including age, body mass index (BMI), and lesion size, and other potentially related factors including number of PRP/HA injection and change in BMI from baseline. Results: Included were 355 patients, all with minimum 5-year follow-up data. The overall reoperation rate was 9.0% (32 of 355). According to multivariable analysis, 5-year AOFAS scores were associated with number of PRP injections (correlation coefficient, 3.12 [95% CI, 2.36 to 3.89]; P < .001), BMI at baseline (correlation coefficient, -0.222 [95% CI, -0.363 to -0.082]; P = .002), and mean BMI change from baseline (correlation coefficient, -1.15 [95% CI, -1.32 to -0.98]; P < .001). When comparing number of PRP injections (0, 1-2, or ≥3), we found that patients who had serial PRP injection (≥3 with at least a 3-month interval between injections) had diminished functional and radiographic deterioration over time. Conclusion: Arthroscopic microfracture improved patient-reported and structural outcomes for patients with OLT at 5 years after surgery. Serial PRP injections and reduction in BMI from baseline were able to slow radiographic and functional deterioration. Future trials regarding the combination of microfracture and PRP in treatment of OLT should focus on the efficacy of longer term, intra-articular, serial injections of PRP instead of single injections.

30 September 2022 in Frontiers in Bioengineering and Biotechnology
Verified
Journal Article
Open access
CQCheng Qu
Cheng Qu
HYHeng Yang
Heng Yang
CWCong Wang
Cong Wang
CWChongyang Wang
Chongyang Wang
MYMengjie Ying
Mengjie Ying
XLXudong Liu
Xudong Liu
Purpose: To develop and evaluate a deep learning-based method to localize and classify anterior cruciate ligament (ACL) ruptures on knee MR images by using arthroscopy as the reference standard.Methods: We proposed a fully automated ACL rupture localization system to localize and classify ACL ruptures. The classification of ACL ruptures was based on the projection coordinates of the ACL rupture point on the line connecting the center coordinates of the femoral and tibial footprints. The line was divided into three equal parts and the position of the projection coordinates indicated the classification of the ACL ruptures (femoral side, middle and tibial side). In total, 85 patients (mean age: 27; male: 56) who underwent ACL reconstruction surgery under arthroscopy were included. Three clinical readers evaluated the datasets separately and their diagnostic performances were compared with those of the model. The performance metrics included the accuracy, error rate, sensitivity, specificity, precision, and F1-score. A one-way ANOVA was used to evaluate the performance of the convolutional neural networks (CNNs) and clinical readers. Intraclass correlation coefficients (ICC) were used to assess interobserver agreement between the clinical readers.Results: The accuracy of ACL localization was 3.77 ± 2.74 and 4.68 ± 3.92 (mm) for three-dimensional (3D) and two-dimensional (2D) CNNs, respectively. There was no significant difference in the ACL rupture location performance between the 3D and 2D CNNs or among the clinical readers (Accuracy, p < 0.01). The 3D CNNs performed best among the five evaluators in classifying the femoral side (sensitivity of 0.86 and specificity of 0.79), middle side (sensitivity of 0.71 and specificity of 0.84) and tibial side ACL rupture (sensitivity of 0.71 and specificity of 0.99), and the overall accuracy for sides classifying of ACL rupture achieved 0.79.Conclusion: The proposed deep learning-based model achieved high diagnostic performances in locating and classifying ACL fractures on knee MR images.

1 March 2023 in Journal of Orthopaedic Surgery and Research
Verified
Journal Article
Open access
PDF file
JZJieyuan Zhang
Jieyuan Zhang
XLXueqian Li
Xueqian Li
SFShaoling Fu
Shaoling Fu
Kai YangKai Yang
Kai Yang
ZSZhongmin Shi
Zhongmin Shi
Purpose: The aim of this study was to evaluate the relation between anteromedial ankle osteophytes (AMAO) and anteromedial ankle impingement (AMAI) in chronic lateral ankle instability (CLAI) through visualization and quantification.Methods: Forty-three patients with unilateral CLAI between September 2018 and March 2020 accepted arthroscopic repair of an anterior talofibular ligament (ATFL) and were split into two groups: AMAI (AMAI including intraoperative AMAO resection) and pure CLAI (with AMAO but without AMAI, no AMAO resection). The AMAO protrusion lengths in each direction were measured and compared after all of the ankles were reconstructed. All patients were assessed preoperatively and at 2-year follow-up with ankle dorsiflexion, the American Orthopedic Foot and Ankle Society (AOFAS) ankle–hindfoot score, and visual analog scale (VAS) score.Results: Intelligent analysis showed that a large extent of osteophytes was found at the dorsomedial surface of the talar neck in AMAI group. The upper and inner bound protrusion distances of AMAO in AMAI group were greater than in the pure CLAI group. There was no significant difference in anterior bound protrusion distance of AMAO between the two groups. Preoperatively, the ankle dorsiflexion of AMAI group (7.6 ± 1.4°) was considerably lower than that of pure CLAI group (22.4 ± 1.9°) (p < 0.001). When compared to the pure CLAI group, the AMAI group had a substantially worse AOFAS score (62.2 ± 6.7 vs 71.1 ± 9.1;p < 0.001) and VAS score (6.0 ± 1.0 vs 4.9 ± 0.8;p < 0.05). However, there was no significant difference in postoperative ankle dorsiflexion, AOFAS score, or VAS score between the two groups.Conclusion: AMAO is formed mostly on the dorsomedial surface of the talar neck in CLAI with AMAI, and the upper and inner bound protrusion lengths of AMAO were shown to be significantly correlated with the existence of AMAI in CLAI.Level of evidence IV.

19 September 2025 in Bioengineering
Verified
Journal Article
Open access
PDF file
LZLi Zhengliang
Li Zhengliang
JXJie Xia
Jie Xia
CWCong Wang
Cong Wang
ZZZhemin Zhu
Zhemin Zhu
FZFan Zhang
Fan Zhang
Kai YangKai Yang
Kai Yang
Full-length radiography is essential for evaluating spinal deformities, limb length discrepancies, and preoperative planning in orthopedics, yet the measurement accuracy of different radiographic methods remains unclear. This phantom study compared the accuracy of rotational stitching, one-shot and slot-scanning full-length radiography across six radiographic systems in quantifying distances between anatomical landmarks. Measurement errors were statistically analyzed using appropriate nonparametric tests. The results demonstrated significant differences in measurement accuracy among the three methods (H (2) = 15.86, p < 0.001). Slot-scanning exhibited the highest accuracy, with a mean error of −1.19 ± 10.13 mm, while both rotational stitching and one-shot imaging showed greater systematic underestimation, with mean errors of −18.95 ± 13.77 mm and −15.32 ± 12.38 mm, respectively. These negative biases (approximately 1.9 cm and 1.5 cm) are clinically meaningful because, if unrecognized, they can alter mechanical axis estimation and alignment planning in procedures such as high tibial osteotomy (HTO). Post hoc analysis confirmed the superior accuracy of slot-scanning compared to the other two methods, while no significant difference was found between rotational stitching and one-shot imaging. These findings indicate that system choice substantially impacts measurement accuracy, supporting preferential use of slot-scanning when precise quantitative assessment is required.