Yi Xu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Yi Xu has a total of 24 co-authors. Here are the top co-authors by number of shared publications:
- Cunyi Fan
Department of Orthopedics, Shanghai Sixth People’s Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, People’s Republic of China
5 shared publications - Ziyang Sun
Department of Orthopedics, Shanghai Sixth People’s Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, PR China
4 shared publications - Hang Liu
Department of Orthopedics, Shanghai Sixth People’s Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, PR China
3 shared publications - Hao Xiong
Department of Orthopaedics, Shanghai Sixth People’s Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
3 shared publications - Juehong Li
Department of Orthopedics, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, 200233, PR China
3 shared publications - Gang Luo
Department of Orthopedics, Shanghai Sixth People’s Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, PR China
2 shared publications
Latest Publications
- 22 October 2025
- Published by British Editorial Society of Bone & Joint Surgery in Orthopaedic Proceedings
- Vol. 107-B (SUPP_9) , 50
DEVELOPMENT AND VALIDATION OF A DEEP-LEARNING MODEL TO PREDICT TOTAL HIP REPLACEMENT ON RADIOGRAPHS: THE TOTAL HIP REPLACEMENT PREDICTION (THREP) MODEL
The methods for accurately assessing risk of total hip replacement (THR) in patients with osteoarthritis are scarce. A novel and reliable method should be investigated to play a significant role in research and clinical routine. Purpose: To develop a deep learning model that can reliably predict the risk of THR based on radiographic images and clinical symptoms. This retrospective, multi-center, case-control study assessed hip joints on weighted-bearing anterior-posterior pelvic radiographs from the Osteoarthritis Initiative (OAI) participants. Participants received THR were matched to controls according to age, sex, ethnicity, body mass index and ethnicity. Cases and controls were uniformly split at 72% ( n = 528), 14% ( n = 104), 14% ( n = 104) into training, validation, and testing data sets. Images and clinical symptoms were passed through a detection model and a deep convolutional neural network (DCNN) model to predict the probability of THR within 9 years and the most likely time period of THR (0-3 years, 3–6 years, 6–9 years). Model performance was assessed using area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity in the testing set. A total of 736 participants were evaluated, including 184 cases and 552 controls. The prediction model achieved an overall accuracy, sensitivity, and specificity of 91.35%, 92.59% and 86.96%, with an AUC of 0.944 within 9 years. Meanwhile, the AUC of the DCNN model for assessing the most likely time period was 0.907 for 0-3 years, 0.916 for 3–6 years, and 0.841 for 6–9 years. Gradient-weighted class activation mapping closely corresponded to regions affecting the prediction of the DCNN model. Conclusion: The proposed DCNN model is a reliable and valid method to predict the probability of THR within limitations. It could assist clinicians in patient counselling and decision-making about the timing of the intervention. In the future, by increasing the size of the dataset, enhancing the ethnic and socioeconomic diversity of the participants, and improving the follow-up rate, the quality of the conclusions can be further improved.
- 22 October 2025
- Published by British Editorial Society of Bone & Joint Surgery in Orthopaedic Proceedings
- Vol. 107-B (SUPP_9) , 50
DEVELOPMENT AND VALIDATION OF A DEEP-LEARNING MODEL TO PREDICT TOTAL HIP REPLACEMENT ON RADIOGRAPHS: THE TOTAL HIP REPLACEMENT PREDICTION (THREP) MODEL
The methods for accurately assessing risk of total hip replacement (THR) in patients with osteoarthritis are scarce. A novel and reliable method should be investigated to play a significant role in research and clinical routine. Purpose: To develop a deep learning model that can reliably predict the risk of THR based on radiographic images and clinical symptoms. This retrospective, multi-center, case-control study assessed hip joints on weighted-bearing anterior-posterior pelvic radiographs from the Osteoarthritis Initiative (OAI) participants. Participants received THR were matched to controls according to age, sex, ethnicity, body mass index and ethnicity. Cases and controls were uniformly split at 72% ( n = 528), 14% ( n = 104), 14% ( n = 104) into training, validation, and testing data sets. Images and clinical symptoms were passed through a detection model and a deep convolutional neural network (DCNN) model to predict the probability of THR within 9 years and the most likely time period of THR (0-3 years, 3–6 years, 6–9 years). Model performance was assessed using area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity in the testing set. A total of 736 participants were evaluated, including 184 cases and 552 controls. The prediction model achieved an overall accuracy, sensitivity, and specificity of 91.35%, 92.59% and 86.96%, with an AUC of 0.944 within 9 years. Meanwhile, the AUC of the DCNN model for assessing the most likely time period was 0.907 for 0-3 years, 0.916 for 3–6 years, and 0.841 for 6–9 years. Gradient-weighted class activation mapping closely corresponded to regions affecting the prediction of the DCNN model. Conclusion: The proposed DCNN model is a reliable and valid method to predict the probability of THR within limitations. It could assist clinicians in patient counselling and decision-making about the timing of the intervention. In the future, by increasing the size of the dataset, enhancing the ethnic and socioeconomic diversity of the participants, and improving the follow-up rate, the quality of the conclusions can be further improved.
- Open Access
- 20 October 2025
- Published by Springer Nature in Bone Research
- Vol. 13 (1) , 1-19
Pak4-mediated crosstalk between necroptotic macrophages and tendon stem/progenitor cells contributes to traumatic heterotopic ossification formation
The formation of traumatic heterotopic ossification (HO) is an abnormal repair process after soft tissue injury. Recent studies establish the involvement of immune cells and cellular metabolism in the tissue healing process; however, their role in HO remains unknown. Here, by using murine burn/tenotomy model in vivo and tendon stem/progenitor cells (TSPCs) osteogenic differentiation model in vitro, together with techniques including transgenic knockout, gene knockdown, transcriptome and proteome sequencings, mass spectrometry, co-immunoprecipitation, seahorse, etc., we reveal a novel p21-activated kinase 4 (Pak4) mediated crosstalk where the necroptotic macrophages arouse TSPCs with reduced fatty acid β-oxidation (FAO), to promote aberrant osteogenic differentiation during HO formation. Necroptosis blockade with Mlkl knockout (C57BL/6JGpt-Mlkl em1Cd1679 /Gpt) significantly reduces HO than WT mice. Extracellular vesicle (EVs) secreted from necroptotic bone marrow-derived macrophages (BMDMs, NecroMφ-EVs) are determined to motivate FAO reduction in TSPCs and result in higher osteogenic activity. Pak4 conditional knockout (C57BL/6JGpt-Pak4 em1Cflox /Gpt) in macrophage significantly increases FAO and reduces HO than Flox mice, as well as local injection of PAK4 −/− -EVs (NecroMφ-EVs with Pak4 knockout) than NecroMφ-EVs, and the protective effects are reversed after transfection of Fabp3 S122D , a phosphomimetic mutant of S122 on fatty acid binding protein 3 (Fabp3) phosphorylation site. Mechanically, after soft tissue injury, macrophages infiltrate, and necroptosis occurs, accompanied by paracrine EVs-derived Pak4, which binds directly to Fabp3 and phosphorylates it at the S122 site in TSPCs, results in reduced FAO, finally osteogenic behavior, and HO formation. This study adds perceptiveness into abnormal regeneration-based theory for traumatic HO and raises treatment strategy development.
- 20 October 2025
- Published by Springer Nature in Bone Research
- Vol. 13 (1) , 1-19
Open AccessPak4-mediated crosstalk between necroptotic macrophages and tendon stem/progenitor cells contributes to traumatic heterotopic ossification formation
The formation of traumatic heterotopic ossification (HO) is an abnormal repair process after soft tissue injury. Recent studies establish the involvement of immune cells and cellular metabolism in the tissue healing process; however, their role in HO remains unknown. Here, by using murine burn/tenotomy model in vivo and tendon stem/progenitor cells (TSPCs) osteogenic differentiation model in vitro, together with techniques including transgenic knockout, gene knockdown, transcriptome and proteome sequencings, mass spectrometry, co-immunoprecipitation, seahorse, etc., we reveal a novel p21-activated kinase 4 (Pak4) mediated crosstalk where the necroptotic macrophages arouse TSPCs with reduced fatty acid β-oxidation (FAO), to promote aberrant osteogenic differentiation during HO formation. Necroptosis blockade with Mlkl knockout (C57BL/6JGpt-Mlkl em1Cd1679 /Gpt) significantly reduces HO than WT mice. Extracellular vesicle (EVs) secreted from necroptotic bone marrow-derived macrophages (BMDMs, NecroMφ-EVs) are determined to motivate FAO reduction in TSPCs and result in higher osteogenic activity. Pak4 conditional knockout (C57BL/6JGpt-Pak4 em1Cflox /Gpt) in macrophage significantly increases FAO and reduces HO than Flox mice, as well as local injection of PAK4 −/− -EVs (NecroMφ-EVs with Pak4 knockout) than NecroMφ-EVs, and the protective effects are reversed after transfection of Fabp3 S122D , a phosphomimetic mutant of S122 on fatty acid binding protein 3 (Fabp3) phosphorylation site. Mechanically, after soft tissue injury, macrophages infiltrate, and necroptosis occurs, accompanied by paracrine EVs-derived Pak4, which binds directly to Fabp3 and phosphorylates it at the S122 site in TSPCs, results in reduced FAO, finally osteogenic behavior, and HO formation. This study adds perceptiveness into abnormal regeneration-based theory for traumatic HO and raises treatment strategy development.
- Open Access
- 4 September 2025
- Published by Elsevier in The Journal of Arthroplasty
- Vol. 41 (4) , 1241-1247.e4
Effect of Oral Direct Factor Xa Inhibitors on the Prevalence of Hip Heterotopic Ossification Following Arthroplasty for Acute Femoral Neck Fracture: A Propensity Score-matched Cohort Study
- 4 September 2025
- Published by Elsevier in The Journal of Arthroplasty
- Vol. 41 (4) , 1241-1247.e4
Open AccessEffect of Oral Direct Factor Xa Inhibitors on the Prevalence of Hip Heterotopic Ossification Following Arthroplasty for Acute Femoral Neck Fracture: A Propensity Score-matched Cohort Study
- 3 June 2025
- Published by Wiley in International Journal of Rheumatic Diseases
- Vol. 28 (6) , e70314
Inflammatory Bowel Disease and Risk of Rheumatoid Arthritis: A UK Biobank Cohort Study
Objectives: Inflammatory bowel disease (IBD) and rheumatoid arthritis (RA) are both prevalent inflammatory conditions within the population. Our objective was to explore the relationship between IBD and RA, while examining the role of inflammatory mediators in the observed association. Methods: We used data from the UK Biobank, a population‐based prospective cohort study that recruited adults aged 37–73 years from 22 centers in England, Scotland, and Wales between 2006 and 2010. We included patients diagnosed with IBD at baseline and excluded those with RA at baseline or missing follow‐up information. Cox regression proportional hazard models were employed to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) between patients with IBD (Ulcerative colitis and Crohn's disease) at baseline and the risk of RA. Additionally, we conducted mediation analysis to examine the roles of C‐reactive protein (CRP) and several composite inflammatory indices as potential mediators. Results: After excluding participants with RA at baseline ( N = 6769), those lacking IBD subtype information ( N = 475), and those with missing covariate data ( N = 121 195), a total of 373 693 individuals were included in the analysis. Compared with individuals without IBD, those with IBD had a significantly higher risk of developing RA (hazard ratio 2.06, 95% confidence interval 1.69–2.51). This association remained robust after adjustment for multiple confounders and across all major subgroups. Notably, the risk of RA associated with IBD was even higher among individuals with a low polygenic risk score for RA. Mediation analysis showed that systemic inflammatory markers, such as CRP, explained only a modest proportion of the association between IBD and RA, with the highest mediation proportion observed being 9.56%. Conclusion: In the UK Biobank cohort study, individuals with IBD demonstrated an increased risk of developing RA. Future research should aim to gain insight into these underlying mechanisms and explore ways to improve long‐term health outcomes in these patients.
- 3 June 2025
- Published by Wiley in International Journal of Rheumatic Diseases
- Vol. 28 (6) , e70314
Inflammatory Bowel Disease and Risk of Rheumatoid Arthritis: A UK Biobank Cohort Study
Objectives: Inflammatory bowel disease (IBD) and rheumatoid arthritis (RA) are both prevalent inflammatory conditions within the population. Our objective was to explore the relationship between IBD and RA, while examining the role of inflammatory mediators in the observed association. Methods: We used data from the UK Biobank, a population‐based prospective cohort study that recruited adults aged 37–73 years from 22 centers in England, Scotland, and Wales between 2006 and 2010. We included patients diagnosed with IBD at baseline and excluded those with RA at baseline or missing follow‐up information. Cox regression proportional hazard models were employed to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) between patients with IBD (Ulcerative colitis and Crohn's disease) at baseline and the risk of RA. Additionally, we conducted mediation analysis to examine the roles of C‐reactive protein (CRP) and several composite inflammatory indices as potential mediators. Results: After excluding participants with RA at baseline ( N = 6769), those lacking IBD subtype information ( N = 475), and those with missing covariate data ( N = 121 195), a total of 373 693 individuals were included in the analysis. Compared with individuals without IBD, those with IBD had a significantly higher risk of developing RA (hazard ratio 2.06, 95% confidence interval 1.69–2.51). This association remained robust after adjustment for multiple confounders and across all major subgroups. Notably, the risk of RA associated with IBD was even higher among individuals with a low polygenic risk score for RA. Mediation analysis showed that systemic inflammatory markers, such as CRP, explained only a modest proportion of the association between IBD and RA, with the highest mediation proportion observed being 9.56%. Conclusion: In the UK Biobank cohort study, individuals with IBD demonstrated an increased risk of developing RA. Future research should aim to gain insight into these underlying mechanisms and explore ways to improve long‐term health outcomes in these patients.
- 12 December 2023
- Published by Wolters Kluwer Health in Journal of Bone and Joint Surgery
- Vol. 106 (5) , 389-396
Development and Validation of a Deep-Learning Model to Predict Total Hip Replacement on Radiographs
Background: There are few methods for accurately assessing the risk of total hip arthroplasty (THA) in patients with osteoarthritis. A novel and reliable method that could play a substantial role in research and clinical routine should be investigated. The purpose of the present study was to develop a deep-learning model that can reliably predict the risk of THA with use of radiographic images and clinical symptom data. Methods: This retrospective, multicenter, case-control study assessed hip joints on weighted-bearing anteroposterior pelvic radiographs obtained from Osteoarthritis Initiative (OAI) participants. Participants who underwent THA were matched to controls according to age, sex, body mass index, and ethnicity. Cases and controls were uniformly split into training, validation, and testing data sets at proportions of 72% (n = 528), 14% (n = 104), and 14% (n = 104), respectively. Images and clinical symptom data were passed through a detection model and a deep convolutional neural network (DCNN) model to predict the probability of THA within 9 years as well as the most likely time period for THA (0 to 2 years, 3 to 5 years, 6 to 9 years). Model performance was assessed with use of the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity in the testing set. Results: A total of 736 participants were evaluated, including 184 cases and 552 controls. The prediction model achieved an overall accuracy, sensitivity, and specificity of 91.35%, 92.59% and 86.96%, respectively, with an AUC of 0.944, for THA within 9 years. The AUC of the DCNN model for assessing the most likely time period was 0.907 for 0 to 2 years, 0.916 for 3 to 5 years, and 0.841 for 6 to 9 years. Gradient-weighted class activation mapping closely corresponded to regions affecting the prediction of the DCNN model. Conclusions: The proposed DCNN model is a reliable and valid method to predict the probability of THA—within limitations. It could assist clinicians in patient counseling and decision-making regarding the timing of the intervention. In the future, by increasing the size of the data set, enhancing the ethnic and socioeconomic diversity of the participants, and improving the follow-up rate, the quality of the conclusions can be further improved. Level of Evidence: Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.
- 12 December 2023
- Published by Wolters Kluwer Health in Journal of Bone and Joint Surgery
- Vol. 106 (5) , 389-396
Development and Validation of a Deep-Learning Model to Predict Total Hip Replacement on Radiographs
Background: There are few methods for accurately assessing the risk of total hip arthroplasty (THA) in patients with osteoarthritis. A novel and reliable method that could play a substantial role in research and clinical routine should be investigated. The purpose of the present study was to develop a deep-learning model that can reliably predict the risk of THA with use of radiographic images and clinical symptom data. Methods: This retrospective, multicenter, case-control study assessed hip joints on weighted-bearing anteroposterior pelvic radiographs obtained from Osteoarthritis Initiative (OAI) participants. Participants who underwent THA were matched to controls according to age, sex, body mass index, and ethnicity. Cases and controls were uniformly split into training, validation, and testing data sets at proportions of 72% (n = 528), 14% (n = 104), and 14% (n = 104), respectively. Images and clinical symptom data were passed through a detection model and a deep convolutional neural network (DCNN) model to predict the probability of THA within 9 years as well as the most likely time period for THA (0 to 2 years, 3 to 5 years, 6 to 9 years). Model performance was assessed with use of the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity in the testing set. Results: A total of 736 participants were evaluated, including 184 cases and 552 controls. The prediction model achieved an overall accuracy, sensitivity, and specificity of 91.35%, 92.59% and 86.96%, respectively, with an AUC of 0.944, for THA within 9 years. The AUC of the DCNN model for assessing the most likely time period was 0.907 for 0 to 2 years, 0.916 for 3 to 5 years, and 0.841 for 6 to 9 years. Gradient-weighted class activation mapping closely corresponded to regions affecting the prediction of the DCNN model. Conclusions: The proposed DCNN model is a reliable and valid method to predict the probability of THA—within limitations. It could assist clinicians in patient counseling and decision-making regarding the timing of the intervention. In the future, by increasing the size of the data set, enhancing the ethnic and socioeconomic diversity of the participants, and improving the follow-up rate, the quality of the conclusions can be further improved. Level of Evidence: Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.
- Open Access
Multifunctional Extracellular Matrix Hydrogel with Self-Healing Properties and Promoting Angiogenesis as an Immunoregulation Platform for Diabetic Wound Healing
Treating chronic wounds is a global challenge. In diabetes mellitus cases, long-time and excess inflammatory responses at the injury site may delay the healing of intractable wounds. Macrophage polarization (M1/M2 types) can be closely associated with inflammatory factor generation during wound healing. Quercetin (QCT) is an efficient agent against oxidation and fibrosis that promotes wound healing. It can also inhibit inflammatory responses by regulating M1-to-M2 macrophage polarization. However, its limited solubility, low bioavailability, and hydrophobicity are the main issues restricting its applicability in wound healing. The small intestinal submucosa (SIS) has also been widely studied for treating acute/chronic wounds. It is also being extensively researched as a suitable carrier for tissue regeneration. As an extracellular matrix, SIS can support angiogenesis, cell migration, and proliferation, offering growth factors involved in tissue formation signaling and assisting wound healing. We developed a series of promising biosafe novel diabetic wound repair hydrogel wound dressings with several effects, including self-healing properties, water absorption, and immunomodulatory effects. A full-thickness wound diabetic rat model was constructed for in vivo assessment of QCT@SIS hydrogel, in which hydrogels achieved a markedly increased wound repair rate. Their effect was determined by the promotion of the wound healing process, the thickness of granulation tissue, vascularization, and macrophage polarization during wound healing. At the same time, we injected the hydrogel subcutaneously into healthy rats to perform histological analyses of sections of the heart, spleen, liver, kidney, and lung. We then tested the biochemical index levels in serum to determine the biological safety of the QCT@SIS hydrogel. In this study, the developed SIS showed convergence of biological, mechanical, and wound-healing capabilities. Here, we focused on constructing a self-healing, water-absorbable, immunomodulatory, and biocompatible hydrogel as a synergistic treatment paradigm for diabetic wounds by gelling the SIS and loading QCT for slow drug release.
Open AccessMultifunctional Extracellular Matrix Hydrogel with Self-Healing Properties and Promoting Angiogenesis as an Immunoregulation Platform for Diabetic Wound Healing
Treating chronic wounds is a global challenge. In diabetes mellitus cases, long-time and excess inflammatory responses at the injury site may delay the healing of intractable wounds. Macrophage polarization (M1/M2 types) can be closely associated with inflammatory factor generation during wound healing. Quercetin (QCT) is an efficient agent against oxidation and fibrosis that promotes wound healing. It can also inhibit inflammatory responses by regulating M1-to-M2 macrophage polarization. However, its limited solubility, low bioavailability, and hydrophobicity are the main issues restricting its applicability in wound healing. The small intestinal submucosa (SIS) has also been widely studied for treating acute/chronic wounds. It is also being extensively researched as a suitable carrier for tissue regeneration. As an extracellular matrix, SIS can support angiogenesis, cell migration, and proliferation, offering growth factors involved in tissue formation signaling and assisting wound healing. We developed a series of promising biosafe novel diabetic wound repair hydrogel wound dressings with several effects, including self-healing properties, water absorption, and immunomodulatory effects. A full-thickness wound diabetic rat model was constructed for in vivo assessment of QCT@SIS hydrogel, in which hydrogels achieved a markedly increased wound repair rate. Their effect was determined by the promotion of the wound healing process, the thickness of granulation tissue, vascularization, and macrophage polarization during wound healing. At the same time, we injected the hydrogel subcutaneously into healthy rats to perform histological analyses of sections of the heart, spleen, liver, kidney, and lung. We then tested the biochemical index levels in serum to determine the biological safety of the QCT@SIS hydrogel. In this study, the developed SIS showed convergence of biological, mechanical, and wound-healing capabilities. Here, we focused on constructing a self-healing, water-absorbable, immunomodulatory, and biocompatible hydrogel as a synergistic treatment paradigm for diabetic wounds by gelling the SIS and loading QCT for slow drug release.