Qinchang Chen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Qinchang Chen has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
Churong Lin
Department of Radiology, The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, People’s Republic of China
1 shared publicationGaochen Bai
Department of Paediatric Urology, Guangzhou Women and Children's Medical Center Guangzhou Medical University Guangzhou China
1 shared publicationGuiwu Huang
Department of Orthopedics The First Affiliated Hospital of Sun Yat‐Sen University Guangzhou China
1 shared publication- Guo-Chang Liu
Department of Pediatric Urology, Guangzhou Women and Children’s Medical Center, Guangzhou Medical University, Guangzhou, China
1 shared publication - Jiahao Cai
Department of Pediatric Neurology, Guangzhou Women and Children's Medical Center Guangzhou Medical University Guangzhou China
1 shared publication Jun-Jun Shen
Department of Pediatric Surgery , Zhujiang Hospital No. 253, Industrial Avenue Middle, Guangzhou 510282, Guangdong
1 shared publication
Latest Publications
- 29 July 2024
- Published by Oxford University Press (OUP) in Postgraduate Medical Journal
- Vol. 101 (1191) , 37-44
Facial recognition models for identifying genetic syndromes associated with pulmonary stenosis in children
Williams–Beuren syndrome, Noonan syndrome, and Alagille syndrome are common types of genetic syndromes (GSs) characterized by distinct facial features, pulmonary stenosis, and delayed growth. In clinical practice, differentiating these three GSs remains a challenge. Facial gestalts serve as a diagnostic tool for recognizing Williams–Beuren syndrome, Noonan syndrome, and Alagille syndrome. Pretrained foundation models (PFMs) can be considered the foundation for small-scale tasks. By pretraining with a foundation model, we propose facial recognition models for identifying these syndromes. A total of 3297 (n = 1666) facial photos were obtained from children diagnosed with Williams–Beuren syndrome (n = 174), Noonan syndrome (n = 235), and Alagille syndrome (n = 51), and from children without GSs (n = 1206). The photos were randomly divided into five subsets, with each syndrome and non-GS equally and randomly distributed in each subset. The proportion of the training set and the test set was 4:1. The ResNet-100 architecture was employed as the backbone model. By pretraining with a foundation model, we constructed two face recognition models: one utilizing the ArcFace loss function, and the other employing the CosFace loss function. Additionally, we developed two models using the same architecture and loss function but without pretraining. The accuracy, precision, recall, and F1 score of each model were evaluated. Finally, we compared the performance of the facial recognition models to that of five pediatricians. Among the four models, ResNet-100 with a PFM and CosFace loss function achieved the best accuracy (84.8%). Of the same loss function, the performance of the PFMs significantly improved (from 78.5% to 84.5% for the ArcFace loss function, and from 79.8% to 84.8% for the CosFace loss function). With and without the PFM, the performance of the CosFace loss function models was similar to that of the ArcFace loss function models (79.8% vs 78.5% without PFM; 84.8% vs 84.5% with PFM). Among the five pediatricians, the highest accuracy (0.700) was achieved by the senior-most pediatrician with genetics training. The accuracy and F1 scores of the pediatricians were generally lower than those of the models. A facial recognition-based model has the potential to improve the identification of three common GSs with pulmonary stenosis. PFMs might be valuable for building screening models for facial recognition. What is already known on this topic: Early identification of genetic syndromes (GSs) is crucial for the management and prognosis of children with pulmonary stenosis (PS). Facial phenotyping with convolutional neural networks (CNNs) often requires large-scale training data, limiting its usefulness for GSs. What this study adds: We successfully built multi-classification models based on face recognition using a CNN to accurately identify three common PS-associated GSs. ResNet-100 with a pretrained foundation model (PFM) and CosFace loss function achieved the best accuracy (84.8%). Pretrained with the foundation model, the performance of the models significantly improved, although the impact of the type of loss function appeared to be minimal. How this study might affect research, practice, or policy: A facial recognition-based model has the potential to improve the identification of GSs in children with PS. The PFM might be valuable for building identification models for facial detection.
- 29 July 2024
- Published by Oxford University Press (OUP) in Postgraduate Medical Journal
- Vol. 101 (1191) , 37-44
Facial recognition models for identifying genetic syndromes associated with pulmonary stenosis in children
Williams–Beuren syndrome, Noonan syndrome, and Alagille syndrome are common types of genetic syndromes (GSs) characterized by distinct facial features, pulmonary stenosis, and delayed growth. In clinical practice, differentiating these three GSs remains a challenge. Facial gestalts serve as a diagnostic tool for recognizing Williams–Beuren syndrome, Noonan syndrome, and Alagille syndrome. Pretrained foundation models (PFMs) can be considered the foundation for small-scale tasks. By pretraining with a foundation model, we propose facial recognition models for identifying these syndromes. A total of 3297 (n = 1666) facial photos were obtained from children diagnosed with Williams–Beuren syndrome (n = 174), Noonan syndrome (n = 235), and Alagille syndrome (n = 51), and from children without GSs (n = 1206). The photos were randomly divided into five subsets, with each syndrome and non-GS equally and randomly distributed in each subset. The proportion of the training set and the test set was 4:1. The ResNet-100 architecture was employed as the backbone model. By pretraining with a foundation model, we constructed two face recognition models: one utilizing the ArcFace loss function, and the other employing the CosFace loss function. Additionally, we developed two models using the same architecture and loss function but without pretraining. The accuracy, precision, recall, and F1 score of each model were evaluated. Finally, we compared the performance of the facial recognition models to that of five pediatricians. Among the four models, ResNet-100 with a PFM and CosFace loss function achieved the best accuracy (84.8%). Of the same loss function, the performance of the PFMs significantly improved (from 78.5% to 84.5% for the ArcFace loss function, and from 79.8% to 84.8% for the CosFace loss function). With and without the PFM, the performance of the CosFace loss function models was similar to that of the ArcFace loss function models (79.8% vs 78.5% without PFM; 84.8% vs 84.5% with PFM). Among the five pediatricians, the highest accuracy (0.700) was achieved by the senior-most pediatrician with genetics training. The accuracy and F1 scores of the pediatricians were generally lower than those of the models. A facial recognition-based model has the potential to improve the identification of three common GSs with pulmonary stenosis. PFMs might be valuable for building screening models for facial recognition. What is already known on this topic: Early identification of genetic syndromes (GSs) is crucial for the management and prognosis of children with pulmonary stenosis (PS). Facial phenotyping with convolutional neural networks (CNNs) often requires large-scale training data, limiting its usefulness for GSs. What this study adds: We successfully built multi-classification models based on face recognition using a CNN to accurately identify three common PS-associated GSs. ResNet-100 with a pretrained foundation model (PFM) and CosFace loss function achieved the best accuracy (84.8%). Pretrained with the foundation model, the performance of the models significantly improved, although the impact of the type of loss function appeared to be minimal. How this study might affect research, practice, or policy: A facial recognition-based model has the potential to improve the identification of GSs in children with PS. The PFM might be valuable for building identification models for facial detection.
- Open Access
- 7 December 2022
- Published by Wiley in Cancer Innovation
- Vol. 1 (4) , 316-327
Causal relationship between 14 site‐specific cancers and venous thromboembolism
Background: It has been observed that cancer and venous thromboembolism (VTE) are associated, but anticancer therapy may violate the causality. Therefore, this study aimed to elucidate the causal relationship of various cancers to VTE using Mendelian randomization (MR).Methods: Three MR methods were used to estimate causal effects: Inverse variance weighted (IVW), MR‐Egger and weighted median. Sensitivity analyses included Cochran's Q‐test, MR‐Egger intercept test and MR‐PRESSO. Gene ontology enrichment analysis was performed to elucidate the underlying mechanisms of VTE development in cancer patients.Results: The primary IVW approach showed that non‐Hodgkin's lymphoma (NHL) might increase the risk of VTE (odds ratio [OR]: 1.20, 95% confidence interval [95% CI]: 1.00–1.44, p = 0.045), while melanoma possibly reduced the risk of VTE (OR: 0.89, 95% CI: 0.82–0.97, p = 0.006), although there was no significance after adjustment for multiple testing. No association was observed between VTE risk and other site‐specific cancers. Gene ontology enrichment analysis revealed that vitamin D played an important role in the development of VTE in cancer patients.Conclusions: Our findings suggested that genetically predicted NHL was associated with higher VTE risk, whereas melanoma had lower VTE risk compared with other site‐specific cancers. Moreover, this study suggested that anticancer therapy and increased extensive examination might play a more important role in VTE development than the nature of cancer.
- 7 December 2022
- Published by Wiley in Cancer Innovation
- Vol. 1 (4) , 316-327
Open AccessCausal relationship between 14 site‐specific cancers and venous thromboembolism
Background: It has been observed that cancer and venous thromboembolism (VTE) are associated, but anticancer therapy may violate the causality. Therefore, this study aimed to elucidate the causal relationship of various cancers to VTE using Mendelian randomization (MR).Methods: Three MR methods were used to estimate causal effects: Inverse variance weighted (IVW), MR‐Egger and weighted median. Sensitivity analyses included Cochran's Q‐test, MR‐Egger intercept test and MR‐PRESSO. Gene ontology enrichment analysis was performed to elucidate the underlying mechanisms of VTE development in cancer patients.Results: The primary IVW approach showed that non‐Hodgkin's lymphoma (NHL) might increase the risk of VTE (odds ratio [OR]: 1.20, 95% confidence interval [95% CI]: 1.00–1.44, p = 0.045), while melanoma possibly reduced the risk of VTE (OR: 0.89, 95% CI: 0.82–0.97, p = 0.006), although there was no significance after adjustment for multiple testing. No association was observed between VTE risk and other site‐specific cancers. Gene ontology enrichment analysis revealed that vitamin D played an important role in the development of VTE in cancer patients.Conclusions: Our findings suggested that genetically predicted NHL was associated with higher VTE risk, whereas melanoma had lower VTE risk compared with other site‐specific cancers. Moreover, this study suggested that anticancer therapy and increased extensive examination might play a more important role in VTE development than the nature of cancer.
- Open Access
- 1 December 2021
- Published by Taylor & Francis in International Journal of General Medicine
- Vol. ume 14, 9273-9286
The Association of Waist Circumference and the Risk of Deep Vein Thrombosis
Objective: In this study, we aimed to use a two sample Mendelian randomization (MR) method to identify a potentially causality between waist circumference and the risk of deep vein thrombosis (DVT). Methods: With a two‐sample MR approach, we analyzed the summary data. The main analysis was performed by using the summary genetic data from two large consortium cohorts. Three MR approaches were used to explore MR estimates of waist circumference for DVT (inverse‐variance weighted [IVW] approach, weighted median method and MR‐Egger method). A total of 224 single nucleotide polymorphisms (SNPs) were identified associated with the level of waist circumference at statistical significance (P < 5*10− 8; linkage disequilibrium r2 < 0.1). Results: The result of IVW indicated the positive association between waist circumference and the risk of DVT (OR 1.012, 95% CI 1.009– 1.014, P 7.627E-17). The other two methods were observed with consistent result. MR-Egger regression analysis indicated that no evidence for the presence of directional horizontal pleiotropy. Additionally, DVT was not a causal factor for waist circumference. Conclusion: In summary, we used the GWAS genetic data from two large consortium cohorts and indicated the positive association between waist circumference and DVT. Further researches are needed to investigate potential mechanism and clarify the role of waist circumference on DVT.
- 1 December 2021
- Published by Taylor & Francis in International Journal of General Medicine
- Vol. ume 14, 9273-9286
Open AccessThe Association of Waist Circumference and the Risk of Deep Vein Thrombosis
Objective: In this study, we aimed to use a two sample Mendelian randomization (MR) method to identify a potentially causality between waist circumference and the risk of deep vein thrombosis (DVT). Methods: With a two‐sample MR approach, we analyzed the summary data. The main analysis was performed by using the summary genetic data from two large consortium cohorts. Three MR approaches were used to explore MR estimates of waist circumference for DVT (inverse‐variance weighted [IVW] approach, weighted median method and MR‐Egger method). A total of 224 single nucleotide polymorphisms (SNPs) were identified associated with the level of waist circumference at statistical significance (P < 5*10− 8; linkage disequilibrium r2 < 0.1). Results: The result of IVW indicated the positive association between waist circumference and the risk of DVT (OR 1.012, 95% CI 1.009– 1.014, P 7.627E-17). The other two methods were observed with consistent result. MR-Egger regression analysis indicated that no evidence for the presence of directional horizontal pleiotropy. Additionally, DVT was not a causal factor for waist circumference. Conclusion: In summary, we used the GWAS genetic data from two large consortium cohorts and indicated the positive association between waist circumference and DVT. Further researches are needed to investigate potential mechanism and clarify the role of waist circumference on DVT.