ABSTRACT
BACKGROUND: 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. METHODS: 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. RESULTS: 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. CONCLUSIONS: 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. Key messages 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.
ABSTRACT
OBJECTIVE: To study the selective killing effects of adenovirus (Ad)-mediated double suicide gene system driven by KDR promoter (KDR-CdglyTK) on the human hepatic carcinoma cells and human umbilical vein endothelial cells (HUVECs). METHODS: KDR-expressing BEL-7402 and HUVECs and HepG2 cells that did not express KDR were infected by KDR-CdglyTK, and the infection efficiency and the expression of CdglyTK in the cells was detected by RT-PCR. The infected cells were treated with the the prodrugs 5-FC and GCV at different concentrations, and the cell-killing effects and bystander effects were evaluated by MTT method. RESULTS: At the multiplicity of infection (MOI) of 100, the recombinant AdKDR-CDglyTK showed similar infection efficiency in the 3 cell lines. RT-PCR demonstrated CDglyTK expression in the recombinant adenovirus and the 3 infected cell lines. BEL-7402 and HUVECs infected by the KDR-CdglyTK, but not the HepG2 cells, were highly sensitive to the prodrugs (P<0.001). Bystander effects of the double suicide gene system were observed in the coculture of the infected and non-infected BEL-7402 and HUVECs. CONCLUSION: The double suicide gene system driven by KDR promoter has specific killing effect on KDR-expressing hepatocellular carcinoma cells and HUVECs.
Subject(s)
Cytosine Deaminase/genetics , Genes, Transgenic, Suicide/genetics , Liver Neoplasms/pathology , Promoter Regions, Genetic/genetics , Vascular Endothelial Growth Factor Receptor-2/genetics , Adenoviridae/genetics , Apoptosis/genetics , Cells, Cultured , Cytosine Deaminase/metabolism , Endothelial Cells/cytology , Genetic Therapy , Genetic Vectors , Humans , Recombinant Fusion Proteins/genetics , Recombinant Fusion Proteins/metabolism , Thymidine Kinase/genetics , Thymidine Kinase/metabolism , Tumor Cells, Cultured , Umbilical Veins/cytology , Vascular Endothelial Growth Factor Receptor-2/metabolismABSTRACT
OBJECTIVE: To evaluate the effect of adenovirus-mediated double suicide gene (CD/TK) for selective killing of breast cancer cells. METHODS: Vascular endothelial growth factor (VEGF)-expressing MCF-7 cells and normal human mammary epithelial cells that did not express VEGF were infected with the adenovirus containing VEGFP-CD/TK-GFP genes. CD/TK gene expression in the infected cells was detected by RT-PCR. After treatment of the infected cells with GCV and/or 5-FC, the cell growth status was evaluated using MTT assay, and the cell cycle changes were detected with flow cytometry. In nude mice bearing human breast cancer, the recombinant adenovirus vector was injected directly into the tumor followed by intraperitoneal injection of the prodrugs GCV and/or 5-FC, and the subsequent tumor growth was observed. RESULTS: The recombinant adenovirus achieved similar infection rates in MCF-7 and human mammary epithelial cells, and the rates increased gradually with the multiplicity of infection (MOI) of the virus. RT-PCR demonstrated the presence of CD/TK gene product in infected MCF-7 cells, but not in the infected mammary epithelial cells. The infected MCF-7 cells, but not the mammary epithelial cells, were highly sensitive to the pro-drugs. The CD/TK fusion gene system showed significantly greater efficiency than either of the single suicide gene in killing the target cells (P<0.01). At the MOI of 100, treatment of the infected cells with the pro-drugs resulted in increased cell percentage in G(0)-G(1) phase and decreased percentage in S phase. In nude mice bearing MCF-7 cell-derived subcutaneous tumor, treatment with the double suicide gene system significantly inhibited the tumor growth, showing much stronger effect than either of the single suicide gene (P<0.01). CONCLUSION: The adenovirus-mediated CD/TK double suicide gene driven by VEGF promoter combined with GCV and 5-FC treatment can be an effective therapy against experimental breast cancer, and produces much greater efficacy than the single suicide gene CD/TK combined with GCV or 5-FC.