Facial recognition models for identifying genetic syndromes associated with pulmonary stenosis in children.

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.

Discrepancy between arterial oxygen saturation and pulse oximetry measurement in a Chinese pediatric patient cohort.

Background: Increasing evidence suggest a racial bias in pulse oximetry measurement, but this was under investigated in Asian pediatric populations. Methods: Via the Pediatric Intensive Care database, this retrospective study included pediatric patient records of arterial oxygen saturation (SaO2) and oxygen saturation on pulse oximetry (SpO2) measured within 10 min. Discrepancy was examined, and potential predictors of occult hypoxemia (defined as SaO2 <88% with the paired SpO2 ≥92%) as well as its association with outcomes were explored by logistic regression. Results: A total of 390 patients were included with 454 pairs of SaO2-SpO2 readings. The study population consisted of Han Chinese (99.0%) and 43.6% were female. Occult hypoxemia was observed in 20.0% of the patients, with a mean SaO2 of 71.4 ± 15.8%. Potential predictors of occult hypoxemia included female, being first admitted to cardiac ICU, congenital heart disease, increased heart rate, while patients with prior surgery records were less likely to experience occult hypoxemia. Patients with occult hypoxemia had numerically higher in-ICU mortality (16.7% versus 10.9%) and in-hospital mortality (17.9% versus 10.9%), but the associations were not statistically significant. Conclusions: There was a substantial proportion of hypoxemia that was not detected by pulse oximetry in the Chinese pediatric patients, which might be predicted by several characteristics and seemed to associate with mortality.

Risk factors associated with unplanned endotracheal self-extubation of hospitalized intubated patients: a 3-year retrospective case-control study.

This 3-year retrospective case-control study aimed to identify risk factors associated with unplanned endotracheal self-extubation (UESE) of hospitalized intubated patients and to compare unplanned and planned extubation groups' characteristics of patients and nurses, vital signs, serum laboratory values, Glasgow Coma Scale scores, Acute Physiology and Chronic Health Evaluation II (APACHE II) scores, and use of physical restraints and sedatives. The study found that most UESEs occurred during evening or night shifts or during shifts staffed by nurses with less experience and less education. Most of the self-extubated patients (80%) were physically restrained. Pulse rate and APACHE II score were both significant predictors of UESE. Efforts to prevent UESEs should include identification of patients at higher risk.

Transcriptionally mediated inhibition of telomerase of fungal immunomodulatory protein from Ganoderma tsugae in A549 human lung adenocarcinoma cell line.

Telomerase expression is the hallmark of tumor cells, and activation of this ribonucleoprotein complex may be a rate-limiting or critical step in cellular immortalization and oncogenesis. Fungal immunomodulatory protein, FIP-gts, has been isolated from Ganoderma tsugae. In the present study, we expressed and purified the recombinant fungal immunomodulatory protein reFIP-gts in E. coli. We found that reFIP-gts significantly and selectively inhibits the growth of A549 cancer cells while not affecting the growth of normal MRC-5 fibroblasts. The reFIP-gts suppression of telomerase activity is concentration-dependent, due to the downregulation of the telomerase catalytic subunit (hTERT). It also happens at the mRNA level. These results were confirmed by transient transfections of A549 cells with pGL3-Basic plasmid constructs containing the functional hTERT promoter and its E-box-deleted sequences cloned upstream of a luciferase reporter gene. With electrophoretic mobility shift assays and Western blotting, we demonstrated that in response to reFIP-gts, binding of c-myc transcriptional factor to the E-box sequence on the hTERT promoter is inhibited. These results show that reFIP-gts suppresses telomerase activity and inhibits transcriptional regulation of hTERT via a c-myc-responsive element-dependent mechanism. Our findings provide new insight into both the anticancer function of reFIP-gts and the regulation of hTERT/telomerase expression, which may be valuable in the development of a promising chemopreventive agent.

Fungal immunomodulatory protein from Flammulina velutipes induces interferon-gamma production through p38 mitogen-activated protein kinase signaling pathway.

FIP-fve is a fungal immunomodulatory protein purified from Flammulina velutipes, an edible golden needle mushroom thought to possess potent immunomodulatory properties. When examined for its effects on lymphocytes, FIP-fve exhibited potent mitogenic effects on human peripheral blood lymphocytes, inducing G1/G0 to S phase proliferation. T cells activated by FIP-fve show significant production and secretion of interferon-gamma (IFN-gamma) associated with intercellular adhesion molecule 1 expression but low detectable levels of interleukin-4 in vitro or in vivo. However, SB203580, the p38 mitogen-activated protein kinase (p38 MAPK) inhibitor, can fully abolish the production of IFN-gamma induced by FIP-fve. At the same time, SB203580 only partially prevents the lymphocytes from progressing from G1 to S phase of the cell cycle. These findings demonstrate that FIP-fve is a potent T-cell activator, mediating its effects via cytokine regulation of p38 MAPK. The immunoprophylatic effects of FIP-fve in Th2-mediated allergic anaphylaxis are believed to be associated with the ability of FIP-fve to enhance activation of IFN-gamma-releasing Th1 cells.