scholarly journals Genetic syndromes screening by facial recognition technology: VGG-16 screening model construction and evaluation

2021 ◽  
Vol 16 (1) ◽  
Author(s):  
Dian Hong ◽  
Ying-Yi Zheng ◽  
Ying Xin ◽  
Ling Sun ◽  
Hang Yang ◽  
...  
Keyword(s):  
Facial Recognition ◽  
Characteristic Curve ◽  
Facial Dysmorphism ◽  
Healthy Children ◽  
Genetic Syndromes ◽  
Recognition Model ◽  
Oxford University ◽  
Screening Model ◽  
Fold Cross Validation ◽  
Visual Geometry

Abstract Background Many genetic syndromes (GSs) have distinct facial dysmorphism, and facial gestalts can be used as a diagnostic tool for recognizing a syndrome. Facial recognition technology has advanced in recent years, and the screening of GSs by facial recognition technology has become feasible. This study constructed an automatic facial recognition model for the identification of children with GSs. Results A total of 456 frontal facial photos were collected from 228 children with GSs and 228 healthy children in Guangdong Provincial People's Hospital from Jun 2016 to Jan 2021. Only one frontal facial image was selected for each participant. The VGG-16 network (named after its proposal lab, Visual Geometry Group from Oxford University) was pretrained by transfer learning methods, and a facial recognition model based on the VGG-16 architecture was constructed. The performance of the VGG-16 model was evaluated by five-fold cross-validation. Comparison of VGG-16 model to five physicians were also performed. The VGG-16 model achieved the highest accuracy of 0.8860 ± 0.0211, specificity of 0.9124 ± 0.0308, recall of 0.8597 ± 0.0190, F1-score of 0.8829 ± 0.0215 and an area under the receiver operating characteristic curve of 0.9443 ± 0.0276 (95% confidence interval: 0.9210–0.9620) for GS screening, which was significantly higher than that achieved by human experts. Conclusions This study highlighted the feasibility of facial recognition technology for GSs identification. The VGG-16 recognition model can play a prominent role in GSs screening in clinical practice.

scholarly journals Automatic Facial Recognition of Williams-Beuren Syndrome Based on Deep Convolutional Neural Networks

2021 ◽  
Vol 9 ◽  
Author(s):  
Hui Liu ◽  
Zi-Hua Mo ◽  
Hang Yang ◽  
Zheng-Fu Zhang ◽  
Dian Hong ◽  
...  
Keyword(s):  
Neural Networks ◽  
Face Recognition ◽  
Transfer Learning ◽  
Convolutional Neural Networks ◽  
Facial Recognition ◽  
Small Scale ◽  
Healthy Children ◽  
Genetic Syndromes ◽  
Deep Convolutional Neural Networks ◽  
Genetic Syndrome

Background: Williams-Beuren syndrome (WBS) is a rare genetic syndrome with a characteristic “elfin” facial gestalt. The “elfin” facial characteristics include a broad forehead, periorbital puffiness, flat nasal bridge, short upturned nose, wide mouth, thick lips, and pointed chin. Recently, deep convolutional neural networks (CNNs) have been successfully applied to facial recognition for diagnosing genetic syndromes. However, there is little research on WBS facial recognition using deep CNNs.Objective: The purpose of this study was to construct an automatic facial recognition model for WBS diagnosis based on deep CNNs.Methods: The study enrolled 104 WBS children, 91 cases with other genetic syndromes, and 145 healthy children. The photo dataset used only one frontal facial photo from each participant. Five face recognition frameworks for WBS were constructed by adopting the VGG-16, VGG-19, ResNet-18, ResNet-34, and MobileNet-V2 architectures, respectively. ImageNet transfer learning was used to avoid over-fitting. The classification performance of the facial recognition models was assessed by five-fold cross validation, and comparison with human experts was performed.Results: The five face recognition frameworks for WBS were constructed. The VGG-19 model achieved the best performance. The accuracy, precision, recall, F1 score, and area under curve (AUC) of the VGG-19 model were 92.7 ± 1.3%, 94.0 ± 5.6%, 81.7 ± 3.6%, 87.2 ± 2.0%, and 89.6 ± 1.3%, respectively. The highest accuracy, precision, recall, F1 score, and AUC of human experts were 82.1, 65.9, 85.6, 74.5, and 83.0%, respectively. The AUCs of each human expert were inferior to the AUCs of the VGG-16 (88.6 ± 3.5%), VGG-19 (89.6 ± 1.3%), ResNet-18 (83.6 ± 8.2%), and ResNet-34 (86.3 ± 4.9%) models.Conclusions: This study highlighted the possibility of using deep CNNs for diagnosing WBS in clinical practice. The facial recognition framework based on VGG-19 could play a prominent role in WBS diagnosis. Transfer learning technology can help to construct facial recognition models of genetic syndromes with small-scale datasets.

A-176 Undercover Performance Validity Testing: Proposed Embedded Validity Indicators for Pediatric Neuropsychological Assessment

2021 ◽  
Vol 36 (6) ◽  
pp. 1231-1231
Author(s):  
Naomi R Kaswan ◽  
Ryan C Thompson ◽  
Yelena Markiv ◽  
Aubrey Deenen ◽  
Haig V Pilavjian ◽  
...  
Keyword(s):  
Sport Participation ◽  
Word Reading ◽  
Digit Span ◽  
Characteristic Curve ◽  
Test Word ◽  
Healthy Children ◽  
Performance Validity ◽  
Validity Indicators ◽  
Neuropsychological Evaluations ◽  
Embedded Validity Indicators

Abstract Objective Literature supports the use of the Delis-Kaplan Executive Function System Trail Making Test Conditions 4/2 ratio (TMT 4/2) and Stroop Color Word Test Word Reading (WR) as embedded validity indicators (EVIs) with adults (Erdodi et al., 2018; Guise et al., 2012) and the Wechsler Abbreviated Scale of Intelligence, 2nd Edition Matrix Reasoning (MR) as an EVI with children (Sussman et al., 2017). This study assessed the utility of these measures as EVIs in healthy children, compared to the Test of Memory Malingering Trial 1 (TOMM1 < 45; Perna & Loughan, 2013) and Reliable Digit Span (RDS). Method Participants (n = 99, 68.7% male, Mage = 11.9) completed baseline neuropsychological evaluations for sport participation, including the aforementioned measures. Receiver operator characteristic curve analysis was used to determine whether TMT 4/2, MR, and WR accurately categorized valid performance. Results TMT 4/2 yielded adequate sensitivity (0.83–1.00) but poor specificity (0.07–0.09) when predicting TOMM1 and RDS pass/fail performances. MR yielded adequate sensitivity (1.00) and specificity (0.92) when predicting RDS pass/fail performance and adequate specificity (0.92) and poor sensitivity (0.18) when predicting TOMM1 pass/fail performance. The only EVI that produced better than chance accuracy was MR when predicting RDS pass/fail performance (area under the curve [AUC] = 0.98). All participants failed the WR cutoff, suggesting poor specificity. Conclusion Results suggest that MR was the only EVI that achieved minimally acceptable specificity (≥0.90) in children. MR performed adequately when detecting valid performances but variably when detecting invalid performances; therefore, MR may be used alongside well-established performance validity tests with children but not independently.

Biomechanical and Anthropometric Factors That Differentiate National- and Regional-Level Judo Players

2021 ◽  
pp. 1-7
Author(s):  
Felipe Guimarães Teixeira ◽  
Paulo Tadeu Cardozo Ribeiro Rosa ◽  
Roger Gomes Tavares Mello ◽  
Jurandir Nadal
Keyword(s):  
Cross Validation ◽  
Center Of Pressure ◽  
Characteristic Curve ◽  
Multivariable Analysis ◽  
Stepwise Logistic Regression ◽  
Mean Velocity ◽  
Bone Width ◽  
Longitudinal Length ◽  
Judo Athletes ◽  
Fold Cross Validation

Purpose: The study aimed to identify the variables that differentiate judo athletes at national and regional levels. Multivariable analysis was applied to biomechanical, anthropometric, and Special Judo Fitness Test (SJFT) data. Method: Forty-two male judo athletes from 2 competitive groups (14 national and 28 state levels) performed the following measurements and tests: (1) skinfold thickness, (2) circumference, (3) bone width, (4) longitudinal length, (5) stabilometric tests, (6) dynamometric tests, and (7) SJFT. The variables with significant differences in the Wilcoxon rank-sum test were used in stepwise logistic regression to select those that better separate the groups. The authors considered models with a maximum of 3 variables to avoid overfitting. They used 7-fold cross validation to calculate optimism-corrected measures of model performance. Results: The 3 variables that best differentiated the groups were the epicondylar humerus width, the total number of throws on the SJFT, and the stabilometric mean velocity of the center of pressure in the mediolateral direction. The area under the receiver-operating-characteristic curve for the model (based on 7-fold cross validation) was 0.95. Conclusion: This study suggests that a reduced set of anthropometric, biomechanical, and SJFT variables can differentiate judo athlete’s levels.

scholarly journals Postprocessing Next-Day Ensemble Probabilistic Precipitation Forecasts Using Random Forests

2019 ◽  
Vol 34 (6) ◽  
pp. 2017-2044 ◽  
Author(s):  
Eric D. Loken ◽  
Adam J. Clark ◽  
Amy McGovern ◽  
Montgomery Flora ◽  
Kent Knopfmeier
Keyword(s):  
Random Forest ◽  
Operating Characteristic ◽  
Characteristic Curve ◽  
Ensemble Forecast ◽  
Grid Spacing ◽  
Relative Operating Characteristic ◽  
Advantages And Disadvantages ◽  
Skill Scores ◽  
Systematic Biases ◽  
Fold Cross Validation

Abstract Most ensembles suffer from underdispersion and systematic biases. One way to correct for these shortcomings is via machine learning (ML), which is advantageous due to its ability to identify and correct nonlinear biases. This study uses a single random forest (RF) to calibrate next-day (i.e., 12–36-h lead time) probabilistic precipitation forecasts over the contiguous United States (CONUS) from the Short-Range Ensemble Forecast System (SREF) with 16-km grid spacing and the High-Resolution Ensemble Forecast version 2 (HREFv2) with 3-km grid spacing. Random forest forecast probabilities (RFFPs) from each ensemble are compared against raw ensemble probabilities over 496 days from April 2017 to November 2018 using 16-fold cross validation. RFFPs are also compared against spatially smoothed ensemble probabilities since the raw SREF and HREFv2 probabilities are overconfident and undersample the true forecast probability density function. Probabilistic precipitation forecasts are evaluated at four precipitation thresholds ranging from 0.1 to 3 in. In general, RFFPs are found to have better forecast reliability and resolution, fewer spatial biases, and significantly greater Brier skill scores and areas under the relative operating characteristic curve compared to corresponding raw and spatially smoothed ensemble probabilities. The RFFPs perform best at the lower thresholds, which have a greater observed climatological frequency. Additionally, the RF-based postprocessing technique benefits the SREF more than the HREFv2, likely because the raw SREF forecasts contain more systematic biases than those from the raw HREFv2. It is concluded that the RFFPs provide a convenient, skillful summary of calibrated ensemble output and are computationally feasible to implement in real time. Advantages and disadvantages of ML-based postprocessing techniques are discussed.

scholarly journals Apolipoprotein A-V Is a Novel Diagnostic and Prognostic Predictor in Pediatric Patients with Sepsis: A Prospective Pilot Study in PICU

2020 ◽  
Vol 2020 ◽  
pp. 1-9 ◽  
Author(s):  
Chunxia Wang ◽  
Yun Cui ◽  
Huijie Miao ◽  
Xi Xiong ◽  
Jiaying Dou ◽  
...  
Keyword(s):  
Pilot Study ◽  
Pediatric Patients ◽  
Characteristic Curve ◽  
Kidney Injury ◽  
Healthy Children ◽  
Prognostic Predictor ◽  
Prospective Pilot Study ◽  
Serum Proteomics ◽  
Apolipoprotein A

Background. Sepsis induces the release of lipid mediators, which control both lipid metabolism and inflammation. However, the role of serum apolipoprotein A-V (ApoA5) in sepsis is poorly understood in pediatric patients. Methods. ApoA5 was screened from serum proteomics profile in lipopolysaccharide- (LPS-) treated mice for 2 h, 24 h, and controls. Then, we conducted a prospective pilot study, and patients with sepsis admitted to a pediatric intensive care unit (PICU) were enrolled from January 2018 to December 2018. Serum ApoA5 levels on PICU admission were determined using enzyme-linked immunosorbent assays (ELISA). Blood samples from 30 healthy children were used as control. The correlation of ApoA5 with the clinical and laboratory parameters was analyzed. Logistic regression analyses and receiver operating characteristic curve (ROC) analysis were used to investigate the potential role of serum ApoA5 as a prognostic predictor for PICU mortality in pediatric patients with sepsis. Results. A total of 101 patients with sepsis were enrolled in this study. The PICU mortality rate was 10.9% (11/101). Serum ApoA5 levels on PICU admission were significantly lower in nonsurvivors with sepsis compared with survivors (P=0.009). In subgroup analysis, serum levels of ApoA5 were significantly correlated with sepsis-associated multiple organ dysfunction syndrome (MODS) (P<0.001), shock (P=0.002), acute kidney injury (AKI) (P<0.001), acute liver injury (ALI) (P=0.002), and gastrointestinal (GI) dysfunction (P=0.012), but not respiratory failure, brain injury, and pathogenic species (all P>0.05). Correlation analyses revealed significant correlations of serum ApoA5 with Ca2+ concentration. Remarkably, the area under ROC curve (AUC) for serum ApoA5 levels on PICU admission was 0.789 for prediction of PICU mortality with a sensitivity of 75% and a specificity of 84.5% at a threshold value of 822 ng/mL. Conclusions. Serum ApoA5 level is associated with sepsis-associated shock, AKI, ALI, GI dysfunction, or MODS in children. Moreover, the findings of the present study suggest a prognostic value of ApoA5 in children with sepsis, and lower serum ApoA5 than 822 ng/mL predicts worse outcome in pediatric sepsis.

Predicting Procedure Step Performance From Operator and Text Features: A Critical First Step Toward Machine Learning-Driven Procedure Design

2020 ◽  
pp. 001872082095858
Author(s):  
Anthony D. McDonald ◽  
Nilesh Ade ◽  
S. Camille Peres
Keyword(s):  
Machine Learning ◽  
Language Processing ◽  
Characteristic Curve ◽  
Design Guidelines ◽  
Machine Learning Algorithms ◽  
Design Alternatives ◽  
Subjective Assessments ◽  
Text Features ◽  
Partial Dependence ◽  
Fold Cross Validation

Objective The goal of this study is to assess machine learning for predicting procedure performance from operator and procedure characteristics. Background Procedures are vital for the performance and safety of high-risk industries. Current procedure design guidelines are insufficient because they rely on subjective assessments and qualitative analyses that struggle to integrate and quantify the diversity of factors that influence procedure performance. Method We used data from a 25-participant study with four procedures, conducted on a high-fidelity oil extraction simulation to develop logistic regression (LR), random forest (RF), and decision tree (DT) algorithms that predict procedure step performance from operator, step, readability, and natural language processing-based features. Features were filtered using the Boruta approach. The algorithms were trained and optimized with a repeated 10-fold cross-validation. After training, inference was performed using variable importance and partial dependence plots. Results The RF, DT, and LR algorithms with all features had an area under the receiver operating characteristic curve (AUC) of 0.78, 0.77, and 0.75, respectively, and significantly outperformed the LR with only operator features (LROP), with an AUC of 0.61. The most important features were experience, familiarity, total words, and character-based metrics. The partial dependence plots showed that steps with fewer words, abbreviations, and characters were correlated with correct step performance. Conclusion Machine learning algorithms are a promising approach for predicting step-level procedure performance, with acknowledged limitations on interpolating to nonobserved data, and may help guide procedure design after validation with additional data on further tasks. Application After validation, the inferences from these models can be used to generate procedure design alternatives.

scholarly journals Efficacy of Quantitative Muscle Ultrasound Using Texture-Feature Parametric Imaging in Detecting Pompe Disease in Children

Entropy ◽  
2019 ◽  
Vol 21 (7) ◽  
pp. 714
Author(s):  
Hong-Jen Chiou ◽  
Chih-Kuang Yeh ◽  
Hsuen-En Hwang ◽  
Yin-Yin Liao
Keyword(s):  
Pompe Disease ◽  
Late Onset ◽  
Characteristic Curve ◽  
Texture Feature ◽  
Rectus Femoris ◽  
Parametric Imaging ◽  
Imaging Method ◽  
Healthy Children ◽  
Linear Discriminant ◽  
Optimal Feature

Pompe disease is a hereditary neuromuscular disorder attributed to acid α-glucosidase deficiency, and accurately identifying this disease is essential. Our aim was to discriminate normal muscles from neuropathic muscles in children affected by Pompe disease using a texture-feature parametric imaging method that simultaneously considers microstructure and macrostructure. The study included 22 children aged 0.02–54 months with Pompe disease and six healthy children aged 2–12 months with normal muscles. For each subject, transverse ultrasound images of the bilateral rectus femoris and sartorius muscles were obtained. Gray-level co-occurrence matrix-based Haralick’s features were used for constructing parametric images and identifying neuropathic muscles: autocorrelation (AUT), contrast, energy (ENE), entropy (ENT), maximum probability (MAXP), variance (VAR), and cluster prominence (CPR). Stepwise regression was used in feature selection. The Fisher linear discriminant analysis was used for combination of the selected features to distinguish between normal and pathological muscles. The VAR and CPR were the optimal feature set for classifying normal and pathological rectus femoris muscles, whereas the ENE, VAR, and CPR were the optimal feature set for distinguishing between normal and pathological sartorius muscles. The two feature sets were combined to discriminate between children with and without neuropathic muscles affected by Pompe disease, achieving an accuracy of 94.6%, a specificity of 100%, a sensitivity of 93.2%, and an area under the receiver operating characteristic curve of 0.98 ± 0.02. The CPR for the rectus femoris muscles and the AUT, ENT, MAXP, and VAR for the sartorius muscles exhibited statistically significant differences in distinguishing between the infantile-onset Pompe disease and late-onset Pompe disease groups (p < 0.05). Texture-feature parametric imaging can be used to quantify and map tissue structures in skeletal muscles and distinguish between pathological and normal muscles in children or newborns.

scholarly journals Quantitative EEG predicts outcomes in children after cardiac arrest

Neurology ◽  
2019 ◽  
Vol 92 (20) ◽  
pp. e2329-e2338 ◽  
Author(s):  
Seungha Lee ◽  
Xuelong Zhao ◽  
Kathryn A. Davis ◽  
Alexis A. Topjian ◽  
Brian Litt ◽  
...  
Keyword(s):  
Cardiac Arrest ◽  
Predictive Value ◽  
Cross Validation ◽  
Characteristic Curve ◽  
Cerebral Performance Category ◽  
Pediatric Intensive Care ◽  
Quantitative Eeg ◽  
Line Length ◽  
Neurologic Outcomes ◽  
Fold Cross Validation

ObjectiveTo determine whether quantitative EEG (QEEG) features predict neurologic outcomes in children after cardiac arrest.MethodsWe performed a single-center prospective observational study of 87 consecutive children resuscitated and admitted to the pediatric intensive care unit after cardiac arrest. Full-array conventional EEG data were obtained as part of clinical management. We computed 8 QEEG features from 5-minute epochs every hour after return of circulation. We developed predictive models utilizing random forest classifiers trained on patient age and 8 QEEG features to predict outcome. The features included SD of each EEG channel, normalized band power in alpha, beta, theta, delta, and gamma wave frequencies, line length, and regularity function scores. We measured outcomes using Pediatric Cerebral Performance Category (PCPC) scores. We evaluated the models using 5-fold cross-validation and 1,000 bootstrap samples.ResultsThe best performing model had a 5-fold cross-validation accuracy of 0.8 (0.88 area under the receiver operating characteristic curve). It had a positive predictive value of 0.79 and a sensitivity of 0.84 in predicting patients with favorable outcomes (PCPC score of 1–3). It had a negative predictive value of 0.8 and a specificity of 0.75 in predicting patients with unfavorable outcomes (PCPC score of 4–6). The model also identified the relative importance of each feature. Analyses using only frontal electrodes did not differ in prediction performance compared to analyses using all electrodes.ConclusionsQEEG features can standardize EEG interpretation and predict neurologic outcomes in children after cardiac arrest.

scholarly journals Reference values of fecal calgranulin C (S100A12) in school aged children and adolescents

2017 ◽  
Vol 56 (1) ◽  
Author(s):  
Anke Heida ◽  
Anneke C. Muller Kobold ◽  
Lucie Wagenmakers ◽  
Koos van de Belt ◽  
Patrick F. van Rheenen
Keyword(s):  
Characteristic Curve ◽  
Reference Value ◽  
Reference Interval ◽  
Enzyme Linked Immunosorbent Assay ◽  
Healthy Children ◽  
Healthy Controls ◽  
Reference Limit ◽  
School Aged Children ◽  
Activated Neutrophils ◽  
Upper Reference Limit

AbstractBackground:Calgranulin C (S100A12) is an emerging marker of inflammation. It is exclusively released by activated neutrophils which makes this marker potentially more specific for inflammatory bowel disease (IBD) compared to established stool markers including calprotectin and lactoferrin. We aimed to establish a reference value for S100A12 in healthy children and investigated whether S100A12 levels can discriminate children with IBD from healthy controls.Methods:In a prospective community-based reference interval study we collected 122 stool samples from healthy children aged 5–19 years. Additionally, feces samples of 41 children with suspected IBD (who were later confirmed by endoscopy to have IBD) were collected. Levels of S100A12 were measured with a sandwich enzyme-linked immunosorbent assay (ELISA) (InflamarkResults:The upper reference limit in healthy children was 0.75 μg/g (90% confidence interval: 0.30–1.40). Median S100A12 levels were significantly higher in patients with IBD (8.00 μg/g [interquartile range (IQR) 2.5–11.6] compared to healthy controls [0.22 μg/g (IQR<0.22); p<0.001]). The best cutoff point based on receiver operating characteristic curve was 0.33 μg/g (sensitivity 93%; specificity 97%).Conclusions:Children and teenagers with newly diagnosed IBD have significantly higher S100A12 results compared to healthy individuals. We demonstrate that fecal S100A12 shows diagnostic promise under ideal testing conditions. Future studies need to address whether S100A12 can discriminate children with IBD from non-organic disease in a prospective cohort with chronic gastrointestinal complaints, and how S100A12 performs in comparison with established stool markers.

Sign in / Sign up

Export Citation Format

Share Document