Deep Learning Model and its Application for the Diagnosis of Exudative Pharyngitis.

OBJECTIVES: Telemedicine is firmly established in the healthcare landscape of many countries. Acute respiratory infections are the most common reason for telemedicine consultations. A throat examination is important for diagnosing bacterial pharyngitis, but this is challenging for doctors during a telemedicine consultation. A solution could be for patients to upload images of their throat to a web application. This study aimed to develop a deep learning model for the automated diagnosis of exudative pharyngitis. Thereafter, the model will be deployed online. METHODS: We used 343 throat images (139 with exudative pharyngitis and 204 without pharyngitis) in the study. ImageDataGenerator was used to augment the training data. The convolutional neural network models of MobileNetV3, ResNet50, and EfficientNetB0 were implemented to train the dataset, with hyperparameter tuning. RESULTS: All three models were trained successfully; with successive epochs, the loss and training loss decreased, and accuracy and training accuracy increased. The EfficientNetB0 model achieved the highest accuracy (95.5%), compared to MobileNetV3 (82.1%) and ResNet50 (88.1%). The EfficientNetB0 model also achieved high precision (1.00), recall (0.89) and F1-score (0.94). CONCLUSIONS: We trained a deep learning model based on EfficientNetB0 that can diagnose exudative pharyngitis. Our model was able to achieve the highest accuracy, at 95.5%, out of all previous studies that used machine learning for the diagnosis of exudative pharyngitis. We have deployed the model on a web application that can be used to augment the doctor's diagnosis of exudative pharyngitis.

The relationship between sleep-disordered breathing, blood pressure, and urinary cortisol and catecholamines in children.

STUDY OBJECTIVES: Hypertension is a complication of obstructive sleep apnea (OSA) syndrome in adults. A correlation between OSA syndrome and elevated blood pressure (BP) is suggested in children, but its pathogenesis remains unclear. Our aim was to study the effects of sleep and sleep apnea on BP and sympathetic nervous system activation as measured by serum cortisol and urinary catecholamines. We hypothesized that children with OSA syndrome would have higher BP, urinary catecholamines, and cortisol compared with controls. METHODS: We measured BP during polysomnography in 78 children with suspected sleep-disordered breathing and 18 nonsnoring controls. BP was measured during wakefulness and every 30-60 minutes throughout the night. All participants had 24-hour urinary catecholamine and free cortisol collections 48 hours before polysomnography. RESULTS: BP varied with sleep stage; it was highest during wakefulness and N1 and lowest during non-rapid eye movement stage 3. Children classified as high apnea-hypopnea index (AHI) snorers (AHI >5 events/h) had a greater prevalence of systolic hypertension (57%) than low-AHI snorers (22%) and nonsnoring controls (22%; P = .04). The high-AHI snorers also had higher diastolic BP (P < .02) as well as blunted nocturnal diastolic BP changes during sleep (P = .02) compared with low-AHI snorers (AHI <5 events/h). Twenty-hour urinary free cortisol and 24-hour urinary catecholamines were not associated with BP. CONCLUSIONS: BP in children varies with sleep stage. OSA is associated with systolic hypertension, higher BP during rapid eye movement sleep, as well as elevation of diastolic BP and blunted BP changes with sleep.

A cross-sectional study on reference ranges of normal oral temperatures among students in Singapore.

INTRODUCTION: During the 2003 Severe Acute Respiratory Syndrome (SARS) outbreak, all schools in Singapore implemented twice-daily temperature monitoring for students to curtail the spread of the disease. Students were not allowed to attend school if their temperature readings were >37.8 degrees C for students < or =12 years old, or > or =37.5 degrees C for students >12 years old. These values had been arbitrarily determined with professional inputs. The aim of this study is to determine the reference ranges of normal oral temperatures of students in Singapore and recommend the cut-off values for febrile patients. This may be used in another similar outbreak of an infectious disease with fever. MATERIALS AND METHODS: Four co-ed primary schools and 4 co-ed secondary schools were selected for this study. Four thousand and two hundred primary 1 to secondary 3 students responded (96.8%) and participated in this cross-sectional study. The mean ages of the students in the lowest (primary 1) and highest educational levels (secondary 3) were 7.4 years old and 15.3 years old, respectively. Twelve oral temperature readings per student (i.e. measurements taken 4 times a day in 3 consecutive days) were collected. Forty-six thousand seven hundred and eighty-three (92.8%) out of 50,400 temperature readings were used for the analysis as missing data were excluded. A quantile regression model was applied to estimate reference ranges of normal oral temperatures for students with adjustment for potential confounding factors. RESULTS: The age-specific reference ranges of normal oral temperature from this study for students < or =12 years old and >12 years old were 35.7 degrees C to 37.7 degrees C and 35.6 degrees C to 37.4 degrees C, respectively. Temperatures of 37.8 degrees C and 37.5 degrees C are therefore recommended as the oral temperature cut-offs for those < or =12 years old and >12 years old, respectively. CONCLUSION: This study has provided empirical data on normal oral temperature cut-offs which could be used during temperature screening in schools.

Sleep disorders in children: the Singapore perspective.

This review article summarises the current available literature on sleep patterns and sleep problems in Singapore children. Co-sleeping is a culturally dependent practice and its prevalence in Singapore has been determined to be 73%. Co-sleeping is not associated with significant sleep problems in Singapore children. Snoring and habitual snoring occur in 28.1% and 6.0% of Singapore children, respectively. Habitual snoring in Singapore children was significantly associated with obesity, allergic rhinitis, atopic dermatitis, maternal smoking and breastfeeding. Atopy was the strongest risk factor for habitual snoring in Singapore, and the effect was cumulative. Children attending psychiatric services in Singapore may also have sleep disorders, the highest prevalence being in children with attention deficit hyperactivity disorder. The knowledge on childhood sleep disorders (including obstructive sleep apnoea) amongst the public, patients, parents and future doctors in Singapore are inadequate and there is an urgent need for increased education in this area given the importance of good sleep in children. There is also a need to change parental attitudes about sleep disorders and encourage early medical consultation.

Relationship between exhaled nitric oxide and atopy in Asian young adults.

OBJECTIVES: The relationship between exhaled nitric oxide and atopy is controversial. The aim of this study was to determine the relationship between exhaled nitric oxide (FE(NO)) and atopy in Asian young adults. METHODOLOGY: Subjects were assessed by: (i) the International Study of Asthma and Allergies in Childhood questionnaire to differentiate asthmatic from nonasthmatic and rhinitis from non-rhinitis subjects; (ii) skin prick testing to 10 allergens; and (iii) FE(NO) measurements performed online at a flow rate of 50 mL/s. RESULTS: Complete results were available for 84 subjects. FE(NO) values were highest in atopic asthmatics (n = 34; median FE(NO), 59.8 p.p.b.; interquartile range, 30.4-85.5 p.p.b), followed by atopic nonasthmatics (n = 34; median, 38.4 p.p.b.; range, 16.7-49.3 p.p.b), nonatopic asthmatics (n = 5; median, 19.1 p.p.b.; range, 17.9-33.4 p.p.b), and lowest in nonatopic nonasthmatics (n = 11; median, 15.7 p.p.b.; range, 11.5-21.7 p.p.b). FE(NO) values were significantly higher in atopic (n = 68; median, 44.7 p.p.b.; range, 27.3-75.2 p.p.b) compared to nonatopic subjects (n = 16; median, 17.0 p.p.b.; range, 11.7-23.8 p.p.b.; P < 0.0001), regardless of asthma and rhinitis status. FE(NO) levels correlated with the severity of atopy (wheal size) for both asthmatic (r = 0.44, P = 0.005) and nonasthmatic subjects (r = 0.48, P = 0.001). There was no significant difference in FE(NO) levels between nonatopic asthmatics and nonatopic nonasthmatic subjects (P = 0.25). CONCLUSIONS: Increased FE(NO) levels are more reflective of atopy rather than asthma, and increased nitric oxide production may be predominantly a feature of atopy in asthmatics.

Snoring and atopic disease: a strong association.

The prevalence and factors associated with snoring and habitual snoring in Asian children are largely unknown. Our objectives were to evaluate the prevalence and factors associated with snoring and habitual snoring in preschool and primary school children in Singapore. A self-response questionnaire on snoring was administered to parents of children aged 4-7 years in randomly selected preschools and primary schools in Singapore. The overall response rate was 91.3% (nt = 11,114). Snoring and habitual snoring were reported in 28.1% and in 6.0% of the children, respectively. On multivariate logistic regression analysis, snoring was significantly associated with male gender, race, atopy (asthma, allergic rhinitis, or atopic dermatitis), maternal atopy (allergic rhinitis or atopic dermatitis), maternal smoking, and breastfeeding. Habitual snoring was significantly associated with obesity (odds ratio (OR), 3.75; 95% confidence interval (CI), 1.67-8.42), allergic rhinitis (OR, 2.90; 95% CI, 2.06-4.08), atopic dermatitis (OR, 1.80; 95% CI, 1.28-2.54), maternal smoking (OR, 2.22; 95% CI, 1.09-4.53), and breastfeeding (OR, 1.49; 95% CI, 1.11-1.98). Atopy was the strongest risk factor for habitual snoring, and the effect was cumulative. The odds ratio of a child with all three atopic diseases (asthma, allergic rhinitis, and atopic dermatitis) to have habitual snoring was 7.45 (95% CI, 3.48-15.97). In conclusion, snoring and habitual snoring are common in Asian children. Atopy is strongly associated with snoring and habitual snoring. We suggest that children who are significantly atopic receive additional attention during screening for snoring, habitual snoring, and other features of obstructive sleep apnea syndrome.

Childhood severe acute respiratory syndrome, coronavirus infections and asthma.

Severe acute respiratory syndrome (SARS) is a new infectious disease caused by a novel coronavirus. Children appear to be less susceptible to the SARS coronavirus, although the other non-SARS coronaviruses can cause respiratory infections in adults and in children of all ages. The exact reasons as to why SARS preferentially affects adults, and not children, are still unknown. Many hypotheses exist and need to be explored. During the outbreak of SARS, there did not appear to be an increase in asthma exacerbations in children.