Variability in the Evaluation and Treatment of Healthy Neonates With Uncomplicated Conjunctivitis.

Purpose: To explore the current state of diagnosis and management of neonatal conjunctivitis.

Methods: Cosmos, an EHR-based, de-identified data set including more than 200 million patients, was used for this study. Neonates born between January 1, 2016 and December 31, 2022, discharged from the hospital by day 3 of life, and with an ambulatory visit within the first 4 weeks of life associated with a new diagnosis of neonatal conjunctivitis (SNOMED) or conjunctivitis (ICD-10 H10.

Comparison of Machine Learning Models to a Novel Score in the Identification of Patients at Low Risk for Diabetic Retinopathy.

Purpose: To develop an easily applicable predictor of patients at low risk for diabetic retinopathy (DR).

Design: An experimental study on the development and validation of machine learning models (MLMs) and a novel retinopathy risk score (RRS) to detect patients at low risk for DR.

Subjects: All individuals aged ≥18 years of age who participated in the telemedicine retinal screening initiative through Temple University Health Systems from October 1, 2016 through December 31, 2020.

Knowing Your Patient Population: Techniques to Capture Infants at High Risk for Physical Abuse in a Trauma Registry.

Background: Physical abuse is a major public health concern and a leading cause of morbidity and mortality in infants. Clinical decision tools derived from trauma registries can facilitate timely risk-stratification. The Trauma Quality Improvement Program (TQIP) database does not report age for children <1 year who are at highest risk for abuse.

Systematic review and meta-analysis of the etiology of heavy menstrual bleeding in 2,770 adolescent females.

Background: Adolescent heavy menstrual bleeding(HMB), menorrhagia or abnormal uterine bleeding commonly occur in adolescent women. The differential diagnosis can be challenging. The pneumonic: PALM-COEIN (polyp, adenomyosis, leiomyoma, malignancy and hyperplasia, coagulopathy, ovulatory dysfunction, endometrial, iatrogenic, and not yet classified), is commonly used but it does not stratify as to the likelihood of a disorder.

Derivation and Validation of a Machine Learning Algorithm for Predicting Venous Thromboembolism in Injured Children.

Background: Venous thromboembolism (VTE) causes significant morbidity in pediatric trauma patients. We applied machine learning algorithms to the Trauma Quality Improvement Program (TQIP) database to develop and validate a risk prediction model for VTE in injured children.

Methods: Patients ≤18 years were identified from TQIP (2017-2019, n = 383,814).