A metabolomics-based classification system as a screening tool for EC will be clinically validated.
There were two cohorts in this diagnostic trial. To train multiple classification models, a multicenter prospective cohort of 50 cases (postmenopausal women with EC; International Federation of Gynecology and Obstetrics stages I-III and grades G1-G3) and 70 controls (no EC but matched on age, years from menopause, tobacco use, and comorbidities) was used. Each trained model’s accuracy was then utilized as a statistical weight to create an ensemble machine learning method for testing, which was verified using a second prospective cohort of 1430 postmenopausal women. The research was carried out at the University Hospital of San Giovanni di Dio e Ruggi d’Aragona in Salerno (Italy) and the Lega Italiana per la Lotta contro I Tumori clinic in Avellino (Italy). The data was collected between January 2018 and February 2019, and the analysis was placed between January and March 2019.
Based on blood metabolome analysis, the presence or absence of EC. All participants’ metabolites were collected and evaluated using gas chromatography-mass spectrometry from dried blood samples. Test findings were summarized using a confusion matrix. Sensitivity, specificity, positive and negative predictive values, positive and negative probability ratios, and accuracy were among the performance measures considered. In women who had a positive test result, hysteroscopy was used to confirm or rule out EC. Participants who had negative findings were followed up one year later to look into the emergence of EC symptoms.
Reference:jamanetwork.com/journals/jamanetworkopen/fullarticle/2770880?resultClick=1
