Artificial Intelligence and amniotic fluid multi-omics analysis: The prediction of perinatal outcome in the asymptomatic short cervix

Buket Aydas

Abstract

 The Amniotic Fluid (AF) had been gotten from second trimester patients with asymptomatic short CL (<15 mm). CL, channeling and the nearness of AF 'slop' were evaluated in all cases. Consolidated focused on metabolomic and proteomic examination of AF was performed. A blend of fluid Chromatography-Mass spectrometry (LC-MS-MS) and proton Nuclear Mass Spectrometry (1H-NMR) based metabolomics and focused on proteomics examination Bioplex Human cytokine Group-1 measure (Bio-Rad) comprising of chemokines, cytokines and development factors, were performed on the AF tests. To decide the power of the markers we utilized various AI methods including profound learning (DL) to foresee moderate rashness, <34 weeks, dormancy period before conveyance, and NICU remains. We utilized standard strategic relapse moreover. Omics biomarkers were assessed alone and in the mix with standard sonographic, clinical, and segment components to anticipate results. Prescient exactness was determined utilizing region under the beneficiary working attributes bend (AUC) and 95% CI, affectability, and explicitness esteem.