Quantitative magnetic resonance imaging and automated textural radiomics for assessment of human placenta accreta spectrum

点击翻译按钮获取中文摘要

PROJECT SUMMARY Placenta Accreta Spectrum (PAS) disorder is a condition in which the placenta fails to separate partially or totally from the uterine wall, resulting in significant intrapartum maternal morbidity and mortality at delivery. Depending on the extent of the placental invasion into surrounding tissue, hysterectomy may be lifesaving. Assessment of PAS severity prior to delivery is crucial for multidisciplinary surgical planning and patient counseling. However, there is an unmet clinical need for quantitative, objective measures of abnormal placentation that can serve as predictors of risk of invasion and hysterectomy. The overall goal of the proposed project is to prospectively assess PAS in high-risk women across gestation using quantitative multi-parametric magnetic resonance (MR) imaging and automated textural radiomics and to correlate findings with surgical outcomes. Coming from a highly interdisciplinary background of MR chemistry, biology, and physics, the candidate's long-term career goal is to become an independent investigator with in-depth knowledge and skills to conduct translational placental imaging research. The candidate's short-term goal is to complete a pilot study under the mentorship of an interdisciplinary advisory committee to evaluate PAS in women with high-risk pregnancies using multi-parametric MRI and textural radiomics. To meet these goals, the following research aims are proposed: 1) Evaluate placental invasion across gestation using multi-parametric MR imaging and deep learning based novel quantification tools in PAS high- risk pregnancies, and 2) Extract the radiomic pattern across gestation of high-risk PAS placenta and determine its association with clinical surgical outcome. These research aims will serve as the platform for the career development plan and training aims which include: 1) Gain knowledge in translational prospective research design, 2) Improve technical skills in advanced clinical MR acquisition development, 3) Acquire applied statistical and data analysis skills, and 4) Develop mentorship and leadership skills while under training. Together, the research and training aims will provide the training, experience and preliminary data for an R01 application. The future R01 will develop a clinical and surgical outcome prediction model for high-risk PAS pregnancies, potentially improving maternal and fetal outcomes. The training will position the candidate to become a leader in translational placental imaging research.