DPU - Symposium 2022

Medica l Image Ana lysis & Ar t i f icia l Intel l igence Symposium 2022 Nr. 5: Ass.-Prof. Dr. Sepideh Hatamikia &Dr. Geevarghese George: Feature selection strategies for improved predictions of histopathologic response to neoadjuvant chemotherapy in patients with ovarian cancer Curriculum Vitae: Dr. Geevarghese George obtained his PhD in Physics from the University of Strasbourg (France) for his work at the Institut Charles Sadron (Theory and Simulation of Polymers group, CNRS), supervised by Dr. Joachim Wittmer. His thesis was on the numerical study of statistical and rheological properties of polymer films. During his PhD, he also developed an interest in machine learning and pivoted into the field through self-directed projects and coursework. His research interest at MIAAI (DPU) is in the area of machine/deep learning for medical image based diagnosis/prognosis. Curriculum Vitae: Ass.-Prof. Dr. Sepideh Hatamikia completed her PhD and also a postdoctoral position in biomedical engineering field at Medical University of Vienna at the Department of Medical Physics and Biomedical Engineering at the Digital Image Processing Laboratory. Since march 2022, she joined Danube Private University (DPU) and currently she is the Head of Computational Imaging in the Medical Image Analysis and Artificial Intelligence (MIAAI) research group. In addition, Ass.-Prof. Dr. Hatamikia is a researcher at Austrian Center for Medical Innovation and Technology (ACMIT) since January 2018 and continues collaboration with this research center on different research projects. Abstract: Feature selection strategies for improved predictions of histopathologic response to neoadjuvant chemotherapy in patients with ovarian cancer Abstract: Response to neoadjuvant chemotherapy (NACT) in patients with high grade serous ovarian carcinoma (HGSOC) is evaluated histopathologically using the chemotherapy response score (CRS) for omental deposits. Recent radiomics-based models have shown promise in predicting CRS. This study investigates how different feature selection strategies improve CRS prediction. Our results demonstrated that feature selection methods can improve the prediction of response to NACT based on CRS in terms of accuracy, sensitivity and specificity.

RkJQdWJsaXNoZXIy NTA3MTM=