Medica l Image Ana lysis & Ar t i f icia l Intel l igence Symposium 2022 Nr. 21: Lorenz Perschy: Prediction of breast cancer in a high-risk patient cohort using deep learning in MR imaging data Curriculum Vitae: Lorenz Perschy is a completing a Masters in Bioinformatics at the University of Vienna and simultaneously studying Medicine at the Medical University of Vienna. He is currently working on his Master Thesis in the group of Georg Langs at the Computational Imaging Research Lab. The aim of his project is to improve the prediction of breast cancer in high-risk patients by employing deep learning methods. Conducting research at the intersection of medicine and informatics is very rewarding to me as I know my efforts have the potential to improve the life of many patients. Abstract: Prediction of breast cancer in a high-risk patient cohort using deep learning in MR imaging data Breast cancer contributed to approximately 685,000 deaths worldwide in 2020 [1]. Early detection is crucial for effective treatment and patient survival. In patients at an elevated risk of developing breast cancer due to family history or known mutations (e.g. BRCA1, BRCA2, p53) magnetic resonance imaging (MRI) is used for screening in regular check-ups [2]. However, manual diagnosis of MRI scans takes years of radiological experience, is time consuming and sometimes misses lesions. While most automated machine learning methods focus on the detection/classification of lesions at a specific timepoint [3], this project aims to develop an imaging based deep learning method to predict the likelihood of new lesions forming until the next medical examination. A dataset of 1,000 high risk patients collected over 20 years at the AKH Wien will be used to uncover patterns associated with early onset of breast cancer, potentially allowing radiologists to detect lesions earlier and thereby to improve treatment outcome.
RkJQdWJsaXNoZXIy NTA3MTM=