MTW seminar: “Deep learning for medical image analysis”
När: 01/11/2022 , 11:45 - 13:00
In this seminar Ida Häggström will talk about two of her ongoing projects with her colleagues from Memorial Sloan Kettering in New York. She has developed a deep learning model that automatically predicts whether a patient has avid lymphoma or not based on unannotated positron emission tomography (PET) images. In her other project she developed a model to prognosticate the risk of disease relapse after surgery in early-stage lung cancer patients, by analyzing pre-treatment PET images.
Ida will also talk about her new Chalmers-based projects with colleagues at Sahlgrenska that are just starting up. She is working on identifying outliers in computed tomography (CT) images from the large national SCAPIS cohort of random volunteers, and building a model for early detection of lung cancer in the same cohort. She will also analyze clinical lung cancer patients. In another project, Ida also analyzes CT images in elderly people to predict and prognosticate bone fractures.
Ida Häggström completed two Master’s degrees in Engineering Physics followed by Medical Physics, and proceeded with a PhD in Medical Physics at Umeå University, graduating in 2015. She then moved to Memorial Sloan Kettering Cancer Center in New York, USA for a postdoctoral fellowship, followed by working as a Research Associate and Senior Research Scientist. Ida returned to Sweden and Chalmers University of Technology in late 2021 and am currently an Associate Professor in the Computer Vision group at the department of Electrical Engineering, working with machine and deep learning methods for medical image analysis.
The seminar will be online and held in English.
DATE: November 1
TIME: 11:45 – 13:00
PLACE: Online, via Zoom
Chalmers – Scaniasalen, Chalmers kårhus