With advancements in Deep Learning, computer vision has transformed healthcare from diagnosis to prognosis and treatment to prevention. Its far-reaching applications include data synthesis, surgical assistants, patient monitoring, and cancer screening. Nevertheless, before these algorithms make their way into real-world medical imaging scenarios, there is exciting research to develop accurate, robust, interpretable, grounded, and human-centered approaches. For this reason, the course aims to teach introductory computer vision topics focusing on biomedical applications. The focus of this course will be developing knowledge in biomedical imaging and fundamental computational skills, including data acquisition, formats, filtering, segmentation, feature extraction, machine learning-based image analysis, and deep introductory learning for computer vision (i.e., convolutional neural networks for classification). Frontal lessons to introduce theoretical concepts and ad-hoc hands-on sessions to apply the learned methods, all of which will be in Python, will be carried out during the course. In the end, students will take part in an internal challenge to apply the computer vision strategies learned to a novel research problem on real-world medical image data. Upon completion of the course, students will have the fluency necessary to dive into medical computer vision research literature and be able to engineer and develop solutions for different biomedical imaging tasks.

dal October 2022 a November 2022