12:30 - 12:50 - Naomi Duits, PhD candidate, Biophotonics & Medical Imaging
Titel: Visualization of Cell and Tissue Dynamics in Human Lung Tissue Using Higher Harmonic Generation Microscopy
Abstract: Treatment selection of patients with lung cancer and interstitial lung disease is challenging due to variety in treatment response. Current selection approaches are insufficient and underlying disease mechanisms are not fully understood, which leads to over- and undertreatment of patients. In our research, we aim to develop a biopsy-based drug testbed based on timelapse imaging using higher harmonic generation microscopy. Higher harmonic generation microscopy is a label-free and non-damaging imaging technique capable of visualising relevant tissue structures ((immune) cells, elastin and collagen elastin fibers). This has enabled us to study dynamic tissue features in cultured human lung tissue through 3D timelapse (3D+t) imaging. In our experiments, we use lung tissue containing normal, ILD and tumor tissue. During tissue culture, we can visualize dynamic tissue metrics such as (immune) cell motion, cellular interactions and changes in tissue morphology. We expect that these dynamic features are predictive of treatment response and that our testbed facilitates testing of different treatment options to prevent over- and undertreatment of patients.
12:50 -13:45 - Dr. Nachi Stern, group leader of the Learning Machines group, AMOLF
Learning without neurons in physical systems
Abstract:
From electrically responsive neuronal networks to immune repertoires, biological systems can learn to perform complex tasks. In this talk, we explore physical learning, a framework inspired by computational learning theory and biological systems, where networks physically adapt to applied forces to adopt desired functions. Unlike traditional engineering approaches or artificial intelligence, physical learning is facilitated by physically realizable learning rules, requiring only local responses and no explicit information about the desired functionality. Our research shows that such local learning rules can be derived for broad classes of physical networks and that physical learning is indeed physically realizable, without computer aid, through laboratory experiments. We take further inspiration from learning in the brain to demonstrate the success of physical learning beyond the quasi-equilibrium regime, leading to faster learning with little penalty. By leveraging the advances of statistical learning theory in physical machines, we propose physical learning as a promising bridge between computational machine learning and biology, with the potential to enable the development of new classes of smart metamaterials that adapt in-situ to users’ needs.