2024/06/24 Computational fiber imaging using digital holography and deep learning for biomedicine and communication

Speaker

Prof. Juergen Czarske, Dresden University of Technology

Time

2024/06/24 1:30PM

Location

Room 1-418, SEIEE Building

Host

Dr. Yuye Ling, Associate Professor, Department of Electronic Engineering

Abstract

Light has the potential to recognize the development of diseases, to prevent them or to heal them early and gently. Traditionally, however, lens-based imaging results in bulky systems. We outline the 3D imaging with needle-sized, lensless multicore fiber endoscopes using holography. Deep learning is promising for label-free cancer diagnostics in neurosurgery using minimally invasive endoscopy. Besides biomedicine data transmission with fiber networks will be high lighted. Multimode fibers show promise for improving data rates in optical communications but come with the challenge of compensating for scattering effects. Physics-informed deep learning enables to correct the scattering, resulting in advancing security and data rate. Fiber imaging promises advances in various fields such as information, biomedicine and environmental protection.