Guest Seminar - Next-Generation Optical and Photoacoustic Imaging Devices Empowered by Deep Learning

There are pressing need to provide better biomedical imaging devices for medical professionals to improve the current standard of care for patients. For instance, rapid and slide-free tissue imaging with histological contrast and minimal tissue preparation has long been a challenging yet appealing medical pursuit. In the first part of the talk, I will review a promising and transformative histological imaging method recently developed in my laboratory, termed computational high-throughput autofluorescence microscopy by pattern illumination (CHAMP), which can provide rapid and label-free imaging of thick and unprocessed tissues with large surface irregularity at an acquisition speed of 10 mm2/10 s with 1.1-µm lateral resolution. CHAMP images can be subsequently transformed into virtually stained histological images (Deep-CHAMP) through unsupervised learning in a minute, ensuring that medical professionals can understand our images immediately without additional training.

In the second part of the talk, I will share the new research directions related to CHAMP, including (1) the incorporation of a sectioning vibratome for three-dimensional whole-organ imaging with subcellular resolution in histopathology color tone, (2) the multi-color virtual staining with weakly supervised learning, and (3) the assistive medical diagnosis with artificial intelligence.

In the third part of the talk, I will share our recent research achievements in photoacoustic imaging with the co-development of the imaging system, contrast agent, and reconstruction algorithm. I will end the talk by sharing my view on the future of biomedical imaging devices in medicine.