Summary

Computer vision and image analysis are both great examples of successes with machine learning especially deep learning. While computer vision and image analysis primarily deal with existing images, tomographic reconstruction produces images of internal structures from externally measured indirect data. Recently, deep learning techniques are being actively explored for tomographic reconstruction by multiple groups worldwide, with encouraging results and potential major impacts. We believe that deep reconstruction is a next major target of deep learning. Sponsored by RPI Center for Biotechnology & Interdisciplinary Studies/Biomedical Imaging Center/NIH Training Program in Biomolecular Science and Engineering, we are honored to host this workshop for brainstorming and collaboration.

Agenda

Speaker Affiliation Title
Oct. 18
12:00–12:55pm Registration & Lunch
12:55–01:00pm Ge Wang RPI, BME Opening Remarks
01:00–01:30pm Bruno De Man GE GRC Raw Data Learning Opportunities in CT
01:30–02:00pm Quanzheng Li Harvard MGH Deep Learning Based Image Reconstruction with Noise Inputs
02:00–02:30pm Birsen Yazici RPI, ECSE Deep Learning for Computational Imaging
02:30–03:00pm Hengyong Yu Univ. of Mass. Lowell CNN-based Metal Artifact Reduction in CT
03:00–03:30pm Pingkun Yan RPI, BME Hybrid Imaging via Deep Learning
03:30–04:00pm Ruoyang Yao RPI, BME Compressed Fluorescence Lifetime Imaging with Deep Learning
04:00–05:00pm Free Discussion
05:00–07:00pm Dinner
Oct. 19
09:00–09:30am Kerstin Hammernik NYU Medical Center Learning a Variational Network for Accelerated MRI
09:30–10:00am Ye Duan Univ. of Missouri Synthesis of X-ray Projections via Deep Learning
10:00–10:30am Enhao Gong Stanford Deep GAN for Compressed Sensing MRI
10:30–11:00am Qingsong Yang RPI, BME GAN-based Image Denoising with Perceptual Measurement
11:00–11:30am Wenxiang Cong RPI, BME Beam Hardening Correction with Deep Learning
11:30–12:00pm Ruibin Feng RPI, BME Machine Learning for Photon-counting CT
12:00–01:00pm Lunch
01:00–01:30pm Fenglei Fan RPI, BME Machine Learning with Quadratic Neurons
01:30–02:00pm Hongming Shan RPI, BME Transfer Learning for Low-dose CT Denoising
02:00–02:30pm Florian Knoll NYU Medical Center Clinical Evaluation of Machine Learning for MRI
02:30–03:00pm Lars Gjesteby RPI, BME Metal Artifact Reduction via Deep Learning
03:00–03:30pm Uwe Kruger RPI, BME Projection-based Chemometric and Deep Reconstruction
03:30–04:00pm He Yang Augusta Univ. Deep Learning for Dual-energy CT
04:00–04:10pm Ge Wang RPI, BME Close remarks
04:10–05:00pm Free Discussion

Survey