Completed
Model and learning strategies for Computational 3D phase microscopy
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Computational Microscopy Using LED Array for 3D Phase Imaging - Prof. Lei Tian
Automatically move to the next video in the Classroom when playback concludes
- 1 Computational microscopy using an LED array
- 2 LED array microscope for multimodal computational imaging
- 3 Intensity diffraction tomography with controllable LED illumination
- 4 Linear single-scattering approximation model Sample
- 5 Single scattering information is directly visible in the Fourier space
- 6 3D reconstruction by slice-wise deconvolution
- 7 Scan-free 3D phase reconstruction on cell cluster
- 8 Expanding IDT's imaging limits
- 9 Multiplexed Intensity Diffraction Tomography (MIDT) for high volume-rate IDT
- 10 Multi-scale phase tomography
- 11 Annual IDT (alDT) further extends imaging limits
- 12 Real-time volumetric phase imaging
- 13 "Open" computational 3D phase microscopy
- 14 Single-scattering model limits IDT's reconstructio
- 15 Multiple-scattering model to improve accuracy
- 16 Nonlinear SSNP multiple-scattering model
- 17 Reconstruction by optimization
- 18 Reconstruction on multiple-scattering sample
- 19 Reconstruction on dynamic sample
- 20 Deep learning training leverages single-scattering! inversion and learns multiple-scattering physics
- 21 Deep learning training leverages single-scattering inversion and learns multiple-scattering physics
- 22 Network improves weak scattering recovery
- 23 Network generalizes to different optical setups
- 24 Network generalizes to dynamic multiple-scattering sami
- 25 Acknowledgements
- 26 Model and learning strategies for Computational 3D phase microscopy
- 27 Lightweight 2D U-Net for efficient 3D Recovery
- 28 Multiple-scattering model-based reconstruction is still limited
- 29 Compute SSNP forward model stepwise
- 30 Model-based data normalization for generalization across different contrasts
- 31 Network prediction validation
- 32 Multiple-scattering physical simulator for generating training data Neural network trained entirely on simulated data generated using multiple scattering models
- 33 Trained deep learning model generalizes to biological samples
