Deep learning in macroscopic diffuse optical imaging
Problem
Analyzing macroscopic diffuse optical images (optical property retrieval, fluorescence lifetime imaging, tomography) with conventional inverse solvers is slow and limited in reconstruction quality.
Why it matters
Faster, more accurate noninvasive optical imaging analysis could improve biomedical diagnostics and research.
What the AI did
The paper reviews deep learning methods applied across diffuse optical imaging tasks, replacing or augmenting conventional inverse solvers for image formation and analysis.
Results
Literature reviewed reports decreased analysis time (often by many orders of magnitude) and increased quantitative reconstruction quality versus conventional inverse solvers.
Impact
Peer-reviewed review documenting substantial speed and quality gains from deep learning in diffuse optical imaging.
Record
| Venue | Journal of Biomedical Optics |
|---|---|
| Published | 2022 |
| Work type | article |
| Authors | Jason T. Smith, Marien Ochoa, Denzel E. Faulkner, Grant Haskins, Xavier Intes |
| Citations | 45 (OpenAlex, at import time) |
| References | 180 |
| Model or method | not stated in source |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1117/1.jbo.27.2.020901 |
| Open access | gold |
| Keywords | deep learning, diffuse optical imaging, diffuse optical tomography, fluorescence lifetime, reconstruction |
Importance rating
real but narrower advance
Review claims DL brings diffuse optical imaging modalities toward clinical translation with orders-of-magnitude speed gains; potential bedside impact stated but no patient scale quantified.