PMID- 37456517
OWN - NLM
STAT- PubMed-not-MEDLINE
LR  - 20240509
IS  - 2573-0436 (Print)
IS  - 2333-9403 (Electronic)
IS  - 2333-9403 (Linking)
VI  - 9
DP  - 2023
TI  - High-Speed Time-Domain Diffuse Optical Tomography with a Sensitivity 
      Equation-based Neural Network.
PG  - 459-474
LID - 10.1109/tci.2023.3273423 [doi]
AB  - Steady progress in time-domain diffuse optical tomography (TD-DOT) technology is 
      allowing for the first time the design of low-cost, compact, and high-performance 
      systems, thus promising more widespread clinical TD-DOT use, such as for 
      recording brain tissue hemodynamics. TD-DOT is known to provide more accurate 
      values of optical properties and physiological parameters compared to its 
      frequency-domain or steady-state counterparts. However, achieving high temporal 
      resolution is still difficult, as solving the inverse problem is computationally 
      demanding, leading to relatively long reconstruction times. The runtime is 
      further compromised by processes that involve 'nontrivial' empirical tuning of 
      reconstruction parameters, which increases complexity and inefficiency. To 
      address these challenges, we present a new reconstruction algorithm that combines 
      a deep-learning approach with our previously introduced 
      sensitivity-equation-based, non-iterative sparse optical reconstruction (SENSOR) 
      code. The new algorithm (called SENSOR-NET) unfolds the iterations of SENSOR into 
      a deep neural network. In this way, we achieve high-resolution sparse 
      reconstruction using only learned parameters, thus eliminating the need to tune 
      parameters prior to reconstruction empirically. Furthermore, once trained, the 
      reconstruction time is not dependent on the number of sources or wavelengths 
      used. We validate our method with numerical and experimental data and show that 
      accurate reconstructions with 1 mm spatial resolution can be obtained in under 20 
      milliseconds regardless of the number of sources used in the setup. This opens 
      the door for real-time brain monitoring and other high-speed DOT applications.
FAU - Wang, Fay
AU  - Wang F
AD  - Department of Biomedical Engineering, Columbia University, New York, NY 10027.
FAU - Kim, Stephen H
AU  - Kim SH
AD  - Department of Biomedical Engineering, New York University - Tandon School of 
      Engineering, New York, NY 10001.
FAU - Zhao, Yongyi
AU  - Zhao Y
AD  - Department of Electrical and Computer Engineering, Rice University, Houston, TX 
      77005.
FAU - Raghuram, Ankit
AU  - Raghuram A
AD  - Department of Electrical and Computer Engineering, Rice University, Houston, TX 
      77005.
FAU - Veeraraghavan, Ashok
AU  - Veeraraghavan A
AD  - Department of Electrical and Computer Engineering, Rice University, Houston, TX 
      77005.
FAU - Robinson, Jacob
AU  - Robinson J
AD  - Department of Electrical and Computer Engineering, Rice University, Houston, TX 
      77005.
FAU - Hielscher, Andreas H
AU  - Hielscher AH
AD  - Department of Biomedical Engineering, New York University - Tandon School of 
      Engineering, New York, NY 10001.
LA  - eng
GR  - T15 LM007093/LM/NLM NIH HHS/United States
PT  - Journal Article
DEP - 20230508
PL  - United States
TA  - IEEE Trans Comput Imaging
JT  - IEEE transactions on computational imaging
JID - 101660833
PMC - PMC10348778
MID - NIHMS1902768
OTO - NOTNLM
OT  - Deep learning
OT  - diffuse optics
OT  - image reconstruction
OT  - inverse problem
OT  - sensitivity equation
OT  - sparse image reconstruction
EDAT- 2023/07/17 06:42
MHDA- 2023/07/17 06:43
PMCR- 2024/05/08
CRDT- 2023/07/17 04:24
PHST- 2023/07/17 06:43 [medline]
PHST- 2023/07/17 06:42 [pubmed]
PHST- 2023/07/17 04:24 [entrez]
PHST- 2024/05/08 00:00 [pmc-release]
AID - 10.1109/tci.2023.3273423 [doi]
PST - ppublish
SO  - IEEE Trans Comput Imaging. 2023;9:459-474. doi: 10.1109/tci.2023.3273423. Epub 
      2023 May 8.