High Resolution, Deep Imaging Using Confocal Time-of-Flight Diffuse Optical Tomography

Imaging inside highly scattering media using hybrid deep learning and analytical algorithm

Imaging through highly scattering media is a challenging problem with numerous applications in biomedical and remote-sensing fields. Existing methods that use analytical or deep learning tools are limited by simplified forward models or a requirement for prior physical knowledge, resulting in blurry images or a need for large training databases. To address these limitations, we propose a hybrid scheme called Hybrid-DOT that combines analytically derived image estimates with a deep learning network. Our analysis demonstrates that Hybrid-DOT outperforms a state-of-the-art ToF-DOT algorithm by improving the PSNR ratio by 4.6 dB and reducing the resolution by a factor of 2.5. Furthermore, when compared to a deep learning stand-alone model, Hybrid-DOT achieves a 0.8 dB increase in PSNR, 1.5 times the resolution, and a significant reduction in the required dataset size (factor of 1.6-3). The proposed model remains effective at higher depths, providing similar improvements for up to 160 mean-free paths.

An Optical Sensory System for Assessment of Residual Cancer Burden in Breast Cancer Patients Undergoing Neoadjuvant Chemotherapy

Breast cancer patients undergoing neoadjuvant chemotherapy (NAC) require precise and accurate evaluation of treatment response. Residual cancer burden (RCB) is a prognostic tool widely used to estimate survival outcomes in breast cancer. In this study, we introduced a machine-learning-based optical biosensor called the Opti-scan probe to assess residual cancer burden in breast cancer patients undergoing NAC. The Opti-scan probe data were acquired from 15 patients (mean age: 61.8 years) before and after each cycle of NAC. Using regression analysis with k-fold cross-validation, we calculated the optical properties of healthy and unhealthy breast tissues. The ML predictive model was trained on the optical parameter values and breast cancer imaging features obtained from the Opti-scan probe data to calculate RCB values. The results show that the ML model achieved a high accuracy of 0.98 in predicting RCB number/class based on the changes in optical properties measured by the Opti-scan probe. These findings suggest that our ML-based Opti-scan probe has considerable potential as a valuable tool for the assessment of breast cancer response after NAC and to guide treatment decisions. Therefore, it could be a promising, non-invasive, and accurate method for monitoring breast cancer patient's response to NAC.

High-Speed Time-Domain Diffuse Optical Tomography with a Sensitivity Equation-based Neural Network

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.

Single photon single pixel imaging into thick scattering medium

Imaging into thick scattering medium is a long-standing challenge. Beyond the quasi-ballistic regime, multiple scattering scrambles the spatiotemporal information of incident/emitted light, making canonical imaging based on light focusing nearly impossible. Diffusion optical tomography (DOT) is one of the most popular approach to look inside scattering medium, but quantitatively inverting the diffusion equation is ill-posed, and prior information of the medium is typically necessary, which is nontrivial to obtain. Here, we show theoretically and experimentally that, by synergizing the one-way light scattering characteristic of single pixel imaging with ultrasensitive single photon detection and a metric-guided image reconstruction, single photon single pixel imaging can serve as a simple and powerful alternative to DOT for imaging into thick scattering medium without prior knowledge or inverting the diffusion equation. We demonstrated an image resolution of 12 mm inside a 60 mm thick (∼ 78 mean free paths) scattering medium.

Unrolled-DOT: an interpretable deep network for diffuse optical tomography

SIGNIFICANCE

Imaging through scattering media is critical in many biomedical imaging applications, such as breast tumor detection and functional neuroimaging. Time-of-flight diffuse optical tomography (ToF-DOT) is one of the most promising methods for high-resolution imaging through scattering media. ToF-DOT and many traditional DOT methods require an image reconstruction algorithm. Unfortunately, this algorithm often requires long computational runtimes and may produce lower quality reconstructions in the presence of model mismatch or improper hyperparameter tuning.

AIM

We used a data-driven unrolled network as our ToF-DOT inverse solver. The unrolled network is faster than traditional inverse solvers and achieves higher reconstruction quality by accounting for model mismatch.

APPROACH

Our model "Unrolled-DOT" uses the learned iterative shrinkage thresholding algorithm. In addition, we incorporate a refinement U-Net and Visual Geometry Group (VGG) perceptual loss to further increase the reconstruction quality. We trained and tested our model on simulated and real-world data and benchmarked against physics-based and learning-based inverse solvers.

RESULTS

In experiments on real-world data, Unrolled-DOT outperformed learning-based algorithms and achieved over 10× reduction in runtime and mean-squared error, compared to traditional physics-based solvers.

CONCLUSION

We demonstrated a learning-based ToF-DOT inverse solver that achieves state-of-the-art performance in speed and reconstruction quality, which can aid in future applications for noninvasive biomedical imaging.

Algorithms and instrumentation for rapid spatial frequency domain fluorescence diffuse optical imaging

SIGNIFICANCE

Rapid estimation of the depth and margins of fluorescence targets buried below the tissue surface could improve upon current image-guided surgery techniques for tumor resection.

AIM

We describe algorithms and instrumentation that permit rapid estimation of the depth and transverse margins of fluorescence target(s) in turbid media; the work aims to introduce, experimentally demonstrate, and characterize the methodology.

APPROACH

Spatial frequency domain fluorescence diffuse optical tomography (SFD-FDOT) technique is adapted for rapid and computationally inexpensive estimation of fluorophore target depth and lateral margins. The algorithm utilizes the variation of diffuse fluorescence intensity with respect to spatial-modulation-frequency to compute target depth. The lateral margins are determined via analytical inversion of the data using depth information obtained from the first step. We characterize method performance using fluorescent contrast targets embedded in tissue-simulating phantoms.

RESULTS

Single and multiple targets with significant lateral size were imaged at varying depths as deep as 1 cm. Phantom data analysis showed good depth-sensitivity, and the reconstructed transverse margins were mostly within ∼30 % error from true margins.

CONCLUSIONS

The study suggests that the rapid SFD-FDOT approach could be useful in resection surgery and, more broadly, as a first step in more rigorous SFD-FDOT reconstructions. The experiments permit evaluation of current limitations.

Sub-mm resolution tomographic imaging in turbid media by an ultra-high density multichannel approach

We demonstrate an ultra-high-density source-detector (SD) diffuse optical tomography system scalable to thousands of combinatorial SD pairs per cm3 of total voxel volume. We demonstrate the imaging of dynamic targets (including phantom arteries) with 100 um resolution at over 10 Hz frame rate within turbid media (> 60 MFP). Further, as a step toward a wearable mobile imager, we introduce monolithic mm-size dense semiconductor laser array chips as sources for potential unobtrusive epidermal tomographic use.

Advanced high resolution three-dimensional imaging to visualize the cerebral neurovascular network in stroke

As an important method to accurately and timely diagnose stroke and study physiological characteristics and pathological mechanism in it, imaging technology has gone through more than a century of iteration. The interaction of cells densely packed in the brain is three-dimensional (3D), but the flat images brought by traditional visualization methods show only a few cells and ignore connections outside the slices. The increased resolution allows for a more microscopic and underlying view. Today's intuitive 3D imagings of micron or even nanometer scale are showing its essentiality in stroke. In recent years, 3D imaging technology has gained rapid development. With the overhaul of imaging mediums and the innovation of imaging mode, the resolution has been significantly improved, endowing researchers with the capability of holistic observation of a large volume, real-time monitoring of tiny voxels, and quantitative measurement of spatial parameters. In this review, we will summarize the current methods of high-resolution 3D imaging applied in stroke.

Tutorial on the Use of Deep Learning in Diffuse Optical Tomography

Diffuse optical tomography using deep learning is an emerging technology that has found impressive medical diagnostic applications. However, creating an optical imaging system that uses visible and near-infrared (NIR) light is not straightforward due to photon absorption and multi-scattering by tissues. The high distortion levels caused due to these effects make the image reconstruction incredibly challenging. To overcome these challenges, various techniques have been proposed in the past, with varying success. One of the most successful techniques is the application of deep learning algorithms in diffuse optical tomography. This article discusses the current state-of-the-art diffuse optical tomography systems and comprehensively reviews the deep learning algorithms used in image reconstruction. This article attempts to provide researchers with the necessary background and tools to implement deep learning methods to solve diffuse optical tomography.