Towards real-time diffuse optical tomography with a handheld scanning probe

Diffuse optical tomography (DOT) performed using deep-learning allows high-speed reconstruction of tissue optical properties, and could thereby enable image-guided scanning, e.g. to enhance clinical breast imaging. Previously published models are geometry specific, and therefore require extensive data generation and training for each use case, and restrict the scanning protocol at the point of use. To overcome these obstacles, a novel transformer-based architecture is proposed to encode spatially unstructured DOT measurements, enabling a single trained model to handle arbitrary scanning pathways and measurement density. The model is demonstrated with breast tissue-emulating simulated and phantom data, yielding - for 24 mm-deep absorption (μa) and reduced scattering (μs') images respectively - average RMSEs of 0.0095±0.0023 cm¯¹ and 1.95±0.78 cm¯¹, Sørensen-Dice coefficients of 0.55±0.12 and 0.67±0.1, and anomaly contrast of 79±10\% and 93.3±4.6% of the ground-truth contrast, with an effective imaging speed of 14 Hz. The average absolute μa and μs' values of homogeneous simulated examples were within 10% of the true values.