Deep learning in macroscopic diffuse optical imaging

Abstract

. Significance : Biomedical optics system design, image formation, and image analysis have primarily been guided by classical physical modeling and signal processing methodologies. Recently, however, deep learning (DL) has become a major paradigm in computational modeling and has demonstrated utility in numerous scientific domains and various forms of data analysis. Aim : We aim to comprehensively review the use of DL applied to macroscopic diffuse optical imaging (DOI). Approach : First, we provide a layman introduction to DL. Then, the review summarizes current DL work in some of the most active areas of this field, including optical properties retrieval, fluorescence lifetime imaging, and diffuse optical tomography. Results : The advantages of using DL for DOI versus conventional inverse solvers cited in the literature reviewed herein are numerous. These include, among others, a decrease in analysis time (often by many orders of magnitude), increased quantitative reconstruction quality, robustness to noise, and the unique capability to learn complex end-to-end relationships. Conclusions : The heavily validated capability of DL’s use across a wide range of complex inverse solving methodologies has enormous potential to bring novel DOI modalities, otherwise deemed impractical for clinical translation, to the patient’s bedside.

Optical imaging and spectroscopy for the study of the human brain: status report

“…Nonetheless, as we explore novel approaches in Machine Learning, we can expect that DOT will soon become the norm by providing quantitative functional data. 201 …”

Deep learning in fNIRS: a review

“…Hence, there is still not a consensus on which architecture and hyperparameters are optimal for a specific problem and building on current work requires some level of technical expertise to assess which implementation would be optimal. This may be circumvented in the future with the advent of network automated designed via neural architecture search methods, 104 but these have not been yet applied to the field of fNIRS. Moreover, following the principles of the no-free-lunch theorem, it is expected that prior knowledge on the problem at hand should guide in the selection of the ML/DL algorithm.…”

Periodic-net: an end-to-end data driven framework for diffuse optical imaging of breast cancer from noisy boundary data

“…Various deep-learning approaches for numerous scientific domains and various forms of data analysis have been extensively reviewed for current work in optical properties retrieval. 18 , 19 Well-controlled data sets for training and validation are among the most important topics in the neural network, but the lack of large, publicly available data sets leads to unique challenges. The development of a data generation pipeline 20 based on Monte Carlo modeling has shown to be a useful method for rapid, robust, and user-friendly image formation in a wide variety of applications.…”