Binary dose level classification of tumour microvascular response to radiotherapy using artificial intelligence analysis of optical coherence tomography images

Abstract

The dominant consequence of irradiating biological systems is cellular damage, yet microvascular damage begins to assume an increasingly important role as the radiation dose levels increase. This is currently becoming more relevant in radiation medicine with its pivot towards higher-dose-per-fraction/fewer fractions treatment paradigm (e.g., stereotactic body radiotherapy (SBRT)). We have thus developed a 3D preclinical imaging platform based on speckle-variance optical coherence tomography (svOCT) for longitudinal monitoring of tumour microvascular radiation responses in vivo. Here we present an artificial intelligence (AI) approach to analyze the resultant microvascular data. In this initial study, we show that AI can successfully classify SBRT-relevant clinical radiation dose levels at multiple timepoints (t = 2–4 weeks) following irradiation (10 Gy and 30 Gy cohorts) based on induced changes in the detected microvascular networks. Practicality of the obtained results, challenges associated with modest number of animals, their successful mitigation via augmented data approaches, and advantages of using 3D deep learning methodologies, are discussed. Extension of this encouraging initial study to longitudinal AI-based time-series analysis for treatment outcome predictions at finer dose level gradations is envisioned.

Low-Cost 3D-Printed Tools Towards Robust Longitudinal Multi-Modal Pre-Clinical Imaging

“…These automatic micro-stages remove the problem of having to figure out the mouse positioning for each imaging timepoint, saving valuable time and removing imaging variability in these logistically complex experiments (see Figure 5 ). This also facilitates to first-order the inter-timepoint co-registration required for time series analyses 5,7,9 and inter-modality co-registration from intravital optical imaging to ex-vivo histology (see Suppl. 4 ).…”

“…In the presented work we demonstrate a novel DSWC design and imaging toolset enabling more robust 1) longitudinal, and 2) multi-modal pre-clinical imaging experiments. This pipeline has far-ranging potential applications for longitudinal studies in the biomedical sciences [1][2][3][4][5][6][7][8][9][10][11][12][13] . One such application is shedding light on the response kinetics of tumour xenografts to various therapeutics (e.g., chemotherapies, radiotherapies, etc.).…”

“…This pipeline holds potential to reduce variability inter-timepoint and inter-specimen in notoriously complex in-vivo experiments. Beyond this, through improvements to multi-modal data acquisition and subsequent co-registration robustness, this pipeline may further facilitate studies correlating findings between pre-clinical modalities with clinically feasible analogues 7,9,49 . This simplification of translational studies may thus accelerate the transition from benchtop to bedside.…”

“…First developed in the 1930s, surgically implantable window chambers (WCs) enable high-resolution optical interrogation of tissue microstructure and its vasculature, with wideranging applications across the fields of biophysics, biomaterials, and biomedical engineering and biomedicine [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16] . In particular, WC models have been utilized to shed light on the tumour microenvironment (TME), such as the microvasculature which is hypothesized to play a major role in pathology development and tumour therapeutic response [4][5][6][7][8][9][10] . They consist of a surgical opening into the tissues of an organism (most commonly small rodents such as mice or rats), sealed by a biocompatible frame (typically made of titanium) securing a glass window over the target organ to be imaged 17 .…”

Historical Progress of Stereotactic Radiation Surgery

“…Some studies and vendors are trying to employ AI in contouring and dose calculation of TPSs. [ 61 - 63 ]…”