Oct 01

SS 46 - Radiation and Cancer Physics 10: Functional and Quantitative Imaging Method Development

369 - Toward Endogenous Tissue Dosimetry and Radiotherapy Response Monitoring: Quantifying Blood and Oxygenation Levels in Tissue via Multispectral Cherenkov Imaging

11:30am - 11:40am PT

Room 152

Presenter(s)

Roman Vasyltsiv, BS - Dartmouth College, Hanover, NH

R. Vasyltsiv¹, A. L. Matous², S. Wang³, D. J. Gladstone⁴, L. A. Jarvis⁵, and P. Bruza³; ¹Dartmouth College, Hanover, NH, ²Dartmouth-Hitchcock Medical Center, Lebanon, NH, ³Thayer School of Engineering, Dartmouth College, Hanover, NH, ⁴Department of Medicine, Geisel School of Medicine, Dartmouth College, Hanover, NH, ⁵Dartmouth Health, Lebanon, NH

Purpose/Objective(s):

Cherenkov imaging enables real-time qualitative dose visualization and incident detection but is limited in dosimetry by signal distortions caused by light transport in tissue. These distortions are influenced by tissue properties like blood and oxygenation, which affect visible light differently across the spectrum. We propose that spectrally segmented imaging can reveal wavelength-dependent responses, enabling signal intensity corrections. We present a multispectral imaging method that simultaneously captures four Cherenkov emission bands to analyze blood and oxygenation effects across wavelengths, linking physiological variables to spectrally resolved signals.

Materials/Methods:

A spectral imaging camera attachment was designed to split and filter Cherenkov emission into four bands (600–800 nm) captured on a single sensor. Blood concentration effects on spectral emission were studied using a 9-well blood dilution phantom (0–3%). To assess oxygenation impacts, a 1% blood phantom was deoxygenated over 15 minutes, with oxygen levels monitored by an electrode and imaged during periodic irradiation. In vivo spectral imaging was performed on a mantle cell lymphoma (MCL) treatment to correlate with deoxygenation phantom trends. Two breast cancer patients were monitored with spectral imaging to validate blood concentration effects observed in phantoms. Phantom and patient data were analyzed to establish relationships between physiological variables and Cherenkov emission.

Results:

Increase in blood content strongly affected the Cherenkov spectrum, with exponential decay constants of 600 nm and 800 nm bands differing by a factor of 5.4. Attenuation correction in a sample blood phantom achieved <3% deviation from expected intensity. Phantom oxygenation showed a strong correlation with spectral ratios (R² = 0.99), enabling oxygenation prediction with 9.8 mmHg uncertainty (95% CI). Surface tumor tissue achieved >5 SNR across all spectral bands, revealing a central region with characteristics of hypoxia - 26% and 1% drop in intensity for 600nm and 800nm, respectively, compared to healthy tissue. Spectral Cherenkov imaging also revealed areas of increased sub-surface blood concentration during breast radiotherapy, which were indistinguishable with monochromatic imaging.

Conclusion:

This study is the first to systematically examine how blood content and tissue oxygenation affect Cherenkov spectra, revealing relationships between these variables and spectral emission. The established correlations between blood concentration, oxygenation, and wavelength-dependent responses provide a foundation for addressing a key barrier to quantitative in vivo dosimetry, enabling future development of correction methods to produce more accurate surface dose maps during radiotherapy treatments.