An Explicit Estimated Baseline Model for Robust Estimation of Fluorophores Using Multiple-Wavelength Excitation...

Area of Science:

• Biomedical Optics
• Fluorescence Spectroscopy
• Medical Imaging

Background:

• Spectroscopy quantifies fluorophores but struggles with overlapping signals from unknown baselines.
• Accurate quantification is crucial in applications like neuro-oncology for distinguishing tumor margins.

Purpose of the Study:

• To develop a novel spectroscopy method for robust fluorophore quantification without prior baseline assumptions.
• To improve the accuracy of protoporphyrin IX (PpIX) detection for glioma diagnosis.

Main Methods:

• Utilized fluorescence signals across multiple excitation wavelengths.
• Developed a least squares estimator with a closed-form expression.
• Validated the method using a digital phantom calibrated with clinical and experimental data.

Main Results:

• The proposed method demonstrated superior robustness compared to existing techniques in simulated clinical glioma images.
• Achieved up to 87% accuracy in distinguishing healthy from tumor tissue, outperforming methods that dropped to near 0% accuracy.
• Effectively handled high variability in baseline fluorescence signals.

Conclusions:

• The new method offers a significant advancement in accurate fluorophore quantification, particularly for complex biological samples.
• This technique holds promise for enhancing diagnostic accuracy in neuro-oncology and other fields relying on fluorescence imaging.