Predictive model for laser-induced tissue necrosis with immunohistochemistry validation

Photoacoustic imaging holds promise to provide critical guidance in surgical interventions, but its widespread use is challenged by the absence of applicable safety guidelines across diverse target tissues. The biosafety of this technology is primarily associated with the risk of necrosis generation, which is an irreversible thermal effect that can result from prolonged, high-energy laser applications.

We introduce the first known numerical simulation approaches to assess laser-induced necrosis in liver tissue and present a novel microscopy analysis framework to validate performance.

Our simulation methods integrate Monte Carlo simulations of laser-tissue interaction with the COMSOL interface, model local tissue heating, and predict associated tissue damage to quantify the percentage of tissue necrosis resulting from laser application. Our initial predictions are based on 30 and 73 mJ mean laser energies, laser irradiation times of 1, 10, and 20 min, and a 750 nm laser wavelength. Empirical validations with in vivo porcine liver exposed to a mean laser energy of 73 mJ and 750 nm laser wavelength were performed based on H&E and cleaved Caspase-3 immunohistochemistry (IHC) results. Simulation results from the lower 30 mJ laser energy were additionally cross-referenced with previous qualitative H&E-based reports.

Negligible tissue damage was observed with necrosis predictions $\le 15.05\%$ , damage thresholds were determined to be within the 15.05% to 66.23% necrosis prediction range, and necrosis predictions deviated from quantitative IHC results by 0.01% to 8.1%.

We successfully demonstrated an in silico alternative to the otherwise time-consuming and expensive empirical assessments that would be required to create tissue-specific laser safety guidelines. The presented methods have the potential to be translated to multiple tissues and additional laser properties.