Estimation of the temperature field in laser-induced hyperthermia experiments with a phantom

BACKGROUND

One of the challenges faced during the hyperthermia treatment of cancer is to monitor the temperature distribution in the region of interest. The main objective of this work was to accurately estimate the transient temperature distribution in the heated region, by using a stochastic heat transfer model and temperature measurements.

METHODS

Experiments involved the laser heating of a cylindrical phantom, partially loaded with iron oxide nanoparticles. The nanoparticles were manufactured and characterized in this work. The solution of the state estimation problem was obtained with an algorithm of the Particle Filter method, which allowed for simultaneous estimation of state variables and model parameters. Measurements of one single sensor were used for the estimation procedure, which is highly desirable for practical applications in order to avoid patient discomfort.

RESULTS

Despite the large uncertainties assumed for the model parameters and for the coupled radiation-conduction model, discrepancies between estimated temperatures and internal measurements were smaller than 0.7 °C. In addition, the estimated fluence rate distribution was physically meaningful. Maximum discrepancies between the prior means and the estimated means were of 2% for thermal conductivity and heat transfer coefficient, 4% for the volumetric heat capacity and 3% for the irradiance.

CONCLUSIONS

This article demonstrated that the Particle Filter method can be used to accurately predict the temperatures in regions where measurements are not available. The present technique has potential applications in hyperthermia treatments as an observer for active control strategies, as well as to plan personalized heating protocols.

Estimation of Tumor Size Evolution Using Particle Filters

Cancer is characterized by the uncontrolled growth of cells with the ability of invading local organs and/or tissues and of spreading to other sites. Several kinds of mathematical models have been proposed in the literature, involving different levels of refinement, for the evolution of tumors and their interactions with chemotherapy drugs. In this article, we present the solution of a state estimation problem for tumor size evolution. A system of nonlinear ordinary differential equations is used as the state evolution model, which involves as state variables the numbers of tumor, normal and angiogenic cells, as well as the masses of the chemotherapy and anti-angiogenic drugs in the body. Measurements of the numbers of tumor and normal cells are considered available for the inverse analysis. Parameters appearing in the formulation of the state evolution model are treated as Gaussian random variables and their uncertainties are taken into account in the estimation of the state variables, by using an algorithm based on the auxiliary sampling importance resampling particle filter. Test cases are examined in the article dealing with a chemotherapy protocol for pancreatic cancer.

The contrast and convergence of Bayesian and frequentist statistical approaches in pharmacoeconomic analysis

The application of Bayesian statistical analyses has been facilitated in recent years by methodological advances and an increasing complexity necessitated within research. Substantial debate has historically accompanied this analytic approach relative to the frequentist method, which is the predominant statistical ideology employed in clinical studies. While the essence of the debate between the two branches of statistics centres on differences in the use of prior information and the definition of probability, the ramifications involve the breadth of research design, analysis and interpretation. The purpose of this paper is to discuss the application of frequentist and Bayesian statistics in the pharmacoeconomic assessment of healthcare technology. A description of both paradigms is offered in the context of potential advantages and disadvantages, and applications within pharmacoeconomics are briefly addressed. Additional considerations are presented to stimulate further development and to direct appropriate applications of each method such that the integrity and robustness of scientific inference be strengthened.