Application

A novel proton counting detector and accompanying methodology to characterize the dimensions of an artificial implant and human tissues.

Key Benefits

• Provides an alternative way of counting protons utilizing most probable energy as a unit.
• Generates material maps of artificial implants and human tissues required for Monte Carlo (MC) dose calculation.
• Provides a secondary validation of material and density maps provided by metal artifact reduction (MAR) and multi-energy computed tomography (MECT) techniques.

Market Summary

Presence of metal implants in patients complicates the delivery of proton therapy (PT) for cancer treatment due to inaccurate characterization of metal and surrounding tissues and subsequent inability to accurately calculate the dose in treatment planning systems (TPS) because of this. Since 2018, MC dose calculation algorithms are being used in TPSs, but their calculation accuracy depends heavily on accurate composition, density, and geometry information. This information is even more important and needed for patients with metal implants. Methods such as MAR and MECT have been proposed, but these methods do not differentiate the materials due to photon starvation and residual artifacts in surrounding tissues. Due to range uncertainties and dose perturbations from metal implants in the path of proton beam, most PT centers exclude metal implant patients from their general guideline with case-by-case exemptions, suggesting an unmet clinical need.

Technical Summary

Emory researchers have developed a novel method to characterize artificial implants and human tissues using proton counting detector. Specifically, this method provides information about the dimensions of materials in terms of cumulative stopping power (CSP) and relative stopping power (RSP). Each proton is tracked directly by measuring the deposited energy along the proton track using a fast, pixelated spectral detector. To characterize the RSPs, measurements of energy perturbations are performed in the presence of metal implants. Samplings of CSP are made at different locations in an anthropomorphic phantom. CSP and RSP information are extracted from energy spectra at each beam path. Using minimum error of multiple CSP measurements, optimal mass densities are derived for soft tissue and bone. A reconstructed 3D material map of an anthropomorphic phantom is obtained from sparse data by deliberately under sampling the phantom to increase the efficiency of small sensor using compressed sensing algorithms.