Click here to

Evaluation of the DirectDensity Algorithm for Clinical Use

M Schwer1, A Geiersbach2*, J Shah3, G Grousset4, E Klein5, (1) Rhode Island Hospital / Warren Alpert Medical, Providence, RI, (2) Rhode Island Hospital / Warren Alpert Medical, Providence, RI, (3) Siemens Healthineers, Durham, NC, (4) Siemens Healthineers, (5) Rhode Island Hospital / Warren Alpert Medical, Providence, RI


PO-BPC-Virtual-25 (Saturday, 4/17/2021)   [Eastern Time (GMT-4)]

Purpose: To verify the efficacy of the Siemens Healthineers DirectDensity (DD) algorithm materials with densities greater than 1.83 g/cm3. CT scans of Radiation Oncology patients often have densities higher than cortical bone for many reasons, but most notably prosthetic implants. To utilize the DD algorithm, it is important to verify that these high-density materials will not affect treatment quality.

Methods: A 64 slice Siemens Somatom Confidence 64 Slice, Model 10590100, was used to produce these results, and we expect the results to be within the acceptable range for treatment. Monte Carlo (MC) modelling of the Siemens Somatom Confidence CT and a Gammex RMI phantom was performed using TOPAS MC. Common orthopaedic prosthetic materials will be tested against the baseline MC simulation including Ti, Ca, Cr, Al, 50%Ca/50%Cr, and Stainless Steel.

Results: Both the CT simulator and the Gammex RMI phantom have been successfully modelled using the TOPAS MC software, and an IRB to enroll participants in our project has been completed. To proceed with the project, we need to order the custom plugs for our Gammex phantom.

Conclusion: This project has the potential to greatly improve image quality and reduce clinical errors by allowing the clinic to customize scanning protocols for age and body habitus and utilize one calibration curve in the TPS for multiple kV energies. Many radiation oncology clinics scan only at 120 kV to reduce the potential for error in selecting the wrong calibration curve for a given scan, but with DirectDensity, many energies could be utilized with full confidence. If the DirectDensity algorithm proves within 5% of expected for high-Z materials based on the MC measurement, there is no impediment to implementing the DirectDensity algorithm in our Radiation Oncology clinic.

Funding Support, Disclosures, and Conflict of Interest: Research support provided by Siemens Healthineers.


Dual-energy Imaging, Monte Carlo, Calibration


IM- SPECT : Monte Carlo Modeling

Contact Email