Estimation and validation of patient-specific liver elasticity distributions derived from 4DMR for radiotherapy purposes. (10th July 2018)

• Record Type: Journal Article
• Authors: Hasse, Katelyn; Han, Fei; Neylon, John; Min, Yugang; Hu, Peng; Yang, Yingli; Santhanam, Anand
• Is Part Of: Biomedical physics & engineering express. Volume 4:Number 4(2018)
• Publication Date: 2018-07-10
• Subjects: 4DMR -- liver elastography -- biomechanical model -- liver SBRT; Medical physics -- Periodicals; Biophysics -- Periodicals; Biomedical engineering -- Periodicals; Medical sciences -- Periodicals; 610.153
• DOI: 10.1088/2057-1976/aace4d
• Languages: English
• ISSNs: 2057-1976

Abstract

Abstract: Liver diseases, such as liver cancer and cirrhosis, are commonly associated with changes in the biomechanical properties of liver tissue. Functional imaging techniques such as elastography have shown great promise in measuring the biomechanical properties of liver tissue; however, current liver elastography techniques require additional equipment that is conventionally not available within the radiotherapy setup. We present a novel methodology for estimating liver elasticity derived from deformation observed during 0.35 T 4DMR ViewRay (MRIdian System™, ViewRay™, Cleveland, OH, USA) scans within a radiotherapy setup. Phase 1 and phase 8 datasets, categorized by diaphragm position, were first deformably registered. The resulting displacement maps were considered ground-truth. A GPU-based biomechanical model was then assembled from the segmented phase 8 liver dataset and, along with patient-specific boundary constraints, used to iteratively solve for the liver elasticity distribution. The liver elastography process presented here was performed for a set of 11 4DMR patients. Maximum liver deformation was observed to be between 3.99 and 9.04 mm. On average, 95% convergence within 1 mm was observed. A validation study using phase 4 liver datasets illustrated an accuracy of 86%. Normalized cross-correlation quantified high similarity between the results of the estimation and validation studies with their respective ground-truths. Overall, the results suggest that liver elasticity can be measured with approximately 95% convergence using 4DMR scans acquired within the radiotherapy workflow, indicating the potential for the implementation of liver elastography within the clinic.