Nadine Abi-Jaoudeh MD, Presenter: Nothing to Disclose

Niels Noordhoek PhD, Abstract Co-Author: Employee, Koninklijke Philips Electronics NV

Ankur Kapoor PhD, Abstract Co-Author: Nothing to Disclose

Ayele Negussie PhD, Abstract Co-Author: Nothing to Disclose

Ming Li PhD, Abstract Co-Author: Nothing to Disclose

Alessandro Guido Radaelli PHD, MS, Abstract Co-Author: Employee, Koninklijke Philips Electronics NV

Bart Carelsen MSc,PhD, Abstract Co-Author: Employee, Koninklijke Philips Electronics NV

Bradford J. Wood MD, Abstract Co-Author: Support, Koninklijke Philips Electronics NV Support, Celsion Corporation Support, Biocompatibles International plc Support, W. L. Gore & Associates, Inc

High Intensity Focused Ultrasound (HIFU) is applied for tumor ablation and to induce hyperthermia for enhanced drug delivery. Precise targeting of the HIFU focus and monitoring is needed. Ultrasound or MRI imaging is more commonly used. CT for procedure is widely used for image guidance and CT thermometry has demonstrated good correlation between temperature changes and Hounsfield units (HU). Cone beam CT guidance is a convenient option to provide image guidance for ablation targeting and temperature mapping. The purpose of the study was to determine whether cone-beam CT thermometry is reliable for temperature mapping.

Cone Beam CT (CBCT) can theoretically determine tissue temperature. Per published literature, temperature/HU relation is 0.56 HU/degree C. Two sets of experiments were performed. A water container was heated until the water temperature reached 100°C. Two temperature probes were placed in the water container for exact temperature monitoring. Fifteen CBCT scans (Xper CT, Philips Medical Systems, Best, NL) were obtained of the phantom in the same position with the same parameters, at different temperatures as the water cooled down. The second experiment consisted of ablating a 2 cm zone in an agar phantom which was imaged with serial CBCT. The temperature was monitored with three thermocouples placed at different distances from the ablation zone. The HU values were obtained from the CBCT scans (HU=voxel value*slope+ intercept). A 5x5x5 mm voxel size was sampled to obtain an average density/HU which was used to calculate the temperature. The temperature per calculations from CBCT changes of density was compared to actual temperature measurements per thermocouples. U (1))/0.56.  

Clear correlation between temperatures and HU values was present. Noise level is about 10 degrees Celsius for a 5mm voxel. Requirement of 4 degrees noise level in a 5 mm voxel was not met in these experiments. However the results are encouraging and further improvements are needed.

CBCT can determine temperature change with a 10°C noise level. Requirement of 4 degrees noise level in a 5 mm voxel was not met. The results are encouraging but further improvements are needed.

CBCT is increasingly used for image guidance in ablation and CBCT thermometry would provide valuable monitoring moreover CBCT Thermometry can enable CBCT guided HIFU.

Abi-Jaoudeh, N, Noordhoek, N, Kapoor, A, Negussie, A, Li, M, Radaelli, A, Carelsen, B, Wood, B, Cone Beam CT Thermometry.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11034509.html