Nanda Deepa Thimmappa MD, MBBS, Presenter: Nothing to Disclose

Mengchao Pei, Abstract Co-Author: Nothing to Disclose

Yi Wang PhD, Abstract Co-Author: Nothing to Disclose

Constance Chen MD, MPH, Abstract Co-Author: Nothing to Disclose

Christina Y. Ahn MD, Abstract Co-Author: Nothing to Disclose

Joshua Levine MD, Abstract Co-Author: Nothing to Disclose

Martin R. Prince MD, PhD, Abstract Co-Author: Patent agreement, General Electric Company Patent agreement, Hitachi, Ltd Patent agreement, Siemens AG Patent agreement, Toshiba Corporation Patent agreement, Koninklijke Philips Electronics NV Patent agreement, Nemoto Kyorindo Co, Ltd Patent agreement, Bayer AG Patent agreement, Lantheus Medical Imaging, Inc Patent agreement, Bracco Group Patent agreement, Covidien AG Patent agreement, Topspins, Inc Stockholder, Topspins, Inc

Automated perforator flap angiography reporting efficiently creates accurate fat volume measurements; perforator coordinates relative to reference points, and intramuscular distances while eliminating data entry errors.  

Magnetic Resonance (MR) Perforator flap angiography (PFA) reporting requires manual measurements of perforator artery coordinates, diameters, and estimated flap volumes, which is tedious and prone to data entry errors. We have developed a computer reporting system that automatically generates this data for transfer into patient reports.

Reporting software was implemented in xcode/cocoa as an Osirix plugin using objective-C language. The user first identifies the umbilicus as a reference point and then each candidate perforator artery. The software measures the perforating vessel diameters, calculates coordinates and predicts flap volume, assuming an elliptical geometry and the full abdominal subcutaneous fat thickness on a slice by slice basis.Clinical significance: Increasing popularity of breast reconstruction with autologous perforator flaps has created a workflow issue due to tedious perforator measurements that is solved with reporting automation.

In this crossover study, 10 patients underwent MR PFA using manual reporting and this new automated method. Accuracy was assessed by comparing the perforator coordinates relative to umbilicus. Fat volume, intramuscular length, and vessel diameter were also compared. Correlation to surgical findings was made where available.The mean time for each report using the manual method was 122 minutes compared to 46 minutes using the automated method (p < 0.0001, CI 57 – 100). Automation was also used to report perforators in the posterior thigh and gluteus and found to be time efficient and accurate there as well. The margin of error was < 0.5cm in 70% of the cases. Using automation, the fat volume was accurate to 50 g in 90% of the cases. Automated vessel diameter measurements were not reliable probably due to surrounding vessel enhancement.

Thimmappa, N, Pei, M, Wang, Y, Chen, C, Ahn, C, Levine, J, Prince, M, Semi-automated Perforator Flap Reporting.  Radiological Society of North America 2012 Scientific Assembly and Annual Meeting, November 25 - November 30, 2012 ,Chicago IL. http://archive.rsna.org/2012/12023099.html