REGULAR CONTENT
Final ID
573
Type
Original Scientific Research-Oral or Pos
Authors
A Eifler1, A Hoogi2, D Rubin3, L Hofmann4
Institutions
1Stanford University, Palo Alto, CA, 2Stanford University, Stanford, CA, 3Stanford University, Richard M., Stanford, CA, 4Stanford University Medical Center, Palo Alto, CA
Purpose
For May-Thurner Syndrome (MTS), the degree of left common iliac vein stenosis which causes symptoms remains unknown and the characterization of lower extremity symptoms can be subjective. Determining which patients may benefit from treatment, therefore, is a challenge. The purpose of this study was to demonstrate the feasibility of efficiently obtaining quantitative image-based measures of disease severity from contrast enhanced CT studies of MTS patients.
Materials & Methods
Patients seen and treated for MTS at our single academic institution who obtained contrast enhanced CT of the lower extremities were identified. Representative CT studies were selected, de-identified, and transferred to a research server where they were analyzed in Matlab. Novel segmentation algorithms were created to distinguish fat, muscle, skin, and bone regions for each axial image then used to calculate the total volume of fat, percentage volume of fat (PVF) relative to the total leg volume at that slice, fat density, and leg circumference for each leg. Differences between left and right leg values were compared using paired t-tests. Right leg values were considered internal controls.
Results
Following creation of segmentation algorithms, values were successfully obtained from 3 CT studies with results shown the Table. Volume of fat, PVF, fat density, and leg circumference were higher in the left leg compared to right, with PVF being statistically significant, p<0.05.
Conclusions
Novel measures of disease severity in MTS can be efficiently obtained from CT data via automated segmentation methods. With further investigation, these values could be used to help guide patient selection and determine effectiveness of therapy.
Final ID
573
Type
Original Scientific Research-Oral or Pos
Authors
A Eifler1, A Hoogi2, D Rubin3, L Hofmann4
Institutions
1Stanford University, Palo Alto, CA, 2Stanford University, Stanford, CA, 3Stanford University, Richard M., Stanford, CA, 4Stanford University Medical Center, Palo Alto, CA
Purpose
For May-Thurner Syndrome (MTS), the degree of left common iliac vein stenosis which causes symptoms remains unknown and the characterization of lower extremity symptoms can be subjective. Determining which patients may benefit from treatment, therefore, is a challenge. The purpose of this study was to demonstrate the feasibility of efficiently obtaining quantitative image-based measures of disease severity from contrast enhanced CT studies of MTS patients.
Materials & Methods
Patients seen and treated for MTS at our single academic institution who obtained contrast enhanced CT of the lower extremities were identified. Representative CT studies were selected, de-identified, and transferred to a research server where they were analyzed in Matlab. Novel segmentation algorithms were created to distinguish fat, muscle, skin, and bone regions for each axial image then used to calculate the total volume of fat, percentage volume of fat (PVF) relative to the total leg volume at that slice, fat density, and leg circumference for each leg. Differences between left and right leg values were compared using paired t-tests. Right leg values were considered internal controls.
Results
Following creation of segmentation algorithms, values were successfully obtained from 3 CT studies with results shown the Table. Volume of fat, PVF, fat density, and leg circumference were higher in the left leg compared to right, with PVF being statistically significant, p<0.05.
Conclusions
Novel measures of disease severity in MTS can be efficiently obtained from CT data via automated segmentation methods. With further investigation, these values could be used to help guide patient selection and determine effectiveness of therapy.