An Objective Aesthetic Outcome Tool using 3-Dimensional Surface Imaging (3D-SI) to Replace Panel Assessment for Breast Conserving Treatment (BCT).
Association of Breast Surgery ePoster Library. Godden A. 05/13/19; 257082; P038
Ms. Amy Godden
REGULAR CONTENT
Login now to access Regular content available to all registered users.
Abstract
Discussion Forum (0)
Rate & Comment (0)
P038
Topic: Breast surgery
Introduction: The aesthetic goal for BCT is maintenance of symmetry. No gold standard exists to evaluate aesthetic outcome. Panel assessment is most commonly used. With heterogeneous methodology, inherent bias, and poor internal consistency, comparison is unreliable. 3D-SI has advantages over standard photography in that it provides additional views and measures, is quick and simple, and does not require a photographer. It is, however, more expensive. We describe the development of an objective outcome tool using 3D-SI. MethodsREC approved study. 290 women who underwent BCT 1-5 years previously had 3D-SI (VECTRA XT). 3D measures were derived using Mirror™Software, and panel assessment was performed (5 members, blinded to patient ID and surgeon, Harvard 4-point scale). 190 women comprised a training set to create the tool. Measures were entered into a multivariate model to predict panel score. The predicted scores of the remaining 100 women were compared to observed panel assessment for validation. Results6 objective measures were significantly associated with panel score by multivariate analysis and were used in the tool. Correlation between predicted and actual panel score for the training and validation set was moderate (R=0.67 & 0.65 respectively). Limits of agreement in Bland Altman were -1.2to1.2 in the training set and -1.2to1.1 for the validation set. ConclusionsThe preliminary tool has reasonable correlation but defaults towards the median panel score. Adjustment may be required to improve clinical utility. This objective tool will enable the communication and comparison of results in research and provides a method to benchmark clinical performance.
Topic: Breast surgery
Introduction: The aesthetic goal for BCT is maintenance of symmetry. No gold standard exists to evaluate aesthetic outcome. Panel assessment is most commonly used. With heterogeneous methodology, inherent bias, and poor internal consistency, comparison is unreliable. 3D-SI has advantages over standard photography in that it provides additional views and measures, is quick and simple, and does not require a photographer. It is, however, more expensive. We describe the development of an objective outcome tool using 3D-SI. MethodsREC approved study. 290 women who underwent BCT 1-5 years previously had 3D-SI (VECTRA XT). 3D measures were derived using Mirror™Software, and panel assessment was performed (5 members, blinded to patient ID and surgeon, Harvard 4-point scale). 190 women comprised a training set to create the tool. Measures were entered into a multivariate model to predict panel score. The predicted scores of the remaining 100 women were compared to observed panel assessment for validation. Results6 objective measures were significantly associated with panel score by multivariate analysis and were used in the tool. Correlation between predicted and actual panel score for the training and validation set was moderate (R=0.67 & 0.65 respectively). Limits of agreement in Bland Altman were -1.2to1.2 in the training set and -1.2to1.1 for the validation set. ConclusionsThe preliminary tool has reasonable correlation but defaults towards the median panel score. Adjustment may be required to improve clinical utility. This objective tool will enable the communication and comparison of results in research and provides a method to benchmark clinical performance.
P038
Topic: Breast surgery
Introduction: The aesthetic goal for BCT is maintenance of symmetry. No gold standard exists to evaluate aesthetic outcome. Panel assessment is most commonly used. With heterogeneous methodology, inherent bias, and poor internal consistency, comparison is unreliable. 3D-SI has advantages over standard photography in that it provides additional views and measures, is quick and simple, and does not require a photographer. It is, however, more expensive. We describe the development of an objective outcome tool using 3D-SI. MethodsREC approved study. 290 women who underwent BCT 1-5 years previously had 3D-SI (VECTRA XT). 3D measures were derived using Mirror™Software, and panel assessment was performed (5 members, blinded to patient ID and surgeon, Harvard 4-point scale). 190 women comprised a training set to create the tool. Measures were entered into a multivariate model to predict panel score. The predicted scores of the remaining 100 women were compared to observed panel assessment for validation. Results6 objective measures were significantly associated with panel score by multivariate analysis and were used in the tool. Correlation between predicted and actual panel score for the training and validation set was moderate (R=0.67 & 0.65 respectively). Limits of agreement in Bland Altman were -1.2to1.2 in the training set and -1.2to1.1 for the validation set. ConclusionsThe preliminary tool has reasonable correlation but defaults towards the median panel score. Adjustment may be required to improve clinical utility. This objective tool will enable the communication and comparison of results in research and provides a method to benchmark clinical performance.
Topic: Breast surgery
Introduction: The aesthetic goal for BCT is maintenance of symmetry. No gold standard exists to evaluate aesthetic outcome. Panel assessment is most commonly used. With heterogeneous methodology, inherent bias, and poor internal consistency, comparison is unreliable. 3D-SI has advantages over standard photography in that it provides additional views and measures, is quick and simple, and does not require a photographer. It is, however, more expensive. We describe the development of an objective outcome tool using 3D-SI. MethodsREC approved study. 290 women who underwent BCT 1-5 years previously had 3D-SI (VECTRA XT). 3D measures were derived using Mirror™Software, and panel assessment was performed (5 members, blinded to patient ID and surgeon, Harvard 4-point scale). 190 women comprised a training set to create the tool. Measures were entered into a multivariate model to predict panel score. The predicted scores of the remaining 100 women were compared to observed panel assessment for validation. Results6 objective measures were significantly associated with panel score by multivariate analysis and were used in the tool. Correlation between predicted and actual panel score for the training and validation set was moderate (R=0.67 & 0.65 respectively). Limits of agreement in Bland Altman were -1.2to1.2 in the training set and -1.2to1.1 for the validation set. ConclusionsThe preliminary tool has reasonable correlation but defaults towards the median panel score. Adjustment may be required to improve clinical utility. This objective tool will enable the communication and comparison of results in research and provides a method to benchmark clinical performance.
Code of conduct/disclaimer available in General Terms & Conditions
{{ help_message }}
{{filter}}