Published in:
01-09-2020 | Mastectomy | ASO Author Reflections
ASO Author Reflections: Machine Learning Strategies Can Aid Patient Selection in Microvascular Breast Reconstruction
Author:
Anne C. O’Neill, MBBCh, PhD
Published in:
Annals of Surgical Oncology
|
Issue 9/2020
Login to get access
Excerpt
Although there are many effective options for post mastectomy breast reconstruction, microvascular autologous techniques are widely considered to be the “gold standard”, because they offer excellent aesthetic results and long-term patient satisfaction.
1 However, like all microvascular procedures, there is an inherent risk of flap failure with rates of 1–2% reported in most high-volume specialist centres. Preoperative identification of patients who are at increased risk for failure would allow surgeons to recommend an alternative reconstructive technique. Traditional approaches for the identification of significant preoperative risk factors centre on the use of logistic regression models. However, these strategies often are ineffective when the outcome of interest is rare. Previous studies have failed to identify consistent predictors of flap failure in post mastectomy breast reconstruction.
2 …