Identification of Key Breast Features Using a Neural Network: Applications of Machine Learning in the Clinical Setting of Plastic Surgery

BACKGROUND

In plastic surgery, evaluation of breast symmetry is an important aspect of clinical practice. Computer programs have been developed for this purpose, but most of them require operator input. Artificial intelligence has been introduced into many aspects of medicine. In plastic surgery, automated neural networks for breast evaluation could improve quality of care. In this work, the authors evaluate the identification of breast features with an ad hoc trained neural network.

METHODS

An ad hoc convolutional neural network was developed on the YOLOV3 platform to detect key features of the breast that are commonly used in plastic surgery for symmetry evaluation. The program was trained with 200 frontal photographs of patients who underwent breast surgery and was tested on 47 frontal images of patients who underwent breast reconstruction after breast cancer surgery.

RESULTS

The program was able to detect key features in 97.74% of cases (boundaries of the breast in 94 of 94 cases, the nipple-areola complex in 94 of 94 cases, and the suprasternal notch in 41 of 47 cases). Mean time of detection was 0.52 seconds.

CONCLUSIONS

The ad hoc neural network was successful in localizing key breast features, with a total detection rate of 97.74%. Neural networks and machine learning have the potential to improve the evaluation of breast symmetry in plastic surgery by automated and quick detection of features used by surgeons in practice. More studies and development are needed to further knowledge in this area.

A National Review of Insurance Coverage of Noncancerous Breast Reconstruction

BACKGROUND

Breast reconstruction is commonly performed for a multitude of noncancerous indications, such as correction of congenital deformities, acquired tissue disease, burns, and trauma. However, breast reconstruction for noncancerous indications is often considered cosmetic or not explicitly mentioned in insurance policies. The goal of this study was to assess variability in insurance coverage of breast reconstruction for noncancerous indications.

METHODS

The authors conducted a cross-sectional analysis of 102 US insurance companies, including Medicare and Medicaid, for coverage of breast reconstruction for noncancerous indications (Poland syndrome, fibrocystic breast disease, burns and trauma). Insurance companies were selected based on their state enrollment data and market share. A Web-based search and individual telephone interviews were conducted to identify the policy. Medical necessity criteria were abstracted from publicly available policies.

RESULTS

Half of the insurers (49%, n = 50) had no policy for Poland syndrome, 46% (n = 47) had no policy for burns and trauma, and 82% (n = 84) had no policy for fibrocystic breast disease. Fifty-two percent (n = 22) of policies providing coverage for Poland syndrome, 24% (n = 13) of policies providing coverage for burns and trauma, and 58% (n = 7) of policies providing coverage for fibrocystic breast disease had specific, stringent criteria for medical necessity. Thirty-six percent (n = 15) of policies covering Poland syndrome, 47% (n = 26) of policies covering burns and trauma, and 33% (n = 4) of policies covering fibrocystic breast disease include coverage of the contralateral breast.

CONCLUSIONS

There is a paucity of publicly available information on insurance coverage of breast reconstruction for noncancerous indications and a lack of consensus between top US insurance companies on what constitutes medical necessity for surgical correction.