Avoid common mistakes on your manuscript.
Past
For women undergoing cancer-related mastectomy and reconstruction, treatment decisions play a crucial role in postoperative outcomes. However, the current decision-making process often relies on non-randomized group-level evidence or physician preferences,1 potentially resulting in sub-optimal treatment recommendations. Machine learning (ML), an AI technology, has significantly enhanced clinician performance by improving risk predictions and streamlining standardized processes.2,3 While most research has focused on image recognition for clinical diagnosis, there is a growing demand for algorithmic support in treatment decision-making. To address this urgent need, this study aims to develop personalized, data-driven decision aids for women undergoing breast reconstruction, with the goal of improving patient-centered care.
Present
We trained three distinct ML models to predict meaningful changes in health-related quality of life for women undergoing mastectomy and breast reconstruction. This study involved 1454 to 1538 patients from 11 North American study sites, with a 2-year follow-up. To create unbiased clinical algorithms that consider treatment preferences and body image, we excluded socioeconomic and ethnic variables while incorporating baseline patient-reported outcomes (PROs) into our models. The models achieved a good accuracy, with an AUC of up to 0.82. Our findings demonstrate that baseline PROs have a stronger influence on postoperative satisfaction with breasts compared with treatment decisions.4 We highlight the potential of ML algorithms in accurately predicting PROs, offering valuable insights for clinical decision-making.
Future
The integration of ML and PROs enables individualized outcome predictions and shared decision-making. ML algorithms empower clinicians to tailor treatment recommendations confidently. A significant shift toward AI-guided clinical decision-making is expected in the near future. However, challenges remain for ML to fulfill its potential in optimizing clinical care. First, clinical implementation is necessary to validate algorithm effectiveness. Second, addressing biases in training data is crucial for improving algorithm fairness.5 We hope that our research initiates the development of truly individualized and data-driven tools to support patient-centered, clinical decision making. The powerful combination of ML and PROs has the potential to change our current way of clinical decision-making.
References
Pusic AL, Matros E, Fine N, et al. Patient-reported outcomes 1 year after immediate breast reconstruction: results of the mastectomy reconstruction outcomes consortium study. J Clin Oncol. 2017;35(22):2499–506. https://doi.org/10.1200/JCO.2016.69.9561.
Xu C, Subbiah IM, Lu SC, Pfob A, Sidey-Gibbons C. Machine learning models for 180-day mortality prediction of patients with advanced cancer using patient-reported symptom data. Qual Life Res. 2023;32(3):713–27. https://doi.org/10.1007/s11136-022-03284-y.
Pfob A, Sidey-Gibbons C, Lee HB, et al. Identification of breast cancer patients with pathologic complete response in the breast after neoadjuvant systemic treatment by an intelligent vacuum-assisted biopsy. Eur J Cancer. 2021;143(565):134–46. https://doi.org/10.1016/j.ejca.2020.11.006.
Xu C, Pfob A, Mehrara BJ, et al. Enhanced surgical decision-making tools in breast cancer: predicting 2-year postoperative physical, sexual, and psychosocial well-being following mastectomy and breast reconstruction (INSPiRED 004). Ann Surg Oncol. 2023. https://doi.org/10.1245/s10434-023-13971-w.
Pfob A, Sidey-Gibbons C. Systematic bias in medical algorithms: to include or not include discriminatory demographic information? JCO Clin Cancer Informatics. 2022;6:1–2. https://doi.org/10.1200/cci.21.00146.
Funding
Supported by National Cancer Institute Grant No. R01 CA152192 and in part by National Cancer Institute Support Grant No. P30 CA008748.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Disclosure
The authors report no conflicts of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Cai Xu and André Pfob: Joint first authors.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Xu, C., Pfob, A. & Sidey-Gibbons, C. ASO Author Reflections: Enhancing Surgical Decision-Making for Breast Reconstruction—Machine Learning-Driven Prediction of Postoperative Quality of Life. Ann Surg Oncol 30, 7135–7136 (2023). https://doi.org/10.1245/s10434-023-14008-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1245/s10434-023-14008-y