Ensemble machine learning for the prediction of patient-level outcomes following thyroidectomy

Am J Surg. 2021 Aug;222(2):347-353. doi: 10.1016/j.amjsurg.2020.11.055. Epub 2020 Dec 3.

Abstract

Background: Accurate prediction of thyroidectomy complications is necessary to inform treatment decisions. Ensemble machine learning provides one approach to improve prediction.

Methods: We applied the Super Learner (SL) algorithm to the 2016-2018 thyroidectomy-specific NSQIP database to predict complications following thyroidectomy. Cross-validation was used to assess model discrimination and precision.

Results: For the 17,987 patients undergoing thyroidectomy, rates of recurrent laryngeal nerve injury, post-operative hypocalcemia prior to discharge or within 30 days, and neck hematoma were 6.1%, 6.4%, 9.0%, and 1.8%, respectively. SL improved prediction of thyroidectomy-specific outcomes when compared with benchmark logistic regression approaches. For postoperative hypocalcemia prior to discharge, SL improved the cross-validated AUROC to 0.72 (95%CI 0.70-0.74) compared to 0.70 (95%CI 0.68-0.72; p < 0.001) when using a manually curated logistic regression algorithm.

Conclusion: Ensemble machine learning modestly improves prediction for thyroidectomy-specific outcomes. SL holds promise to provide more accurate patient-level risk prediction to inform treatment decisions.

Keywords: Machine learning; Surgical risk prediction; Thyroidectomy.

MeSH terms

  • Adult
  • Aged
  • Algorithms*
  • Female
  • Humans
  • Logistic Models
  • Machine Learning*
  • Male
  • Middle Aged
  • Postoperative Complications / diagnosis*
  • Postoperative Complications / epidemiology*
  • Predictive Value of Tests
  • ROC Curve
  • Risk Factors
  • Thyroid Diseases / complications
  • Thyroid Diseases / diagnosis
  • Thyroid Diseases / surgery*
  • Thyroidectomy / adverse effects*
  • Treatment Outcome