Machine Learning Algorithms to Predict Breast Cancer Recurrence Using Structured and Unstructured Sources from Electronic Health Records

Abstract

Recurrence is a critical aspect of breast cancer (BC) that is inexorably tied to mortality. Reuse of healthcare data through Machine Learning (ML) algorithms offers great opportunities to improve the stratification of patients at risk of cancer recurrence. We hypothesized that combining features from structured and unstructured sources would provide better prediction results for 5-year cancer recurrence than either source alone. We collected and preprocessed clinical data from a cohort of BC patients, resulting in 823 valid subjects for analysis. We derived three sets of features: structured information, features from free text, and a combination of both. We evaluated the performance of five ML algorithms to predict 5-year cancer recurrence and selected the best-performing to test our hypothesis. The XGB (eXtreme Gradient Boosting) model yielded the best performance among the five evaluated algorithms, with precision = 0.900, recall = 0.907, F1-score = 0.897, and area under the receiver operating characteristic AUROC = 0.807. The best prediction results were achieved with the structured dataset, followed by the unstructured dataset, while the combined dataset achieved the poorest performance. ML algorithms for BC recurrence prediction are valuable tools to improve patient risk stratification, help with post-cancer monitoring, and plan more effective follow-up. Structured data provides the best results when fed to ML algorithms. However, an approach based on natural language processing offers comparable results while potentially requiring less mapping effort.

Approaching Personalized Medicine: The Use of Machine Learning to Determine Predictors of Mortality in a Population with SARS-CoV-2 Infection

“…The ML model based on extreme gradient boosting (XGB) was selected in our study because of its generalizability, low risk of overfitting, high interpretability [25], and high scalability [34]. XGB has been confirmed to be a reliable method for recognizing patterns in other diseases such as lupus erythematosus [16], traumatic brain injury-induced coagulopathy [35], epilepsy [36], diabetes [37], Alzheimer's disease [38,39], HIV [40,41], or different types of cancer [42][43][44][45][46]. We, therefore, used the aforementioned ML technique to determine which factors were most predictive of disease severity in a closed group of patients hospitalized for COVID-19 during the first two months of the pandemic, a time when the population did not yet have herd immunity and had not yet been vaccinated.…”

Machine Learning-Based Prediction of Distant Recurrence in Invasive Breast Carcinoma Using Clinicopathological Data: A Cross-Institutional Study

“…Several ML/AI studies have been conducted in oncology [29][30][31][32][33][34][35]50,51]. Some of these have attempted to predict 5-10-year breast cancer recurrences using both structured and unstructured clinicopathological data from Electronic Health Records (EHRs) [50,51]; however, these studies focused primarily on predicting local rather than distant recurrences.…”

A Data-Driven Approach: Investigating Prognostic Factors for Overall Survival in Breast Conserving Surgery (BCS) using Machine Learning

“…Machine learning serves as a pivotal tool in our study enabling us to delve into complex medical data and make precise predictions regarding survival outcomes in patients who have undergone breast-conserving surgery (BCS) [19]. Unlike conventional prognostic models, machine learning facilitates the creation of personalized prognostic models by considering a broad spectrum of patient-speci c data, including clinical, pathological, and demographic information [20].…”