The use of machine learning in sport outcome prediction: A review

Objective: Vagus nerve stimulation (VNS) is an adjunctive and well-established treatment for patients with drug-resistant epilepsy (DRE). However, it is still difficult to identify patients who may benefit from VNS surgery. Our study aims to propose a VNS outcome prediction model based on machine learning with multidimensional preoperative heart rate variability (HRV) indices.Methods: The preoperative electrocardiography (ECG) of 59 patients with DRE and of 50 healthy controls were analyzed. Responders were defined as having at least 50% average monthly seizure frequency reduction at 1-year follow-up. Time domain, frequency domain, and non-linear indices of HRV were compared between 30 responders and 29 non-responders in awake and sleep states, respectively. For feature selection, univariate filter and recursive feature elimination (RFE) algorithms were performed to assess the importance of different HRV indices to VNS outcome prediction and improve the classification performance. Random forest (RF) was used to train the classifier, and leave-one-out (LOO) cross-validation was performed to evaluate the prediction model.Results: Among 52 HRV indices, 49 showed significant differences between DRE patients and healthy controls. In sleep state, 35 HRV indices of responders were significantly higher than those of non-responders, while 16 of them showed the same differences in awake state. Low-frequency power (LF) ranked first in the importance ranking results by univariate filter and RFE methods, respectively. With HRV indices in sleep state, our model achieved 74.6% accuracy, 80% precision, 70.6% recall, and 75% F1 for VNS outcome prediction, which was better than the optimal performance in awake state (65.3% accuracy, 66.4% precision, 70.5% recall, and 68.4% F1).Significance: With the ECG during sleep state and machine learning techniques, the statistical model based on preoperative HRV could achieve a better performance of VNS outcome prediction and, therefore, help patients who are not suitable for VNS to avoid the high cost of surgery and possible risks of long-term stimulation.

Download Full-text

An Application of Machine Learning to Etiological Diagnosis of Secondary Hypertension: Retrospective Study Using Electronic Medical Records (Preprint)

10.2196/preprints.19739 ◽

2020 ◽

Author(s):

Xiaolin Diao ◽

Yanni Huo ◽

Zhanzheng Yan ◽

Haibin Wang ◽

Jing Yuan ◽

...

Keyword(s):

Machine Learning ◽

Aortic Stenosis ◽

Electronic Medical Records ◽

Medical Records ◽

Thyroid Dysfunction ◽

Outcome Prediction ◽

Prediction Models ◽

Secondary Hypertension ◽

Validation Dataset ◽

Composite Outcome

BACKGROUND Secondary hypertension is a kind of hypertension with definite etiology and may be cured. Patients with suspected secondary hypertension can benefit from detection and treatment in time and, conversely, will have higher risk of morbidity and mortality than patients with primary hypertension. OBJECTIVE The aim of this study was to develop and validate machine learning (ML) prediction models of common etiologies in patients with suspected secondary hypertension. METHODS The analyzed dataset was retrospectively extracted from electronic medical records (EMRs) of patients discharged from Fuwai hospital between January 1, 2016 and June 30, 2019. A total of 7532 unique patients were included and divided into two datasets by time: 6302 patients in 2016-2018 as training dataset for model building and 1230 patients in 2019 as validation dataset for further evaluation. Extreme Gradient Boosting (XGBoost) was adopted to develop five prediction models of four etiologies of secondary hypertension and occurrence of any of them, including renovascular hypertension (RVH), primary aldosteronism (PA), thyroid dysfunction and aortic stenosis. Both univariate logistic analysis and Gini impure method were used for feature selection, while grid search and 10-fold cross-validation were used to select the optimal hyperparameters for each model. RESULTS Validation of the composite outcome prediction model showed good performance with an area under the receiver-operating characteristic curve (AUC) of 0.924 in the validation dataset, while the four prediction models of RVH, PA, thyroid dysfunction and aortic stenosis achieved AUC of 0.938, 0.965, 0.959, 0.946, respectively, in the validation dataset. 79 clinical indicators were identified in all and finally used in our prediction models. The result of subgroup analysis on the composite outcome prediction model demonstrated high discrimination with AUCs all higher than 0.890 among all age groups of adults. CONCLUSIONS The ML prediction models in this study showed good performance in detecting four etiologies of patients with suspected secondary hypertension, thus they may potentially facilitate clinical diagnosis decision making of secondary hypertension in an intelligent way. CLINICALTRIAL

Download Full-text