DEANN: A healthcare analytic methodology of data envelopment analysis and artificial neural networks for the prediction of organ recipient functional status

The problem of effectively preprocessing a dataset containing a large number of performance metrics and an even larger number of records is crucial when utilizing an ANN. As such, this study proposes deploying DEA to preprocess the data to remove outliers and hence, preserve monotonicity as well as to reduce the size of the dataset used to train the ANN. The results of this novel data analytic approach, i.e. DEANN, proved that the accuracy of the ANN can be maintained while the size of the training dataset is significantly reduced. DEANN methodology is implemented via the problem of predicting the functional status of patients in organ transplant operations. The results yielded are very promising which validates the proposed method. •A hybrid methodology, DEANN, for prediction improvement was developed.•DEA was utilized to classify the dataset in efficiency frontiers.•Predictions were performed using an ANN due to the complexity of dataset.•High accuracy rates with a reduction in training dataset size validate the DEANN.•This generic data analytic method is applicable to numerous domains.