Table 5 External validation results.
From: A deep neural network approach to predicting clinical outcomes of neuroblastoma patients
Datasets | Balanced accuracy | |||
|---|---|---|---|---|
Training | Test | DNN | SVM | RF |
Clinical outcome = ‘Death from disease’, | ||||
Data = centralities | ||||
Fischer-M | Fischer-M | 87.3% | 75.4% | 75.1% |
Fischer-R | 82.1% | 53.5% | 66.8% | |
Maris | 53.1% | 54.3% | 50.0% | |
Versteeg | 75.0% | 53.3% | 67.5% | |
Fischer-R | Fischer-R | 85.8% | 66.0% | 62.4% |
Fischer-M | 81.5% | 75.4% | 61.2% | |
Maris | 56.2% | 49.7% | 50.0% | |
Versteeg | 70.8% | 68.3% | 67.5% | |
Clinical outcome = ‘Disease progression’, | ||||
Data = centralities | ||||
Fischer-M | Fischer-M | 84.3% | 83.7% | 80.0% |
Fischer-R | 77.0% | 75.2% | 71.8% | |
Maris | 67.5% | 66.0% | 53.8% | |
Versteeg | 78.1% | 82.4% | 78.1% | |
Fischer-R | Fischer-R | 83.7% | 81.0% | 73.3% |
Fischer-M | 80.0% | 76.8% | 75.0% | |
Maris | 67.5% | 58.8% | 58.8% | |
Versteeg | 80.1% | 77.2% | 73.9% | |