Table 2 Classification performance of deep learning models and machine learning algorithm for our method (Accuracy, balanced accuracy, precision,, F1 score, specificity, NPV, AUC calculated using a weighted approach).
From: A deep learning model for diagnosis of inherited retinal diseases
Algorithm | Accuracy | Balanced accuracy | Precision | F1 | Specificity | NPV | AUC | |
|---|---|---|---|---|---|---|---|---|
ML algorithms | Extremely randomized tree26 | 72.22 (71.6-72.84) | 63.88 | 73.46 | 65.07 | 84.02 | 90.91 | 88.22 |
Support vector machine27 | 72.84 (71.6-73.46) | 63.29 | 57.59 | 63.69 | 81.89 | 92.06 | 79.38 | |
XGBoost28 | 75.93 (74.69–76.55) | 66.16 | 59.97 | 66.51 | 83.77 | 93.84 | 87.15 | |
Random forest29 | 75.93 (74.07–77.17) | 66.16 | 58.99 | 66.14 | 85.69 | 94.02 | 80.81 | |
LightGBM30 | 77.16 (75.3-78.14) | 67.92 | 77.66 | 69.81 | 84.53 | 94.02 | 81.17 | |
DL networks | MnasNet-A131 | 83.33 (80.85–86.42) | 78.85 | 82.46 | 81.96 | 90.67 | 93.51 | 94.93 |
AlexNet32 | 87.65 (87.65–90.12) | 86.55 | 87.83 | 87.52 | 94.19 | 93.95 | 97.58 | |
VGG1133 | 92.59 (90.12–93.83) | 91.33 | 92.5 | 92.47 | 96.17 | 96.61 | 97.26 | |
ShuffleNetV21 × 34 | 93.83 (91.36–94.44) | 93.09 | 93.76 | 93.78 | 96.86 | 96.96 | 98.39 | |
VGG1333 | 94.44 (92.59–95.06) | 93.22 | 94.46 | 94.35 | 96.97 | 97.63 | 98.02 | |
Inception V335 | 95.68 (95.06–96.3) | 94.6 | 95.71 | 95.64 | 97.5 | 98.2 | 99.37 | |
ResNet5036 | 95.06 (94.44–95.06) | 93.73 | 95.13 | 95 | 97.08 | 98.01 | 98.25 | |
VGG1633 | 95.68 (95.06–96.3) | 94.98 | 95.64 | 95.65 | 97.82 | 97.98 | 98.52 | |
DenseNet12137 | 95.06 (94.44–95.68) | 94.47 | 95.04 | 95.05 | 97.63 | 97.56 | 98.83 | |
MobileNetV238 | 96.3 (95.06–96.3) | 95.48 | 96.3 | 96.27 | 97.92 | 98.39 | 99.31 | |
