Table 6 Comparison of related works grouped by diseases and the most accurate model
From: Healthcare predictive analytics using machine learning and deep learning techniques: a survey
Disease | # of papers | Adopted techniques | Highest accuracy | |
|---|---|---|---|---|
ML | DL | Â | ||
Diabetes | 16 | logistic regression, KNN, SVM, RF, NB, DT, SVM, ensemble machine learning | ANN, parameter optimization, deep convLSTM, NNs, SVR, ARX, CNNs, DNN, LSTM | The DL model achieved 98.07% accuracy rate |
COVID-19 | 8 | DT, RF, logistic regression, NB, K-means | CNN, CSO-LSTM, RNN, MobileNetV2, ResNetV2, VGG19, DenseNet201, InceptionV3, Xception, RNN-based LSTMs | Logistic regression achieved the highest accuracy with 98.5% |
Heart | 10 | Logistic regression, KNN, RF, DT, NB, SVM, DT, K-means | NN, J4.8, RF, CSO-LSTM, CNN | CSO-LSTM achieved the highest accuracy with 96.16% |
Liver | 1 | Logistic regression, KNN, DT, SVM, NB, and RF | – | Logistic regression achieved the highest accuracy with 75% |
Multiple Disease Detection | 7 | DT, RF, logistic regression, and NB | DNN, MLPs | Logistic regression achieved the highest accuracy with 98.5% in the heart dataset |