Remote diagnosis of diabetics patient through speech engine and fuzzy based machine learning algorithm

A Systematic Review on Metaheuristic Optimization Techniques for Feature Selections in Disease Diagnosis: Open Issues and Challenges

“…The last step of classification was done by Deep Belief Network using Grey Wolf Optimization, which helped detect brain tumor images. [ 89 ] presented a technique for predicting diabetes in patients. This algorithm used fuzzy logic with Grey Wolf Optimization (GWO) for optimized results.…”

“…37% F-mean—97.52% G-mean—97.38% Accuracy—97.43% [ 87 ] Grey wolf optimization Feature selection and identification of Lung Diseases Accuracy The proposed performed well with the classification and consumed less computation cost as well as computed high accuracy The work should include more datasets as well as deep learning techniques to optimize the performance With k-NN—99.4% With random forest– 99.2% With SVM(Linear)– 99.0% With decision tree—98.4% [ 78 ] Bat algorithm Diabetes mellitus detection Accuracy—98.65% The proposed algorithm showed their superiority in terms of their performance The algorithm should be tested on other chronic diseases . [ 70 ] Recurrent neural network using Bat optimizer Anti viral cure drug of SARS CoV-2 Accuracy—96.08% The proposed model showed the best performance of prediction The model should be incorporated in clinical settings [ 88 ] Grey wolf optimization Brain tumor Accuracy—94.1% The performance of the proposed model was better than the existing ones as mentioned in their paper The model was trained with smaller size of dataset Sensitivity- 88.9% Specificity- 100% Precision- 100% [ 89 ] Grey wolf optimization using fuzzy logic Diabetes prediction Accuracy—81% The proposed algorithm showed a great potential in long term outcomes The system worked on few parameters to predict the diab...…”