The ANFIS classifier optimized with FFA and EPSO achieved 98.36% accuracy, 97.85% sensitivity, and 98.78% specificity in detecting epilepsy from EEG signals.
Case-Control
Yes
Does an ANFIS Network optimized with EPSO and FFA Algorithm accurately classify EEG signals for epilepsy detection?
The proposed ANFIS model optimized with FFA and EPSO achieved high accuracy, sensitivity, and specificity for EEG-based epilepsy detection.
- In our present work Intelligent System (IS) is characterized by the technique that integrates with the various applications being operated by the machine learning system. To enhance the performance of the intelligent system, the model is to be learned with training algorithms and finally it has been tested to evaluate its performance. This research work explores thoroughly with various areas of soft computing techniques. Use of soft computing is its capability to reach the decision with low cost and computational time. The model for soft computing is based on the performance of the human brain related to the artificial neural network system. The principal constituents Neuro-Fuzzy Inference System (ANFIS), Evolutionary Computation (EC) and Machine Learning (ML) system. For an investigation regarding the function of brain, electroencephalogram (EEG). Therefore, there is a powerful demand for capturing and analysing the various biomedical information to enhance its clinical application and awareness by using ANFIS Network, with Efficient PSO(EPSO) and FFA Algorithm. The Sensitivity about 97.85%, Specificity about 98.78%, Accuracy about 98.36 %, are achieved
Mahapatra et al. (2026) conducted a case-control in Epilepsy (Idiopathic Partial Epilepsy). ANFIS optimized with FFA and EPSO vs. Existing classification methods was evaluated on Classification accuracy. The ANFIS classifier optimized with FFA and EPSO achieved 98.36% accuracy, 97.85% sensitivity, and 98.78% specificity in detecting epilepsy from EEG signals.