Artificial neural networks play a crucial role in machine learning and there is a need to improve their performance. This paper presents FOXANN, a novel classification model that combines the recently developed Fox optimizer with ANN to solve ML problems. Fox optimizer replaces the backpropagation algorithm in ANN; optimizes synaptic weights; and achieves high classification accuracy with a minimum loss, improved model generalization, and interpretability. The performance of FOXANN is evaluated on three standard datasets: Iris Flower, Breast Cancer Wisconsin, and Wine. The results presented in this paper are derived from 100 epochs using 10-fold cross-validation, ensuring that all dataset samples are involved in both the training and validation stages. Moreover, the results show that FOXANN outperforms traditional ANN and logistic regression methods as well as other models proposed in the literature such as ABC-ANN, ABC-MNN, CROANN, and PSO-DNN, achieving a higher accuracy of 0.9969 and a lower validation loss of 0.0028. These results demonstrate that FOXANN is more effective than traditional methods and other proposed models across standard datasets. Thus, FOXANN effectively addresses the challenges in ML algorithms and improves classification performance.

Epilepsy is one of the most common diseases of the nervous system around the world, affecting all age groups and causing seizures leading to loss of control for a period of time. This study presents a seizure detection algorithm that uses Discrete Cosine Transformation (DCT) type II to transform the signal into frequency-domain and extracts energy features from 16 sub-bands. Also, an automatic channel selection method is proposed to select the best subset among 23 channels based on the maximum variance. Data are segmented into frames of one Second length without overlapping between successive frames. K-Nearest Neighbour (KNN) model is used to detect those frames either to ictal (seizure) or interictal (non-seizure) based on Euclidean distance. The experimental results are tested on 21 patients included in the CHB-MIT dataset. The average F1-score was found to be 93.12, whereas the False-Positive Rate (FPR) average was determined to be 0.07.