Enhancing classification of active and non-active lesions in multiple sclerosis: machine learning models and feature selection techniques

Table 3 Parameters of optimized sequential model

Number of layers	Type of layer	Neurons	Trainable parameters	Activation function
1	Dense input layer	107	11,556	selu
2	Dense internal layer	150	16,200	selus
3	Dense internal layer	100	15,100	selu
4	Dense internal layer	50	5050	selu
5	Dense output layer	1	51	sigmoid

ISSN: 1471-2342