<p>Spiking neural networks (SNN) have arisen as an energy-efficient substitute for traditional deep learning models. The recent advancement of SNN has allowed it to achieve performance comparable with traditional deep learning models. Medical applications such as medical image classification utilize Deep learning-based methods. This study aims to find the applicability of Deep SNN in medical image classification and the construction of a Deep SNN for medical image classification with fewer parameters. A lightweight deep spiking neural network was developed with three convolution layers and three max-pooling layers consisting of only 1.78 M parameters. The activation function was replaced by a threshold-based biological neuron model integrate and fire. The model was trained using surrogate gradient-based backpropagation. The multiclass classification task of Brain tumour detection was chosen for this study. The proposed model achieved an accuracy of 97 % and an F1 score of 0.97. It was compared to state-of-the-art traditional deep learning models. The lightweight architecture facilitates reducing computational resources during training. This study illustrates the effectiveness of spiking neural networks in medical image classification. It provides empirical evidence endorsing them as a viable alternative to traditional artificial neural networks in medical image analysis.</p>