Breast Cancer Detection with Multi Layer Perceptron: Unveiling Neuron Unit Dynamics in Dual Hidden Layers
Simeon Yuda Prasetyo, Ika Dyah Agustia Rachmawati, Ajeng Wulandari · 2024
Breast cancer represents a significant global health burden, with millions of new cases diagnosed each year, making it the most common cancer in women worldwide and a leading cause of cancer-related mortality. Early detection is crucial for improving patient outcomes and reducing mortality rates associated with breast cancer. Machine learning techniques offer promising avenues for enhancing early detection and diagnosis. In this study, we investigate the performance of Multi Layer Perceptron (MLP) models with varying numbers of neuron units in two hidden layers for breast cancer diagnosis. Through systematic experimentation, we identify the optimal configuration, achieving the highest accuracy of 84.91% with the first hidden layer containing 5 neuron units and the second hidden layer containing 10 neuron units. Interestingly, competitive performance is observed in models with 5 neuron units in both hidden layers, showcasing accuracy levels ranging from 84.8% to 84.9%. Variations in other performance metrics across different configurations are minor. Overall, MLP models with 10 neuron units in both hidden layers demonstrate superior performance, with models featuring 5 units in both layers as viable alternatives. These findings underscore the versatility of MLP models in healthcare applications and contribute valuable insights to breast cancer diagnostics.