Balanced Medical Image Classification with Transfer Learning and Convolutional Neural Networks

This paper aims to propose a tool for image classification in medical diagnosis decision support, context where computational power is limited and then specific, high-speed computing infrastructures cannot be used (mainly economic energy consuming reasons). The proposed method combines deep neural networks algorithm with imaging procedures implemented allow an efficient use on affordable hardware. convolutional network (CNN) procedure VGG16 as its base architecture, using the transfer learning technique parameters obtained ImageNet competition. Two blocks one dense block were added this architecture. was developed calibrated basis of five common lung diseases 5430 images from two public datasets technique. holdout ratios 90% 10% training testing, respectively, obtained, regularization tools dropout, early stopping, Lasso (L2). An accuracy (ACC) 56% area under receiver-operating characteristic curve (ROC—AUC) 50% reached which are suitable support resource-constrained environment.