|Advances in neural networks and deep learning have opened a new era in medi-cal imaging technology, health care data analysis and clinical diagnosis. This paper focuses on the classification of MRI for diagnosis of early and progres-sive dementia using transfer learning architectures that employ Convolutional Neural Networks-CNNs, as a base model, and fully connected layers of Soft-max functions or Support Vector Machines-SVMs. The diagnostic process is based on the analysis of the neurodegenerative changes in the brain using seg-mented images of brain asymmetry, which has been identified as a predictive imaging source of early dementia. Results from 300 independent simulation runs on a set of four binary and one multiclass MRI classification tasks illus-trate that transfer learning of CNN-based models equipped with SVM output layer is capable to produce better performing models within a few training epochs compared to commonly used transfer learning architectures that combine CNN pretrained models with fully connected Softmax layers. However, exper-imental findings also confirm that longer training sessions appear to compensate for the shortcomings of the fully connected Softmax layers in the long term. Di-agnosis of early dementia on unseen patients’ brain asymmetry MRI data reached an average accuracy of 90.25% with both transfer learning architec-tures, while progressive dementia was promptly diagnosed with an accuracy that reached 95.90% using a transfer learning architecture that has the SVM layer.|
*** Title, author list and abstract as seen in the Camera-Ready version of the paper that was provided to Conference Committee. Small changes that may have occurred during processing by Springer may not appear in this window.