Hybrid Brain Tumor Classification Scheme of Histopathology Hyperspectral Images Using Linear Unmixing and Deep Learning
Article
Overview
Research
Identity
Additional Document Info
View All
Overview
abstract
Hyperspectral imaging (HSI) has demonstrated its potential to provide correlated spatial and spectral information of a sample by a non-contact technology. In the medical field, especially in histopathology, HSI has been applied for the classification and identification of diseased tissue and for the characterization of its morphological properties. Nevertheless, the identification of a pathology through this spatial-spectral information is not a simple task, due to the variability among samples and the lack of distinctive spectral separability between healthy and diseased tissue. For this reason, in this work, we propose a hybrid methodology to classify non-tumor and tumor histological brain samples by hyperspectral imaging. The proposed approach combines the identification of characteristic components in a hyperspectral image by linear unmixing, as a features engineering step, and the classification robustness of a deep learning approach. For this last step, an ensemble of neural networks is trained using a partition of an augmented dataset. The proposed method can classify histological brain samples with an average accuracy of 88%25, a reduced variability and computational cost, which presents an advantage over comparison methods in the state-of-the-art. This study demonstrates the potential of hybrid classification methodologies to achieve robust and reliable results by combining linear unmixing for features extraction and deep learning for classification.
