AUTHOR=Wang Shuihua , Tang Chaosheng , Sun Junding , Zhang Yudong 

TITLE=Cerebral Micro-Bleeding Detection Based on Densely Connected Neural Network

JOURNAL=Frontiers in Neuroscience

VOLUME=Volume 13 - 2019

YEAR=2019

URL=https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2019.00422

DOI=10.3389/fnins.2019.00422

ISSN=1662-453X

ABSTRACT=Cerebral micro-bleedings (CMBs) as a type of small chronic brain bleeding could have many side effects to patients who suffer from it, such as inducing stroke, loss of brain functions or side effects from other medications and treatment. Therefore, it is important and essential to detect CMBs timely and better to be in an early stage for prompt treatment. In this research, we proposed employing DenseNet 201 as the basic method, and transfer learning for CMBs detection. In order to generate the samples for training, we used sliding window to cover the whole image from left to right and from up to bottom, and based on the central pixel, we can decide target value. Considering the data imbalance, the cost matrix was used in this research. Then, based on the new model, we tested the classification accuracy, and it achieved 97.71%, which provides better performance than the state of art methods.