The latest study by an international group of scientists will allow identifying dangerous brain tumours using a neural network, which will greatly simplify the work with MRI images and make more accurate diagnoses, chelTV reports.
A group of scientists consisting of five representatives from universities in India and scientists from South Ural State University have developed methods for computer analysis of MRI images to detect glioblastoma tumours. The research results are published in the scientific “Journal of Big Data“.
A special neural network is capable of extracting characteristic signs from MRI images, which can be used to identify a dangerous tumour. A neural network can simplify this work, reduce the requirements for a person’s qualifications and provide a higher diagnosis accuracy since it is ahead of its counterparts in terms of training time and image classification accuracy.