Radiomic Synthesis Using Deep Convolutional Neural Networks
This provides a faster tool for radiomic image generation in clinical decision support for breast cancer patients, but it is incremental as it applies an existing method to a specific bottleneck.
The paper tackled the computational expense of generating radiomic images, such as GLCM entropy, by developing RadSynth, a deep CNN model that produced synthetic entropy images with an average percentage difference of 0.07 and correlation of 0.97 compared to traditional methods.
Radiomics is a rapidly growing field that deals with modeling the textural information present in the different tissues of interest for clinical decision support. However, the process of generating radiomic images is computationally very expensive and could take substantial time per radiological image for certain higher order features, such as, gray-level co-occurrence matrix(GLCM), even with high-end GPUs. To that end, we developed RadSynth, a deep convolutional neural network(CNN) model, to efficiently generate radiomic images. RadSynth was tested on a breast cancer patient cohort of twenty-four patients(ten benign, ten malignant and four normal) for computation of GLCM entropy images from post-contrast DCE-MRI. RadSynth produced excellent synthetic entropy images compared to traditional GLCM entropy images. The average percentage difference and correlation between the two techniques were 0.07 $\pm$ 0.06 and 0.97, respectively. In conclusion, RadSynth presents a new powerful tool for fast computation and visualization of the textural information present in the radiological images.