Deep Learning Based Segmentation of Various Brain Lesions for Radiosurgery
This work provides practical insights for medical imaging researchers and clinicians by comparing existing methods on a new dataset, but it is incremental as it does not introduce new algorithms.
The study benchmarked state-of-the-art deep learning segmentation algorithms on a clinical stereotactic radiosurgery dataset to evaluate their performance in segmenting various brain lesions, identifying strengths, weaknesses, and optimal settings.
Semantic segmentation of medical images with deep learning models is rapidly developed. In this study, we benchmarked state-of-the-art deep learning segmentation algorithms on our clinical stereotactic radiosurgery dataset, demonstrating the strengths and weaknesses of these algorithms in a fairly practical scenario. In particular, we compared the model performances with respect to their sampling method, model architecture, and the choice of loss functions, identifying the suitable settings for their applications and shedding light on the possible improvements.