English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Purpose: Image classification may be the fundamental task in imagingartificial intelligence. We have recently shown that reinforcement learning canachieve high accuracy for lesion localization and segmentation even withminuscule training sets. Here, we introduce reinforcement learning for imageclassification. In particular, we apply the approach to normal vs.tumor-containing 2D MRI brain images.  Materials and Methods: We applied multi-step image classification to allowfor combined Deep Q learning and TD(0) Q learning. We trained on a set of 30images (15 normal and 15 tumor-containing). We tested on a separate set of 30images (15 normal and 15 tumor-containing). For comparison, we also trained andtested a supervised deep-learning classification network on the same set oftraining and testing images.  Results: Whereas the supervised approach quickly overfit the training dataand as expected performed poorly on the testing set (57% accuracy, just overrandom guessing), the reinforcement learning approach achieved an accuracy of100%.  Conclusion: We have shown a proof-of-principle application of reinforcementlearning to the classification of brain tumors. We achieved perfect testing setaccuracy with a training set of merely 30 images.

Deep reinforcement learning-based image classification achieves perfect  testing set accuracy for MRI brain tumors with a training set of only 30  images