# Reinforcement learning using Deep Q Networks and Q learning accurately localizes brain tumors on MRI with very small training sets

arXiv: https://arxiv.org/abs/2010.10763 · HTML: https://mlsanity.com/paper/2010.10763/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 Oct 2020.

## Built on by (most cited first, 5 of 5)

- [Deep reinforcement learning with automated label extraction from clinical reports accurately classifies 3D MRI brain volumes](https://mlsanity.com/paper/2106.09812/) (2021)
- [Deep Neuroevolution Squeezes More out of Small Neural Networks and Small Training Sets: Sample Application to MRI Brain Sequence Classification](https://mlsanity.com/paper/2112.12990/) (2021)
- [Deep reinforcement learning-based image classification achieves perfect testing set accuracy for MRI brain tumors with a training set of only 30 images](https://mlsanity.com/paper/2102.02895/) (2021)
- [Sequential Attention-based Sampling for Histopathological Analysis](https://mlsanity.com/paper/2507.05077/) (2025)
- [Deep reinforcement learning for fMRI prediction of Autism Spectrum Disorder](https://mlsanity.com/paper/2206.11224/) (2022)

## Abstract (arXiv)

Purpose Supervised deep learning in radiology suffers from notorious inherent limitations: 1) It requires large, hand-annotated data sets, 2) It is non-generalizable, and 3) It lacks explainability and intuition. We have recently proposed Reinforcement Learning to address all threes. However, we applied it to images with radiologist eye tracking points, which limits the state-action space. Here we generalize the Deep-Q Learning to a gridworld-based environment, so that only the images and image masks are required. Materials and Methods We trained a Deep Q network on 30 two-dimensional image slices from the BraTS brain tumor database. Each image contained one lesion. We then tested the trained Deep Q network on a separate set of 30 testing set images. For comparison, we also trained and tested a keypoint detection supervised deep learning network for the same set of training / testing images. Results Whereas the supervised approach quickly overfit the training data, and predicably performed poorly on the testing set (11% accuracy), the Deep-Q learning approach showed progressive improved generalizability to the testing set over training time, reaching 70% accuracy. Conclusion We have shown a proof-of-principle application of reinforcement learning to radiological images, here using 2D contrast-enhanced MRI brain images with the goal of localizing brain tumors. This represents a generalization of recent work to a gridworld setting, naturally suitable for analyzing medical images.