Deep Reinforcement Learning with Automated Label Extraction from Clinical Reports Accurately Classifies 3D MRI Brain Volumes.
Image classification is probably the most fundamental task in radiology artificial intelligence. To reduce the burden of acquiring and labeling data sets, we employed a two-pronged strategy. We automatically extracted labels from radiology reports in Part 1. In Part 2, we used the labels to train a...
| Published in: | Journal of Digital Imaging Vol. 35; no. 5; pp. 1143 - 1153 |
|---|---|
| Main Authors: | , |
| Format: | diagnostic images equations & formulas research tables/charts Journal Article |
| Published: |
Springer Nature
Oct2022
|
| Online Access: | View this record in EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=159758942&site=ehost-live header: @attributes: shortDbName: ccm uiTerm: 159758942 longDbName: CINAHL Complete uiTag: AN controlInfo: bkinfo: dissinfo: jinfo: jid: 08971889 DOQ jtl: Journal of Digital Imaging issn: 08971889 maglogo: N pubinfo: dt: Oct2022 vid: 35 iid: 5 pid: 237 pub: Springer Nature place: New York, New York artinfo: ui: 159758942 156832454 159758942 159758942 10.1007/s10278-022-00644-5 159758942 ppf: 1143 ppct: 10 formats: fmt: – @attributes: type: T – @attributes: type: P tig: atl: Deep Reinforcement Learning with Automated Label Extraction from Clinical Reports Accurately Classifies 3D MRI Brain Volumes. aug: au: Stember, Joseph Nathaniel Shalu, Hrithwik affil: Memorial Sloan Kettering Cancer Center, 1275 York Avenue, 10065, New York, NY, USA sug: subj: Brain Anatomy and Histology Brain Radiography Imaging, Three-Dimensional Classification Magnetic Resonance Imaging Classification Deep Learning Medical Records Automation Technology, Radiologic Human Radiology Service Comparative Studies Descriptive Statistics ab: Image classification is probably the most fundamental task in radiology artificial intelligence. To reduce the burden of acquiring and labeling data sets, we employed a two-pronged strategy. We automatically extracted labels from radiology reports in Part 1. In Part 2, we used the labels to train a data-efficient reinforcement learning (RL) classifier. We applied the approach to a small set of patient images and radiology reports from our institution. For Part 1, we trained sentence-BERT (SBERT) on 90 radiology reports. In Part 2, we used the labels from the trained SBERT to train an RL-based classifier. We trained the classifier on a training set of 40 images. We tested on a separate collection of 24 images. For comparison, we also trained and tested a supervised deep learning (SDL) classification network on the same set of training and testing images using the same labels. Part 1: The trained SBERT model improved from 82 to 100 % accuracy. Part 2: Using Part 1's computed labels, SDL quickly overfitted the small training set. Whereas SDL showed the worst possible testing set accuracy of 50%, RL achieved 100 % testing set accuracy, with a p -value of 4.9 × 10 - 4 . We have shown the proof-of-principle application of automated label extraction from radiological reports. Additionally, we have built on prior work applying RL to classification using these labels, extending from 2D slices to entire 3D image volumes. RL has again demonstrated a remarkable ability to train effectively, in a generalized manner, and based on small training sets. pubtype: Academic Journal doctype: diagnostic images equations & formulas research tables/charts Journal Article ougenre: Article language: English refInfo: holdings: @attributes: islocal: N |
|---|