AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

MTNeuro: A Benchmark for Evaluating Representations of Brain Structure Across Multiple Levels of Abstraction

Jorge Quesada, Lakshmi Sathidevi, Ran Liu, Nauman Ahad, Joy M Jackson, Mehdi Azabou, Jingyun Xiao, Chris Liding, Matthew Jin, Carolina Urzay, William Gray-Roncal, Erik Christopher Johnson and 1 more

NeurIPS 2022 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Self-supervised neuroimaging representations · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Train with an additional cube of dataBYOL ROI classifier · ROI-C1 to ROI-C3accuracy: 0.88 ± 0.02 -> 0.97 ± 0.01Improved
Train with an additional cube of dataMYOW ROI classifier · ROI-C1 to ROI-C3accuracy: 0.90 ± 0.02 -> 0.98 ± 0.01Improved
Train with an additional cube of dataSupervised ROI classifier · ROI-C1 to ROI-C3accuracy: 0.88 ± 0.03 -> 0.88 ± 0.02No clear effect
Train supervised ResNet18 with MixupSupervised ROI classifier · ROI-C1accuracy: 0.88 ± 0.03 -> 0.90 ± 0.04Improved
Increase image resolution from 2x downsampling to full resolutionSelf-supervised ROI classifiers · ROI-C1accuracy: BYOL: 0.88 ± 0.02 -> 0.98 ± 0.01; MYOW: 0.90 ± 0.02 -> 0.96 ± 0.03; MYOW-m: 0.94 ± 0.02 -> 0.99 ± 0.01Improved
Increase image resolution from 2x downsampling to full resolutionSupervised ROI classifiers · ROI-C1accuracy: Resnet18: 0.88 ± 0.03 -> 0.87 ± 0.1; Resnet18-Mixup: 0.90 ± 0.04 -> 0.91 ± 0.03No clear effect
Use the 3-class setting instead of 4-class segmentation2D segmentation models · Pixel-level segmentationmean IoU: 0.72 to 0.79Improved
Use 3D inputs instead of 2D inputsU-Net segmentation · 3-class segmentationmean IoU: 0.79 ± 0.014 -> 0.80 ± 0.007No clear effect
Train on two subvolumes rather than oneBYOL feature representation · Semantic feature readoutR², vessels: 0.85 ± 0.01 -> 0.88 ± 0.00Improved
Train on two subvolumes rather than oneSupervised feature representation · Semantic feature readoutR², cell count: 0.67 ± 0.06 -> 0.73 ± 0.02Improved
Use MYOW's merged single-projector and predictor design (ablation)MYOW · ROI-C1accuracy: 0.90 ± 0.02 -> 0.94 ± 0.02Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy