LETT-NeXt: A Lightweight RECIST-Guided Model for 3D CT Lesion Segmentation
2026-06-29 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors developed LETT-NeXt, a computer model that takes simple 2D tumor size marks from CT scans and uses them to predict the full 3D shape of tumors. Their model focuses closely around the tumor markings and processes both the scans and markings together to make more accurate 3D tumor outlines. They tested LETT-NeXt on public and hidden data sets, showing good accuracy and efficient speed with low memory needs. This approach helps improve how doctors can understand tumor size and shape beyond basic measurements.
RECISTCT volume3D lesion segmentationDice Similarity CoefficientNormalized Surface DiceMedNeXt-v2encoder-decoder modelAutoZoom inferencetumor response assessmentcomputer-aided diagnosis
Authors
Sebastian Aas, Elias Stenhede, Arian Ranjbar
Abstract
RECIST diameter measurements are widely used for tumor response assessment, but they provide only a limited 2D description of lesion extent. We present LETT-NeXt, a lightweight RECIST-guided model that predicts 3D lesion masks from CT volumes and RECIST markers for the CVPR 2026 Foundation Models for Pan-cancer Segmentation in CT Images competition. LETT-NeXt extracts a RECIST-centered regional crop, encodes the RECIST line and endpoints as two prompt channels, and concatenates them with the CT input. A compact MedNeXt-v2 encoder--decoder predicts the lesion mask, followed by prompt-aware component selection and adaptive AutoZoom inference. On the public validation set, LETT-NeXt achieved a Dice Similarity Coefficient (DSC) of 79.4 $\pm$ 10.1 and a Normalized Surface Dice (NSD) of 72.3 $\pm$ 16.2. On the hidden test set, it achieved a DSC of 73.9 and an NSD of 67.3, corresponding to a challenge score of 70.6\%. On the public validation mirror, LETT-NeXt completed CPU inference in 6.9 $\pm$ 3.0 s per case with a peak memory use of 3.6 GB. Code is available at github.com/Ahus-AIM/lett-next.