LETT-NeXt model enhances 3D CT lesion segmentation with RECIST guidance

By PulseAugur Editorial · [2 sources] · 2026-06-29 10:44

Researchers have developed LETT-NeXt, a lightweight model designed for segmenting 3D lesions in CT scans. This model incorporates RECIST (Response Evaluation Criteria in Solid Tumors) markers as guidance, enhancing its ability to predict lesion masks. LETT-NeXt achieved competitive results in the CVPR 2026 Foundation Models for Pan-cancer Segmentation in CT Images competition, demonstrating efficient inference times and low memory usage. AI

IMPACT This model could improve the accuracy and efficiency of tumor response assessment in clinical settings.

RANK_REASON The cluster describes a research paper detailing a new model for medical image segmentation.

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

COVERAGE [2]

arXiv cs.CV TIER_1 English(EN) · Sebastian Aas, Elias Stenhede, Arian Ranjbar · 2026-06-30 04:00

LETT-NeXt: A Lightweight RECIST-Guided Model for 3D CT Lesion Segmentation

arXiv:2606.30108v1 Announce Type: new 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 …
arXiv cs.CV TIER_1 English(EN) · Arian Ranjbar · 2026-06-29 10:44

LETT-NeXt: A Lightweight RECIST-Guided Model for 3D CT Lesion Segmentation

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 …