arXiv AI
BS: Take the Hint - Interactive Multitracer PET/CT Lesion Segmentation with a Scribble-Conditioned ResEnc U-Net
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv Computer Vision
Aug 25
Three-Phase Scribble-Adaptive Curriculum Learning for autoPETV Grand Challenge
The paper presents Libo Zhang’s algorithmic solution for the autoPETV Grand Challenge, focusing on interactive lesion segmentation in whole-body PET/CT scans. The method encodes user scribbles as two additional input channels and trains a large residual‑encoder U‑Net (≈140 M parameters) through a three‑phase curriculum over 4000 epochs, progressively moving from fully automatic segmentation to handling user‑provided scribbles and finally correcting its own mistakes via simulated error‑driven steps. Using 1811 studies for training and an ensemble of five‑fold checkpoints, the approach achieves a mean AUC‑Dice of 3.836 and a mean AUC‑DMM of 3.869 in interactive five‑fold cross‑validation, with significant gains from the first corrective scribble.
arXiv AI
Sep 1
Pretrained, Curriculum-Tuned, and Ensembled: A Tracer-Aware Interactive Segmentation Pipeline for AutoPET V
arXiv:2608.30844v1 Announce Type: cross Abstract: Interactive lesion segmentation in whole-body PET/CT requires a model to provide a strong initial prediction while also responding efficiently to spa...
arXiv Computer Vision
Sep 16
IMVS: Interactive Medical Volume Segmentation with Test-Time Adaptation - A New Method for Annotating Radiology Datasets
arXiv:2609.16775v1 Announce Type: new Abstract: Annotating large radiology datasets is bottlenecked by the manual effort of delineating structures slice-by-slice in 3D volumes. Interactive methods re...
arXiv Computer Vision
Aug 31
Anatomy-Aware Promptable Segmentation with Online Interactive Training for AUTOPET V
The paper introduces an anatomy-aware, promptable segmentation model for whole-body lesion detection in FDG and PSMA PET/CT scans, tailored for the AUTOPET V challenge. The approach builds on nnU-Net, employing a two-stage training process: an initial pre-training phase for strong baseline segmentation and an online interactive phase that refines predictions using scribble prompts. Anatomical context is integrated via organ supervision with a shared head predicting both lesions and organs, reducing false positives, while a tracer classifier directs studies to either a combined FDG+PSMA model or a PSMA-specific model. Cross-validation results show that organ-supervised training yields the most stable performance, the interactive stage consistently improves Dice scores, and PSMA-specific training delivers the best tracer-wise results.
arXiv Machine Learning
Oct 12
The Predictability Bias of Joint-Embedding Predictive Learning in Medical Imaging
The paper introduces TAP-JEPA, a joint‑embedding predictive architecture trained on over 113,000 CT volumes, which achieves performance comparable to leading encoders in scan‑level classification. It demonstrates a predictability bias: while anatomy is largely predictable from context, pathology is not, leading to significant degradation in pathology detection when predictions are applied. The study audits the model’s predictive pathway, showing that pathology representations shift toward control regions and that removing pathology tokens minimally affects downstream decisions, underscoring that contextual predictability alone is insufficient for clinical relevance.
arXiv Computer Vision
Aug 31
Prompt-Guided Interactive Segmentation of Interstitial Lung Disease in Thoracic CT
The paper introduces the first adaptation of the MedSAM2 foundation model for interactive 3D segmentation of interstitial lung disease (ILD) on thoracic CT scans. It evaluates three fine‑tuning strategies and four prompt types—bounding‑boxes, points, lassos, and scribbles—finding that full model fine‑tuning yields the best performance, improving Dice scores by 4.7 percentage points over the baseline. A proof‑of‑concept workflow is presented where MedSAM2 is first initialized with an automatic prior and then refined by radiologist prompts, with all resources released on GitHub.