Interactive method speeds up radiology volume annotation fourteen times
IMVS: Interactive Medical Volume Segmentation with Test-Time Adaptation - A New Method for Annotating Radiology Datasets
Computer Vision and Pattern Recognition
Summary
Annotating medical 3D images like CT or MRI scans is slow because people must carefully draw around structures on many slices. The authors created IMVS, a system that learns from quick user corrections on each slice and adapts its guesses as it processes the volume, reducing repeated work. This method maintains accuracy but drastically cuts the effort and time needed, especially for tricky areas. It works with different segmentation models and improves over older interactive methods by adapting during the annotation.
What this means in practice
- •For medical imaging specialists: Speed up segmentation annotation in radiology datasets by adapting segmentation models interactively to each patient volume.
- •For clinical data teams: Reduce manual effort when curating labeled datasets for training or validating medical AI with interactive and adaptive segmentation tools.
Authors
Abhilaksh Singh Reen, Kushal Borkar, Ritvik Mahapatra
Abstract
Annotating large radiology datasets is bottlenecked by the manual effort of delineating structures slice-by-slice in 3D volumes. Interactive methods reduce this effort but stay interaction-inefficient: slice-wise methods (including many foundation models) ignore inter-slice continuity, while 3D and video-based methods propagate a prompt with a \emph{fixed} propagator that never adapts to the target volume, so it drifts on low-contrast or pathological structures and must be re-prompted. We present IMVS, a human-in-the-loop annotation framework that composes three components into a closed loop rather than a new segmentation primitive: a lightweight 2D Slice Mask Adapter (SMA) fine-tuned online from user scribbles, a frozen Volume Mask Tracker (VMT) that propagates corrected masks across adjacent slices, and a soft teacher--student alignment that limits forgetting. The SMA is backbone-agnostic (UNet++, DeepLabV3, TransUNet). Across 8 public CT/MRI datasets, IMVS matches strong interactive baselines in quality while sharply cutting annotation effort: $14.4\times$ faster than a proficient copy-based manual workflow ($22.3\times$ over naive manual), $4.6\times$ over slice-wise and $1.9\times$ over 3D interactive methods. MedSAM2 and ScribblePrompt stay competitive or stronger on well-delineated organs; IMVS's advantage is largest on challenging targets and on interaction efficiency. Source code and Demo Video: https://github.com/AbhilakshSinghReen/imvs.