arXiv:2511.19183v1 Announce Type: new 
Abstract: Semantic segmentation is crucial for various biomedical applications, yet its reliance on large annotated datasets presents a bottleneck due to the high cost and specialized expertise required for manual labeling. Active Learning (AL) aims to mitigate this challenge by querying only the most informative samples, thereby reducing annotation effort. However, in the domain of 3D biomedical imaging, there is no consensus on whether AL consistently outperforms Random sampling. Four evaluation pitfalls hinder the current methodological assessment. These are (1) restriction to too few datasets and annotation budgets, (2) using 2D models on 3D images without partial annotations, (3) Random baseline not being adapted to the task, and (4) measuring annotation cost only in voxels. In this work, we introduce nnActive, an open-source AL framework that overcomes these pitfalls by (1) means of a large scale study spanning four biomedical imaging datasets and three label regimes, (2) extending nnU-Net by using partial annotations for training with 3D patch-based query selection, (3) proposing Foreground Aware Random sampling strategies tackling the foreground-background class imbalance of medical images and (4) propose the foreground efficiency metric, which captures the low annotation cost of background-regions. We reveal the following findings: (A) while all AL methods outperform standard Random sampling, none reliably surpasses an improved Foreground Aware Random sampling; (B) benefits of AL depend on task specific parameters; (C) Predictive Entropy is overall the best performing AL method, but likely requires the most annotation effort; (D) AL performance can be improved with more compute intensive design choices. As a holistic, open-source framework, nnActive can serve as a catalyst for research and application of AL in 3D biomedical imaging. Code is at: https://github.com/MIC-DKFZ/nnActive

تم تقديم nnActive كإطار عمل مفتوح المصدر يهدف إلى تحسين تقييم التعلم النشط (AL) في تقسيم البيانات الطبية ثلاثية الأبعاد، حيث يتناول التحديات الكبيرة في هذا المجال مثل الاعتماد على مجموعات بيانات موسعة ومشروطة وغياب الإجماع حول فعالية AL مقارنة بالعينات العشوائية.

nnActive se ha presentado como un marco de código abierto destinado a mejorar la evaluación del Aprendizaje Activo (AL) en la segmentación biomédica en 3D, abordando desafíos significativos en el campo, como la dependencia de grandes conjuntos de datos anotados y la falta de consenso sobre la efectividad del AL en comparación con el muestreo aleatorio.

nnActive a été introduit comme un cadre open-source visant à améliorer l'évaluation de l'apprentissage actif (AL) dans la segmentation biomédicale 3D, en abordant des défis significatifs dans le domaine tels que la dépendance à de grands ensembles de données annotées et le manque de consensus sur l'efficacité de l'AL par rapport à l'échantillonnage aléatoire.

nnActive has been introduced as an open-source framework aimed at enhancing the evaluation of Active Learning (AL) in 3D biomedical segmentation, addressing significant challenges in the field such as reliance on large annotated datasets and the lack of consensus on the effectiveness of AL compared to random sampling.

nnActive: A Framework for Evaluation of Active Learning in 3D Biomedical Segmentation

arXiv:2403.17770v2 Announce Type: replace-cross 
Abstract: Despite the significant success achieved by deep learning methods in medical image segmentation, researchers still struggle in the computer-aided diagnosis of abdominal lymph nodes due to the complex abdominal environment, small and indistinguishable lesions, and limited annotated data. To address these problems, we present a pipeline that integrates the conditional diffusion model for lymph node generation and the nnU-Net model for lymph node segmentation to improve the segmentation performance of abdominal lymph nodes through synthesizing a diversity of realistic abdominal lymph node data. We propose LN-DDPM, a conditional denoising diffusion probabilistic model (DDPM) for lymph node (LN) generation. LN-DDPM utilizes lymph node masks and anatomical structure masks as model conditions. These conditions work in two conditioning mechanisms: global structure conditioning and local detail conditioning, to distinguish between lymph nodes and their surroundings and better capture lymph node characteristics. The obtained paired abdominal lymph node images and masks are used for the downstream segmentation task. Experimental results on the abdominal lymph node datasets demonstrate that LN-DDPM outperforms other generative methods in the abdominal lymph node image synthesis and better assists the downstream abdominal lymph node segmentation task.

تم تطوير خط أنابيب جديد يدمج نماذج الانتشار الشرطية وnnU-Net لتحسين تقسيم العقد اللمفاوية البطنية، حيث يتناول التحديات مثل البيئات المعقدة والبيانات المعلّمة المحدودة. يستخدم هذا النهج LN-DDPM، وهو نموذج احتمالي لإزالة الضوضاء الشرطية، لتوليد بيانات عقد لمفاوية متنوعة وواقعية، مما يحسن أداء التقسيم.

Se ha desarrollado un nuevo pipeline que integra modelos de difusión condicional y nnU-Net para mejorar la segmentación de ganglios linfáticos abdominales, abordando desafíos como entornos complejos y datos anotados limitados. Este enfoque utiliza LN-DDPM, un modelo probabilístico de difusión de eliminación de ruido, para generar datos de ganglios linfáticos diversos y realistas, mejorando así el rendimiento de la segmentación.

Un nouveau pipeline intégrant des modèles de diffusion conditionnels et nnU-Net a été développé pour améliorer la segmentation des ganglions lymphatiques abdominaux, en s'attaquant à des défis tels que des environnements complexes et des données annotées limitées. Cette approche utilise LN-DDPM, un modèle probabiliste de diffusion de débruitage conditionnel, pour générer des données de ganglions lymphatiques diverses et réalistes, améliorant ainsi les performances de segmentation.

A new pipeline integrating conditional diffusion models and nnU-Net has been developed to enhance the segmentation of abdominal lymph nodes, addressing challenges such as complex environments and limited annotated data. This approach utilizes LN-DDPM, a conditional denoising diffusion probabilistic model, to generate diverse and realistic lymph node data, improving segmentation performance.

nnActive: A Framework for Evaluation of Active Learning in 3D Biomedical Segmentation

Was this article worth reading? Share it

LucidQuery AI

Attentive AI

Brainactive