用主动学习减少乳腺影像标注量，兼顾精度与环保。
A Green Solution for Breast Region Segmentation Using Deep Active Learning
- 基于解剖结构设计选样策略，按体位和乳房大小分组数据。
- 30%训练数据下，近邻策略效果最好且碳排放最低。
- 适合医疗影像高效标注与绿色计算研究者参考。
目的：医学乳腺图像标注是提升诊断质量的关键步骤，但耗时费力。本研究聚焦深度主动学习中的样本选择策略，以降低乳腺区域分割（BRS）的训练计算成本并高效利用资源。方法：使用包含59名患者的斯塔万格乳腺MRI数据集，采用FCN-ResNet50作为可持续深度学习模型。提出一种基于乳腺解剖几何（BAG）分析的新样本选择方法，根据患者体位和乳房大小筛选具有相似信息特征的数据。评估了四种策略：随机选择、最近点、乳房大小，以及三者的混合策略。在10%、20%、30%、40%四种训练数据比例下进行模型训练，剩余数据用于测试。通过骰子系数（Dice score）、交并比（IoU）、精确率与召回率，并结合5折交叉验证评估性能。结果：从10%增至40%训练数据，除随机选择外，各策略性能均提升。最近点策略在30%和40%数据比例下碳足迹最低。综合来看，结合最近点策略与30%训练数据，实现了分割性能、效率与环境可持续性的最佳平衡。
原文摘要 · Abstract (English)
Purpose: Annotation of medical breast images is an essential step toward better diagnostic but a time consuming task. This research aims to focus on different selecting sample strategies within deep active learning on Breast Region Segmentation (BRS) to lessen computational cost of training and effective use of resources. Methods: The Stavanger breast MRI dataset containing 59 patients was used in this study, with FCN-ResNet50 adopted as a sustainable deep learning (DL) model. A novel sample selection approach based on Breast Anatomy Geometry (BAG) analysis was introduced to group data with similar informative features for DL. Patient positioning and Breast Size were considered the key selection criteria in this process. Four selection strategies including Random Selection, Nearest Point, Breast Size, and a hybrid of all three strategies were evaluated using an active learning framework. Four training data proportions of 10%, 20%, 30%, and 40% were used for model training, with the remaining data reserved for testing. Model performance was assessed using Dice score, Intersection over Union, precision, and recall, along with 5-fold cross-validation to enhance generalizability. Results: Increasing the training data proportion from 10% to 40% improved segmentation performance for nearly all strategies, except for Random Selection. The Nearest Point strategy consistently achieved the lowest carbon footprint at 30% and 40% data proportions. Overall, combining the Nearest Point strategy with 30% of the training data provided the best balance between segmentation performance, efficiency, and environmental sustainability. Keywords: Deep Active Learning, Breast Region Segmentation, Human-center analysis
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。