Background/Objectives: Developing high-performance artificial intelligence (AI) models for rare diseases is challenging owing to limited data availability. This study aimed to evaluate whether a novel three-class annotation method for preparing training data could enhance AI model performance in detecting osteosarcoma on plain radiographs compared to conventional single-class annotation.Methods: We developed two annotation methods for the same dataset of 468 osteosarcoma X-rays and 378 normal radiographs: a conventional single-class annotation (1C model) and a novel three-class annotation method (3C model) that separately labeled intramedullary, cortical, and extramedullary tumor components. Both models used identical U-Net-based architectures, differing only in their annotation approaches. Performance was evaluated using an independent validation dataset.Results: Although both models achieved high diagnostic accuracy (AUC: 0.99 vs. 0.98), the 3C model demonstrated superior operational characteristics. At a standardized cutoff value of 0.2, the 3C model maintained balanced performance (sensitivity: 93.28%, specificity: 92.21%), whereas the 1C model showed compromised specificity (83.58%) despite high sensitivity (98.88%). Notably, at the 25th percentile threshold, both models showed identical false-negative rates despite significantly different cutoff values (3C: 0.661 vs. 1C: 0.985), indicating the ability of the 3C model to maintain diagnostic accuracy at substantially lower thresholds.Conclusions: This study demonstrated that anatomically informed three-class annotation can enhance AI model performance for rare disease detection without requiring additional training data. The improved stability at lower thresholds suggests that thoughtful annotation strategies can optimize the AI model training, particularly in contexts where training data are limited.
**背景/目的：** 由于数据可用性有限，为罕见疾病开发高性能人工智能模型具有挑战性。本研究旨在评估，与传统的单类别标注方法相比，一种用于准备训练数据的新型三类别标注方法，能否在平片检测骨肉瘤方面提升AI模型的性能。 **方法：** 我们针对同一数据集（包含468张骨肉瘤X光片和378张正常X光片）开发了两种标注方法：一种是传统的单类别标注（1C模型），另一种是新型的三类别标注方法（3C模型），后者分别标注了肿瘤的髓内、皮质和髓外成分。两种模型均采用相同的基于U-Net的架构，仅在标注方法上存在差异。使用独立的验证数据集评估其性能。 **结果：** 尽管两种模型均实现了较高的诊断准确性（AUC：0.99 vs. 0.98），但3C模型表现出更优的操作特性。在标准截断值为0.2时，3C模型保持了平衡的性能（灵敏度：93.28%，特异度：92.21%），而1C模型尽管灵敏度高（98.88%），但特异度较低（83.58%）。值得注意的是，在第25百分位阈值处，尽管截断值差异显著（3C：0.661 vs. 1C：0.985），两种模型的假阴性率相同，这表明3C模型能够在显著更低的阈值下保持诊断准确性。 **结论：** 本研究表明，基于解剖结构信息的三类别标注能够在不增加额外训练数据的情况下，提升AI模型在罕见疾病检测方面的性能。在较低阈值下稳定性的提高表明，深思熟虑的标注策略可以优化AI模型训练，尤其是在训练数据有限的情况下。
肿瘤(癌症)患者之家