优化基于CNN的膝关节骨关节炎诊断:通过CleanLab重新标记来提高模型准确性
Thomures Momenpour1, Arafat Abu Mallouh1
1Department of Computer Science, Manhattan University, Riverdale, NY 10471, USA.
Diagnostics (Basel, Switzerland)
|June 13, 2025
概括
这项研究增强了使用EfficientNetB5和Cleanlab进行精确的Kellgren-Lawrence分级的膝关节骨关节炎分类. 改进的模型对关节疾病严重程度的客观临床评估有希望.
科学领域:
- 医疗成像医学成像
- 人工智能的人工智能
- 整形外科 整形外科 整形外科
背景情况:
- 膝关节关节炎 (KOA) 显著影响生活质量,特别是在老年人群中.
- 准确的Kellgren-Lawrence (KL) 评分对于KOA管理至关重要,但受到主观性和可变性的影响.
- 需要客观的,自动化的方法来克服传统KOA严重程度分类的局限性.
研究的目的:
- 评估EfficientNetB5深度学习模型,将KOA严重程度分为五个KL等级 (0-4).
- 通过以数据为中心的预处理,包括移除异常值和标签纠正,提高KOA分类的准确性.
- 为客观的KOA评估建立可靠的自动化系统.
主要方法:
- 使用Kaggle数据集,其中包括9786张膝盖X射线图像.
- 采用了一个EfficientNetB5模型,并从ImageNet.Net中转移学习.
- 实施了数据预处理管道,包括异常值删除和Cleanlab用于标签校正.
主要成果:
- EfficientNetB5模型的准确度达到82.07%,超过了之前的基准标准,如ResNet-101 (69%).
- 数据预处理,特别是Cleanlab的标签校正,显著提高了模型性能.
- 在健康和严重的KOA之间观察到很强的区别,尽管KL等级1仍然具有挑战性.
结论:
- 将EfficientNetB5与基于Cleanlab的数据预处理集成为KOA严重程度分类提供了一个强大的,准确的方法.
- 该管道显示了KOA评估中客观临床应用的潜力.
- 对于模两可的病例 (KL级1) 和严重的KOA样本大小,需要进一步研究.
更多相关视频
09:47Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
Published on: December 15, 2023
998
07:22Glycemic Impact on Knee Osteoarthritis Symptoms on Physical, Radiographic, and Inflammatory Markers among Individuals Aged 50 and Over with Diabetes
Published on: March 7, 2025
212
相关概念视频
02:07Improving Translational Accuracy
Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...
02:07Improving Translational Accuracy
Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...