PulseAugur
实时 08:53:45
English(EN) Momentum-Anchored Multi-Scale Fusion Model for Long-Tailed Chest X-Ray Classification

新模型锚定动量以改进长尾胸部X光片分类

研究人员开发了一个名为动量锚定多尺度融合网络(Momentum-Anchored Multi-Scale Fusion Network)的新模型,以解决胸部X光片分类中的类别不平衡问题。该模型使用指数移动平均来稳定特征表示,防止对常见病症产生偏见,并提高对罕见疾病的性能。在ChestX-ray14数据集上进行测试,该方法取得了0.8682的平均AUC,在疝气(Hernia)和肺炎(Pneumonia)等病理方面显示出显著的收益。 AI

影响 提高医学影像中罕见病症的诊断准确性,可能带来更早的检测和更好的患者预后。

排序理由 这是一篇发表在arXiv上的研究论文,详细介绍了一种用于医学图像分类的新模型。

在 arXiv cs.CV 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

新模型锚定动量以改进长尾胸部X光片分类

报道来源 [2]

  1. arXiv cs.CV TIER_1 English(EN) · Duy Hoang Khuong, Duy Nguyen Huu, Ngu Huynh Cong Viet ·

    Momentum-Anchored Multi-Scale Fusion Model for Long-Tailed Chest X-Ray Classification

    arXiv:2605.02292v1 Announce Type: new Abstract: Chest X-ray classification suffers from severe class imbalance where gradient updates bias toward majority classes, causing feature drift and poor performance on rare but critical pathologies. We propose a Momentum-Anchored Multi-Sc…

  2. arXiv cs.CV TIER_1 English(EN) · Ngu Huynh Cong Viet ·

    Momentum-Anchored Multi-Scale Fusion Model for Long-Tailed Chest X-Ray Classification

    Chest X-ray classification suffers from severe class imbalance where gradient updates bias toward majority classes, causing feature drift and poor performance on rare but critical pathologies. We propose a Momentum-Anchored Multi-Scale Fusion Network that uses exponential moving …