FreqYOLO:基于本地和全球频率特征学习的子宫疾病检测网络
Ziying Huang1, Shuangshuang Lin2, Kedan Liao2
1School of Biomedical Engineering, Southern Medical University, Guangzhou, Guangdong 510515, China.
FreqYOLO是一种新的AI方法,可以在超声波图像中准确地检测和区分子宫瘤 (LM) 和腺瘤 (AM). 这一进步有助于对这些常见的妇科疾病进行更好的诊断和治疗规划.
科学领域:
- 医疗成像医学成像
- 人工智能的人工智能
- 妇科 妇科医生 妇科
背景情况:
- 子宫乳腺瘤 (LM) 和腺瘤 (AM) 是影响妇女的流行妇科疾病,经常呈现出各种病变特征.
- 在超声波中精确检测和区分LM和AM对于有效治疗至关重要,但由于疾病异质性而具有挑战性.
研究的目的:
- 开发和评估一种基于人工智能的方法,FreqYOLO,用于在超声波图像中更好地检测和区分LM和AM.
- 利用频率特征学习来提高妇科超声分析对象检测模型的性能.
主要方法:
- 提出FreqYOLO,这是一个深度学习模型,使用具有全球和本地频率特征的双分支特征编码器.
- 实施了用于多尺度特征融合的融合,以丰富频率信息.
- 采用了改进的抑制方法,以获得最佳的检测输出.
主要成果:
- FreqYOLO实现了0.734的召回率,0.795的精度,0.763的F1得分,0.788的AP50和0.487的mAP.
- 该方法在检测和区分LM和AM方面表现出优异的性能,与现有的最先进的技术相比.
结论:
- FreqYOLO显示出在超声波中提高LM和AM检测精度的巨大潜力.
- 频率特征学习方法为改善人工智能驱动的妇科诊断工具提供了一个有希望的方向.
更多相关视频
07:54Surface-enhanced Resonance Raman Scattering Nanoprobe Ratiometry for Detecting Microscopic Ovarian Cancer via Folate Receptor Targeting
Published on: March 25, 2019
08:04Fully Automated Leg Tracking in Freely Moving Insects using Feature Learning Leg Segmentation and Tracking (FLLIT)
Published on: April 23, 2020
相关概念视频
Frequency-dependent Selection
IR Frequency Region: Fingerprint Region
The...
Determination of Expected Frequency
Linear Approximation in Frequency Domain
In contrast, nonlinear systems do not inherently possess these properties. However, for small deviations around an operating point, a nonlinear system can often be approximated as linear.
Uterine Tubes
Ultrasound II: Endoscopic Ultrasound and FibroScan
Endoscopic Ultrasound (EUS):
