一个可解释和准确的深度学习诊断框架,以完全和半监督的相互学习为模型
IEEE transactions on medical imaging
|August 21, 2023
概括
一个新的深度学习框架,InterNRL,在临床诊断中实现了高准确性和可解释性. 这种人工智能模型改善了疾病检测和定位,有利于医学图像分析.
科学领域:
- 人工智能在医学中的应用
- 机器学习用于医疗保健
- 医学图像分析 医学图像分析
背景情况:
- 深度学习分类器可以改善临床诊断,但在准确性和可解释性之间面临着权衡.
- 现有的准确模型往往缺乏透明度,阻碍了临床采用.
- 可解释模型可能无法实现具有竞争力的诊断性能.
研究的目的:
- 引入InterNRL,这是一个新的深度学习框架,旨在在临床诊断中实现高准确性和可解释性.
- 解决当前深度学习模型在平衡分类性能和可解释性方面的局限性.
- 提高AI在医疗决策中的可靠性和可信度.
主要方法:
- 开发了InterNRL,这是一个学生-教师框架,使用可解释的基于原型的分类器 (ProtoPNet) 作为学生和全球图像分类器 (GlobalNet) 作为教师.
- 实施了一种新的相互学习范式,以实现学生和教师模型之间的相互优化.
- 在完全和半监督学习场景下优化了框架.
主要成果:
- 在两个监督学习环境下,InterNRL在乳腺癌和视网膜疾病诊断方面取得了最先进的分类表现.
- 该框架在乳腺癌局部化和脑瘤细分方面表现出卓越的性能,使用标签薄弱的图像.
- 相互学习方法使有效的知识转移和模型改进成为可能.
结论:
- 在临床诊断的深度学习中,InterNRL成功地整合了高精度和可解释性.
- 拟议的相互学习模式为训练准确和可解释的医疗AI模型提供了一种灵活和有效的方法.
- 在临床实践中,InterNRL显示了在推进自动诊断和医学图像分析方面显著的潜力.
相关概念视频
Improving Translational Accuracy
11.6K
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...
11.6K
Classification of Illness
7.6K
The meaning of illness is individualized to each person who experiences an alteration in health. In contrast, disease is a medical term indicating a pathological change in the structure and function of the body or mind. It is a condition that has specific symptoms and boundaries.
An illness is a response to a disease in which the person's level of functioning is changed compared with a previous level. The general classification of illness includes acute and chronic.
Acute illness is severe...
An illness is a response to a disease in which the person's level of functioning is changed compared with a previous level. The general classification of illness includes acute and chronic.
Acute illness is severe...
7.6K


