通过利用混合监督和自我和转移学习 (MIST) 来实现医疗图像细分的统一方法
Jianfei Liu1, Sayantan Bhadra1, Omid Shafaat1
1Radiology and Imaging Sciences, National Institutes of Health Clinical Center, 10 Center Dr, Bethesda, 20892, MD, USA.
概括
使用MIST (混合监督,自我和转移学习) 医疗图像细分的准确性得到了显著改善. 这种方法大大减少了手工标签,节省了时间和资源,同时提高了诊断能力.
科学领域:
- 放射学和医学成像学 医学成像学
- 医疗保健中的人工智能
- 计算解剖学的计算解剖学
背景情况:
- 准确的医学图像细分对于定量疾病诊断和治疗规划至关重要.
- 手动的像素智能标签是劳动密集型,耗时,需要专门的专业知识,在医学图像分析中造成瓶.
- 现有的自动化方法往往缺乏临床应用所需的精度.
研究的目的:
- 引入MIST (混合监督,自我和转移学习),这是一种新的框架,可以显著减少医疗图像细分中的手动注释工作.
- 开发一种利用有限的强标签和丰富的弱标签来提高细分精度的方法.
- 为了证明MIST在CT扫描中的脂肪组织区分中的有效性.
主要方法:
- MIST采用双分支网络,具有共享的编码器和两个解码器,在手动注释 (强) 和自动生成 (弱) 标签上进行训练.
- 自我训练反复地改进弱标签,随着时间的推移提高其质量.
- 转移学习是通过结编码器和微调解码器来优化计算效率来利用的.
主要成果:
- 在内部和外部数据集上,MIST在多种组织类型 (肌肉,皮下,内脏脂肪组织) 的细分精度上取得了显著的改进,通过Dice相似系数 (DSC) 测量.
- 例如,皮下脂肪组织的DSC在内部数据集上从75.1%提高到94.2%,在外部数据集上从61.8%提高到82.7% (p<.05).
- MIST框架减少了99%的注释负担,使得大数据集 (102个扫描) 能够使用最小的手动输入 (100片) 来生成准确的标签.
结论:
- MIST有效地减少了医疗图像细分中的手动注释的需要,使其更有效和更容易获得.
- 拟议的方法显著提高了细分的准确性,优于在强弱标签组合上的直接培训.
- 通过提供高质量,自动生成的标签,MIST促进了像3D nnU-Net这样强大的模型的培训,从而推进了定量医学图像分析.
相关概念视频
Observational Studies
Observational studies are a type of analytical study where researchers observe events without any interventions. In other words, the researcher does not influence the response variable or the experiment's outcome.
There are three types of observational studies – Prospective, retrospective, and cross-sectional.
Prospective Study
Prospective studies, also known as longitudinal or cohort studies, are carried out by collecting future data from groups sharing similar characteristics. One example of...
There are three types of observational studies – Prospective, retrospective, and cross-sectional.
Prospective Study
Prospective studies, also known as longitudinal or cohort studies, are carried out by collecting future data from groups sharing similar characteristics. One example of...
Methods of Documentation VII: EMR
Electronic Medical Records (EMRs) primarily center around electronically documenting patients' health information within a single healthcare organization or practice. They contain essential clinical data related to a patient's medical history, diagnoses, medications, treatment plans, lab results, and other pertinent information relevant to the specific encounter or episode of care. EMRs are designed to streamline documentation and workflow processes within individual healthcare settings,...
Documentation in Long-Term and Home Healthcare Setting
Documentation in long-term care facilities and home healthcare settings is crucial for ensuring continuous, coordinated, and comprehensive care for patients. Each setting has its specific documentation processes and tools:
Long-Term Care Facilities
Long-Term Care Facilities
Introduction to Epidemiology
Epidemiology, known as the cornerstone of public health, involves studying the distribution and determinants of health-related events in defined populations and applying these insights to control health issues. This is essential for understanding how diseases spread, identifying populations at greater risk, and implementing measures to control or prevent outbreaks. Epidemiology addresses not only infectious diseases but also non-communicable conditions like cancer and cardiovascular disease,...
Pulmonary Embolism II: Diagnostic Studies and Interprofessional Care
Diagnosing Pulmonary EmbolismDiagnosing pulmonary embolism (PE) involves clinical assessment and advanced imaging tests. The preferred diagnostic tool is the spiral (helical) CT scan or CT angiography (CTA), which uses intravenous contrast media to visualize the pulmonary vasculature and identify emboli.A ventilation-perfusion (V/Q) scan is an alternative for patients unable to receive contrast media. This scan includes both perfusion and ventilation scanning. Perfusion scanning involves...


