基于超声波图像特征的高风险无症状带斑块预测中的可解释的人工智能
Nicoletta Prentzas1, Chara S Skouteli2, Efthyvoulos Kyriacou3
1Department of Computer Science and Biomedical Engineering Research Center, University of Cyprus, Nicosia, Cyprus - nicolep@ucy.ac.cy.
概括
可解释性AI (XAI) 方法改善了从动脉狭窄数据中预测中风风险. 基于论证的可解释机器学习 (ArgEML) 提高了模型透明度和临床决策,而不会牺牲准确性.
科学领域:
- 医疗人工智能
- 可解释的AI (XAI)
- 冠状动脉狭窄的成像
背景情况:
- 使用斑块纹理和临床特征的支持矢量机 (SVM) 模型可以改善无症状大脑动脉狭窄 (ACS) 的中风预测.
- 像SVM这样的黑盒人工智能模型缺乏透明度, 阻碍了临床采用.
- 可解释人工智能 (XAI) 为关键医疗应用提供了可解释人工智能的途径.
研究的目的:
- 研究XAI技术的整合以提高中风风险评估的可解释性.
- 评估XAI是否可以在不影响ACS预测准确性的情况下提高模型的透明度.
主要方法:
- 开发基于论证的可解释机器学习 (ArgEML) 方法和框架.
- 应用ArgEML来从无症状动脉狭窄症患者的现实数据中学习可解释的论证理论.
- 使用标准机器学习指标评估ArgEML性能,并通过模型透明度和解释质量评估可解释性.
主要成果:
- ArgEML模型的预测准确度很高,与传统方法相比.
- 使用ArgEML可以显著提高预测的可解释性.
- 不确定预测被认为是困境,为预测能力提供了有价值的见解.
结论:
- ArgEML有效地提高了从医学数据中预测中风风险的解释性.
- 预测性性能保持,证明了XAI在临床环境中的可行性.
- 通过ArgEML提供的透明度有助于改进模型,指导临床决策,并促进在医疗保健中采用人工智能.
相关概念视频
Imaging Studies for Cardiovascular System VI: Calcium -Scoring CT
119
Calcium-Scoring CT ScanA calcium-scoring CT scan, also known as coronary artery calcium (CAC) scan, detects calcium deposits in the coronary arteries. This test assesses the risk of coronary artery disease (CAD), which can lead to cardiovascular events such as angina, heart failure, and sudden cardiac arrest.A calcium-scoring CT scan is generally recommended for individuals at intermediate risk of CAD without symptoms. It includes:Men aged 40-75 and women aged 50-75: Especially those with a...
119
Atherosclerosis II: Clinical Manifestations and Diagnostic Tests
38
Atherosclerosis is a progressive disorder that leads to the thickening and narrowing of arterial walls due to plaque buildup. This condition can cause various symptoms depending on the arteries affected:Coronary Artery Disease (CAD): This condition affects the coronary arteries and may lead to chest pain (angina), shortness of breath (dyspnea), heart attacks, and other heart disease symptoms.Cerebrovascular Disease: This affects blood flow to the brain, causing transient ischemic attacks (TIAs)...
38


