通过分割联合学习和GAN用于不平衡数据集来增强阿尔茨海默病的分类
G Narayanee Nimeshika1, Subitha D1
1School of Computer Science and Engineering, Vellore Institute of Technology, Chennai, Tamil Nadu, India.
PeerJ. Computer science
|December 9, 2024
概括
这项研究引入了一种新的方法来检测阿尔茨海默病,使用分割联合学习 (SFL) 和条件生成对抗网络 (cGAN) 来解决医疗AI中的数据不平衡和隐私问题.
科学领域:
- 医疗信息学 医疗信息学
- 医疗保健中的人工智能
- 计算神经科学是一种神经科学.
背景情况:
- 医疗保健数据不平衡和隐私问题阻碍了准确的医学分类.
- 现有的模型与像阿尔茨海默氏症这样的疾病的分散和不平衡的数据集作斗争.
- 先进的技术对于提高诊断准确性和患者护理至关重要.
研究的目的:
- 开发一个保护隐私的医疗分类模型,用于阿尔茨海默病的检测.
- 解决医疗AI中不平衡数据集和数据去中心化的挑战.
- 提高AI模型在临床环境中的概括能力.
主要方法:
- 利用分拆联合学习 (SFL) 进行分散型模型培训,而无需共享数据.
- 整合条件生成对抗网络 (cGANs) 合成少数阶级的现实数据.
- 开发了一种混合方法,将SFL和cGAN结合起来,用于强大的阿尔茨海默病分类.
主要成果:
- 在阿尔茨海默病分类中达到约83.54%的准确性.
- 从分散和不平衡的医疗数据集中展示了有效的学习.
- 通过联合学习框架成功保护了患者数据隐私.
结论:
- 拟议的SFL和cGANs模型有效地解决了数据隐私和医疗分类不平衡问题.
- 这种方法提高了阿尔茨海默病的诊断能力,改善了潜在的患者结果.
- 该方法提供了一个可扩展的解决方案,用于在尊重数据保护法规的同时在医疗保健中开发AI.
相关概念视频
Alzheimer's Disease: Treatment
163
Alzheimer's Disease (AD), a neurodegenerative disorder, is pathologically identified by amyloid plaques and neurofibrillary tangles composed of tau protein. AD pharmacotherapy aims to manage cognitive symptoms, delay disease progression, and treat behavioral symptoms. The treatment is primarily symptomatic and palliative, with no definitive disease-modifying therapy available. Cholinesterase inhibitors, including donepezil (Aricept), rivastigmine (Exelon), and galantamine (Razadyne), are...
163
Alzheimer's Disease: Overview
447
Alzheimer's Disease (AD) is a continually advancing neurodegenerative disorder, distinguished by escalating memory loss, cognitive dysfunction, and dementia. The disease unfolds in three stages: preclinical, mild cognitive impairment (MCI), and dementia. Its onset is insidious, and the progression gradual, with the cause not well explained by other disorders.
The clinical diagnosis of AD hinges on the presence of memory and other cognitive impairments. Biomarkers, such as changes in Aβ...
The clinical diagnosis of AD hinges on the presence of memory and other cognitive impairments. Biomarkers, such as changes in Aβ...
447


