多中心3D CNN用于帕金森病的诊断和预后,使用临床和T1加权的MRI数据
Silvia Basaia1, Elisabetta Sarasso2, Francesco Sciancalepore3
1Neuroimaging Research Unit, Division of Neuroscience, IRCCS San Raffaele Scientific Institute, Milan, Italy.
NeuroImage. Clinical
|August 7, 2025
概括
卷积神经网络 (CNN) 在诊断帕金森病 (PD) 和使用MRI扫描预测其进展方面表现有前途. 这项研究开发了一个3D-CNN模型,在区分PD患者和预测疾病进展方面实现了超过70%的准确性.
科学领域:
- 神经成像是一种神经成像.
- 机器学习 机器学习
- 神经学 神经学
背景情况:
- 帕金森病 (PD) 的诊断和进展预测仍然具有挑战性.
- 机器学习,特别是卷积神经网络 (CNN),提供了利用神经成像数据提高准确性的潜力.
研究的目的:
- 使用MRI开发一个3D-CNN来区分对照组和PD患者.
- 利用CNN来预测PD的进展情况.
- 评估转移学习对模型性能的影响.
主要方法:
- 三组PD患者和对照组接受了MRI扫描和临床评估.
- 根据UDPRS-III得分,K-means集群确定了两个PD进展组.
- 一个3D-CNN模型被训练用于二进制分类和进展预测,有或没有转移学习.
主要成果:
- 3D-CNN将中度至重度的PD与单独使用MRI的对照区分开来,准确度为74%.
- 转移学习改善了轻度PD与对照分类的准确度,达到64%.
- 结合MRI和临床数据,该模型以超过70%的准确度预测了PD进展.
结论:
- 集成多式联网数据和转移学习的CNN显示了早期PD分类和监测的潜力.
- 通过激活地图来解释AI可以帮助临床应用在早期诊断和个性化患者管理中.
相关概念视频
Parkinson's Disease: Overview
708
Neurodegenerative disorders are progressive diseases that cause irreversible damage and loss to neurons in specific brain areas. Examples of these disorders include Parkinson's disease, Alzheimer's disease, Multiple Sclerosis (MS), and Amyotrophic Lateral Sclerosis (ALS). These disorders share characteristics such as proteinopathies, selective neuronal vulnerability, and a complex interplay between genetic and environmental factors. The primary therapeutic goal for these conditions is...
708
Parkinson's Disease: Treatment
380
Neurodegenerative disorders, such as Parkinson's Disease (PD), involve the gradual and irreversible destruction of neurons in particular brain areas. These disorders exhibit standard features like proteinopathies, selective vulnerability of some neurons, and an interaction of intrinsic properties, genetics, and environmental influences in neural injury.
Parkinson's Disease is primarily a result of the loss of dopaminergic neurons in the substantia nigra pars compacta. The cornerstone of...
Parkinson's Disease is primarily a result of the loss of dopaminergic neurons in the substantia nigra pars compacta. The cornerstone of...
380


