系统地运用突出度图来解释卷积神经网络的决定,以基于磁盘和杯子几何学来诊断青光眼
Francisco Fumero1, Jose Sigut1, José Estévez1
1Department of Computer Science and Systems Engineering, Universidad de La Laguna, Camino San Francisco de Paula, 19, La Laguna, 38203, Santa Cruz de Tenerife, Spain.
Biomedical physics & engineering express
|January 10, 2025
概括
这项研究评估了用简化眼底图像诊断青光眼的突出性方法. 结果显示可解释性图的变化,这表明医疗应用需要谨慎.
科学领域:
- 眼科医生 眼科 眼科
- 人工智能的人工智能
- 医疗成像医学成像
背景情况:
- 卷积神经网络 (CNN) 显示出从眼底图像诊断青光眼的前景.
- 解释性工具,如突出性方法,对于理解关键医疗应用中的CNN决策至关重要.
- 目前的突出性方法可能不总是与对青光眼诊断的专家医学标准保持一致.
研究的目的:
- 系统地评估突出性方法作为可解释性工具,用于诊断青光眼的CNN.
- 通过突出地图突出显示的特征与专家使用的几何线索联系起来.
- 评估医学图像分析中的突出性方法的可靠性和一致性.
主要方法:
- 为了CNN的训练,利用了简化的眼底图像,只包含盘和杯的轮.
- 采用了九种不同的突出方法来生成归因图.
- 应用了离散算法来规范归因值,并分析了与专家识别的部门的相关性.
主要成果:
- 在简化图像上训练的CNN实现了高分类准确度 (0.83310.8890).
- 在突出地图中观察到显著的变化,受方法,模型和架构的影响.
- 总体结果显示稳定性与数据集之间的强相关性 (0.9289与RIM-ONE DL,0.7806与REFUGE).
结论:
- 度方法在医学成像中表现出相当大的变化,需要在临床使用中谨慎使用.
- 这些方法可能更适合于广泛的图像相关性解释,而不是个别案例评估.
- 识别的网络区域并不总是与已确立的医疗标准保持一致,以确定青光眼的严重程度.
相关概念视频
01:25Glaucoma: Overview
502
Glaucoma is an eye condition characterized by increased intraocular pressure that damages the retina and optic nerve, leading to irreversible blindness if left untreated. The human eye has various components, including the cornea, iris, pupil, lens, and optic nerve. Aqueous humor is secreted by the epithelium of the ciliary body in the posterior chamber and flows through the trabecular meshwork and canal of Schlemm, maintaining normal intraocular pressure. The trabecular meshwork and the canal...
502
01:28Angle Closure Glaucoma: Treatment
434
Angle-closure glaucoma, or closed-angle glaucoma, is an eye condition where the iris bulges out and blocks the iridocorneal angle, resulting in a buildup of aqueous humor and increased intraocular pressure. Immediate medical attention is necessary due to the sudden onset of symptoms. The treatment for angle-closure glaucoma includes short-term and long-term approaches. Short-term treatment involves using eye drops like pilocarpine to lower intraocular pressure by increasing aqueous humor...
434
01:24Vision
52.9K
Vision is the result of light being detected and transduced into neural signals by the retina of the eye. This information is then further analyzed and interpreted by the brain. First, light enters the front of the eye and is focused by the cornea and lens onto the retina—a thin sheet of neural tissue lining the back of the eye. Because of refraction through the convex lens of the eye, images are projected onto the retina upside-down and reversed.
52.9K