Improving visual performance in retinitis pigmentosa by new non-invasive personalizedand adaptive training.

本研究旨在研究色素性视网膜炎（RP）患者的病理视觉行为，并寻求改善残余视觉表现的新方法。早期RP期周边视觉受损，中央视觉相对完整，视力良好。然而，后期RP阶段会影响中央视力和敏锐度。尽管周边视觉的分辨率很低，但对于创建和更新导航空间结构的准确表示非常重要，因此RP是一种损害移动性的条件。一种有希望的恢复视觉表现的方法是利用先天的神经可塑性和提供特定的眼动训练方案，帮助患者自然地“扩大”他们有限的视野（VF）。研究表明，大脑神经的可塑性可能涉及到整体认知策略的改变，以成功应对新的挑战。在一项初步研究中，我们证明了神经可塑性在改善和部分恢复患者行走时的视觉表现方面的潜力。我们发现，训练患者对看不见的VF进行探索性扫视提高了患者的行走速度，但日常生活的其他方面，如与物体碰撞的次数并没有改善。这表明，进一步改进训练策略可能需要最终优化RP患者在日常生活中的导航。我们假设，为了最大限度地将训练效果转移到现实生活中，训练中的运动应该与典型的日常任务相似。目前，这可以通过虚拟现实（VR）应用来实现，还有一个额外的好处，即患者可以锻炼走路和处理不同的情况，而不必暴露在日常生活的危害中。我们的新方案旨在进行有效的训练，以补偿在隧道视觉中落在有限功能VFs之外的区域中同时跟踪多个移动目标的损失：它将指导患者以系统扫描模式（SSP）快速补偿性凝视运动来执行，以便创建场景的全局框架并识别可能的目标。此外，可能的目标的动态扫描跟踪（DST）将被实践，这将只需要一个焦点的关注，将快速循环通过目标，将索引他们的位置，并将返回到每一个之前，它移动得太远。不属于该项目范围，但相关的是将结果转移到其他导致视野丧失的疾病，例如偏盲。我们提出，基于我们的初步工作，神经可塑性训练可能普遍适用于目前不可逆的视网膜或神经元损伤的一系列眼病。

The aim of this proposal is to study the pathological visual behavior in patients with retinitis pigmentosa (RP) and to implement new methods able to improve residual visual performance. In early RP stages, peripheral vision is damaged, while central vision is relatively intact with good visual acuity. Later RP stages, however, impact central vision and acuity. Peripheral vision, despite its low resolution, is important for creating and updating an accurate representation of spatial structure for navigation and therefore RP is a condition that impairs mobility. One promising approach to recover visual performance, is to use the innate neural plasticity and provide specific eye-movement training protocol that helps patients to naturally "expand" their restricted visual field (VF). It has been shown that brain neural plasticity may involve modifications in overall cognitive strategies to successfully cope with new challenges. In a preliminary work we demonstrated the potential of neural plasticity to improve and to partially recover the patients’ visual performance during walking. We found that training the patients to perform explorative saccades to the non-seeing VF improved walking speed of patients, but other aspects of daily life, such as number of collisions with objects did not improve. This suggests that further, improved training strategies might be required to ultimately optimize navigation of RP patients in daily life. We hypothesize that to maximize transfer of training effect in real-life, exercise during training should resemble closely typical daily tasks. This could currently be achieved via virtual reality (VR) applications, with the added benefit that the patients could exercise to walk and deal with different situations without being exposed physically to the hazards of daily life. Our new protocol aims at efficient training that allows to compensate for the loss of simultaneous tracking of multiple moving targets in areas falling outside the limited functional VFs in tunnel vision: It will instruct the patients to perform in a systematic scanning pattern (SSP) fast compensatory gaze movements, in order to create a global framework of the scene and identify possible targets. Further, dynamic scanning tracking (DST) of possible targets will be practiced, which will require only one focus of attention that will cycle rapidly through the targets, will index their locations and will return to each before it moves too far away. Not in the scope of the project, but relevant is the transfer of the results to other diseases with visual field loss, e.g. hemianopia. We propose, based on our preliminary work, that neural-plasticity training may be generally applicable in a range of eye conditions with currently irreversible retinal or neuronal damage.