Richards-Kortum Rebecca, Lorenzoni Cesaltina, Bagnato Vanderlei S, Schmeler Kathleen
Department of Bioengineering, Rice University, Houston, TX, USA.
Institute for Global Health Technologies, Rice University, Houston, TX, USA.
Nat Rev Bioeng. 2024 Jan;2(1):25-43. doi: 10.1038/s44222-023-00135-4. Epub 2023 Dec 22.
Low-cost optical imaging technologies have the potential to reduce inequalities in healthcare by improving the detection of pre-cancer or early cancer and enabling more effective and less invasive treatment. In this Review, we summarise technologies for in vivo widefield, multi-spectral, endoscopic, and high-resolution optical imaging that could offer affordable approaches to improve cancer screening and early detection at the point-of-care. Additionally, we discuss approaches to slide-free microscopy, including confocal imaging, lightsheet microscopy, and phase modulation techniques that can reduce the infrastructure and expertise needed for definitive cancer diagnosis. We also evaluate how machine learning-based algorithms can improve the accuracy and accessibility of optical imaging systems and provide real-time image analysis. To achieve the potential of optical technologies, developers must ensure that devices are easy to use; the optical technologies must be evaluated in multi-institutional, prospective clinical tests in the intended setting; and the barriers to commercial scale-up in under-resourced markets must be overcome. Therefore, test developers should view the production of simple and effective diagnostic tools that are accessible and affordable for all countries and settings as a central goal of their profession.
低成本光学成像技术有潜力通过改善癌前病变或早期癌症的检测，并实现更有效、侵入性更小的治疗，来减少医疗保健方面的不平等。在本综述中，我们总结了用于体内宽视野、多光谱、内窥镜和高分辨率光学成像的技术，这些技术可以提供经济实惠的方法，以改善即时医疗点的癌症筛查和早期检测。此外，我们还讨论了免玻片显微镜检查方法，包括共聚焦成像、光片显微镜和相位调制技术，这些技术可以减少明确癌症诊断所需的基础设施和专业知识。我们还评估了基于机器学习的算法如何提高光学成像系统的准确性和可及性，并提供实时图像分析。为了实现光学技术的潜力，开发者必须确保设备易于使用；光学技术必须在预期环境中的多机构前瞻性临床试验中进行评估；并且必须克服资源匮乏市场中商业规模扩大的障碍。因此，测试开发者应将生产对所有国家和环境都可及且负担得起的简单有效的诊断工具视为其职业的核心目标。
