Institute for Intelligent Systems Research and Innovation, Deakin University, Geelong, Victoria 3216 Australia.
CSIRO Health & Biosecurity, Australian Centre for Disease Preparedness, Geelong, Victoria 3220 Australia.
Acta Trop. 2024 Oct;258:107347. doi: 10.1016/j.actatropica.2024.107347. Epub 2024 Aug 3.
Mosquito-borne diseases such as malaria, dengue, Zika, and chikungunya cause significant morbidity and mortality globally, resulting in over 600,000 deaths from malaria and around 36,000 deaths from dengue each year, with millions of people infected annually, leading to substantial economic losses. The existing mosquito control measures, such as long-lasting insecticidal nets (LLINs) and indoor residual spraying (IRS), helped to reduce the infections. However, mosquito-borne diseases are still among the deadliest diseases, forcing us to improve the existing control methods and look for alternative methods simultaneously. Advanced monitoring techniques, including remote sensing, and geographic information systems (GIS) have significantly enhanced the efficiency and effectiveness of mosquito control measures. Mosquitoes' behavioural traits, such as locomotion, blood-feeding, and fertility are the key determinants of disease transmission and epidemiology. Technological advancements, such as high-resolution cameras, infrared imaging, and artificial intelligence (AI) driven object detection models, including groundbreaking convolutional neural networks, have provided efficient and precise options to monitor various mosquito behaviours, including locomotion, oviposition, fertility, and host-seeking. However, they are not commonly employed in mosquito-based research. This review highlights the novel and significant advancements in behaviour-monitoring tools, mostly from the last decade, due to cutting-edge video monitoring technology and artificial intelligence. These advancements can offer enhanced accuracy, efficiency, and the ability to quickly process large volumes of data, enabling detailed behavioural analysis over extended periods and large sample sizes, unlike traditional manual methods prone to human error and labour-intensive. The use of behaviour-assaying techniques can support or replace existing monitoring techniques and directly contribute to improving control measures by providing more accurate and real-time data on mosquito activity patterns and responses to interventions. This enhanced understanding can help establish the role of behavioural changes in improving epidemiological models, making them more precise and dynamic. As a result, mosquito management strategies can become more adaptive and responsive, leading to more effective and targeted interventions. Ultimately, this will reduce disease transmission and significantly improve public health outcomes.
蚊媒疾病，如疟疾、登革热、寨卡病毒和基孔肯雅热，在全球范围内造成了显著的发病率和死亡率，每年导致超过 60 万人死于疟疾，约 3.6 万人死于登革热，每年有数百万人感染，导致巨大的经济损失。现有的蚊子控制措施，如长效驱虫蚊帐（LLINs）和室内滞留喷洒（IRS），有助于减少感染。然而，蚊媒疾病仍然是最致命的疾病之一，迫使我们改进现有的控制方法，同时寻找替代方法。先进的监测技术，包括遥感和地理信息系统（GIS），极大地提高了蚊子控制措施的效率和效果。蚊子的行为特征，如运动、吸血和繁殖，是疾病传播和流行病学的关键决定因素。技术进步，如高分辨率相机、红外成像和人工智能（AI）驱动的物体检测模型，包括开创性的卷积神经网络，为监测各种蚊子行为提供了高效和精确的选择，包括运动、产卵、繁殖和寻找宿主。然而，它们在基于蚊子的研究中并不常用。本综述强调了行为监测工具的新的和重要的进展，主要来自过去十年，由于尖端的视频监测技术和人工智能。这些进展可以提供更高的准确性、效率和处理大量数据的能力，能够在较长时间内和大样本量上进行详细的行为分析，与传统的易出错和劳动密集型的手动方法不同。行为分析技术的使用可以支持或替代现有的监测技术，并通过提供关于蚊子活动模式和对干预措施的反应的更准确和实时的数据，直接有助于改善控制措施。这种增强的理解有助于确定行为变化在改善流行病学模型中的作用，使它们更精确和动态。因此，蚊子管理策略可以变得更加适应和响应，从而实现更有效和有针对性的干预。最终，这将减少疾病传播，并显著改善公共卫生结果。
