In the heart of Ilam Province, Iran, a silent battle is being waged against Zoonotic Cutaneous Leishmaniasis (ZCL), a vector-borne disease that has long evaded precise prediction and control. But a groundbreaking study, led by Fatemeh Parto Dezfooli from the Department of Photogrammetry and Remote Sensing at K N Toosi University of Technology in Tehran, is changing the game. By integrating remote sensing, GIS, and machine learning, Dezfooli and her team have developed a hybrid model that promises to revolutionize how we understand and combat ZCL.
ZCL, characterized by its distinct spatiotemporal patterns, has long posed a challenge to public health officials and researchers alike. Traditional methods of risk assessment have often fallen short, lacking the precision and comprehensiveness needed to effectively combat the disease. However, Dezfooli’s research, published in the journal *Environmental Research Communications* (translated to English as *Communications in Environmental Research*), is bridging this gap.
The study leverages the strengths of three powerful technologies: geographic information systems (GIS) for trend analysis, remote sensing (RS) for environmental data extraction, and machine learning (ML) for risk assessment. “By combining these technologies, we can gain a more holistic understanding of ZCL’s behavior and spread,” Dezfooli explains. The team utilized data from 2014 to 2019, employing Moran’s I, Getis-Ord Gi* statistics, and the Mann-Kendall (MK) test to investigate spatial and temporal patterns. High-risk ZCL maps were generated through Extreme Gradient Boosting (XGBoost) and Random Forest (RF) models, offering unprecedented accuracy and insight.
The results are striking. The study revealed significant patterns, with a Moran’s I statistic of 0.68 (p < 0.01) and MK values of –2.254 for annual data (p = 0.024) and 3.340 for monthly data (p = 0.001). Temporal analysis indicated a declining trend, with peak incidence observed in late fall and early winter. "This suggests that the critical infection period occurs during summer, due to the incubation period," Dezfooli notes. The risk maps demonstrated high levels of accuracy, with an area under the curve of 0.96 for RF and 0.98 for XGBoost, pinpointing high-risk areas in the western and southern hot deserts and low-risk regions in the northeastern mountainous areas. The implications of this research extend far beyond the realm of public health. For the energy sector, understanding the geographical distribution of diseases like ZCL is crucial. Energy infrastructure, particularly in remote or rural areas, can be significantly impacted by the spread of vector-borne diseases. Accurate risk assessment can inform the placement of new infrastructure, the implementation of protective measures, and the allocation of resources for disease control and prevention. Moreover, the integration of remote sensing, GIS, and machine learning offers a blueprint for future research and development. "This hybrid model can be adapted and applied to a wide range of diseases and environmental challenges," Dezfooli says. As we grapple with the complexities of climate change, urbanization, and global health, the need for robust, data-driven insights has never been greater. In the fight against ZCL and other vector-borne diseases, this research marks a significant step forward. By harnessing the power of technology and data, we can gain a deeper understanding of these diseases and develop more effective strategies for their control and prevention. As Dezfooli's work demonstrates, the future of public health lies in the integration of diverse technologies and the collaborative effort of researchers across disciplines.