Ethnomedicinal knowledge and geographic patterns of medicinal plants used for Leishmaniasis management in Southeastern Morocco: Integrating GIS, multivariate statistics, and machine learning

Abstract

Background: Leishmaniasis remains a major public health concern in Morocco, where medicinal plants continue to play an important role in traditional healthcare. This study aimed to document the medicinal plants used for leishmaniasis management in the Draa-Tafilalet region and to evaluate the influence of sociodemographic and geographical factors on ethnomedicinal knowledge and practices.

Methods: A cross-sectional ethnobotanical survey was conducted across the main provinces of the region using structured questionnaires in 2023. Correspondence analysis, Geographic Information System (GIS)-based mapping, and machine-learning approaches were applied to characterize spatial variation and identify the factors influencing medicinal plant selection.

Results: Women represented 58.62% of respondents, while participants aged ≤22 and 23-38 years each accounted for 44% of the sample. Most respondents were single (77%), economically active (81.6%), and university-educated (78%). However, 59% reported no knowledge of medicinal plants, and 64% were unable to identify any medicinal species. Further, 61 medicinal plant species were documented, with Rosmarinus officinalis (18 citations), Lawsonia inermis (12), and Origanum compactum (8) being the most frequently reported. Errachidia recorded the highest plant diversity (29.03%), whereas no medicinal plants were reported from Rissani. Leaves (40%) and decoction (46%) were the predominant plant part and preparation method, respectively. GIS and correspondence analyses revealed pronounced geographical heterogeneity in medicinal plant use among provinces. Machine Learning Identified High- and Low-Priority Medicinal Plants.

Conclusions: This study represents the first comprehensive ethnobotanical assessment covering the entire Draa-Tafilalet region, integrating GIS, multivariate, and machine learning analyses to identify spatial and sociodemographic determinants of medicinal plant utilization. These findings provide a valuable foundation for biodiversity conservation, ethnopharmacological research, and the development of affordable plant-based strategies for leishmaniasis management.

Keywords: Leishmaniasis; Ethnobotany; Medicinal plants; Drâa-Tafilalet; Geographic Information System (GIS); machine learning; traditional knowledge.