The Taklamakan Desert, the primary source of sand and dust in China, has a significant impact on PM_2.5 concentrations in the urban belt around the Tarim Basin. In this study, moderate-resolution imaging spectroradiometer (MODIS) images and machine learning algorithms were used to simulate PM_2.5 concentrations and analyze deaths from ischemic heart disease (IHD), chronic obstructive pulmonary disease (COPD), lung cancer (LC), and stroke (STK) attributed to PM_2.5. From 2015 to 2024, the aerosol optical depth (AOD) showed a decreasing trend, and the spatial distribution pattern gradually increased from west to east. The counties and cities with higher average PM_2.5 concentrations during the 2024 dust-prone season were Awati County, Alar City, Moyu County, and Aksu City, with average concentrations of 89.7 μg/m³, 88.8 μg/m³, 86.0 μg/m³, and 84.1 μg/m³, respectively. PM_2.5 concentrations during the non-dust season were significantly lower (approximately 50%) than those during the dust-prone season. Among the four major fatal diseases, STK and IHD accounted for the majority of attributable deaths, with proportions exceeding 80% of the total. The results show that PM_2.5 concentrations can be accurately monitored over a large scale, addressing the shortage of fixed monitoring stations, and providing a theoretical basis for sustainable development of ecological environment, health-risk management and control in the urban belt around the Tarim Basin.