The oil palm is a globally vital crop for vegetable oil production, particularly crucial for Southeast Asia's economy. However, the sustainability of oil palm plantations in this region is under significant threat from Ganoderma disease, or Basal Stem Rot. The mechanisms driving the spread of this disease remain poorly understood, and existing management methods have proven ineffective. Consequently, there is a pressing need for monitoring and implementing effective disease management techniques in oil palm plantations. The aim of this study was to utilize time series of ALOS2 PALSAR2 dual polarization SAR imagery over 9 years to identify Ganoderma infected oil palm plants incorporating the XGBoost machine learning model. This study proposes a pipeline starting from: collecting geospatial information from field observations of infected plants in oil palm plantations using a smartphone-based application; examine the potential backscatter variables for infected plants using time series of PALSAR2 measurements, utilizing SAR imagery and the XGBoost machine learning model to predict the Ganoderma-infected plants; and visualizing all the field observations and machine learning-based predictions through a web GIS-based dashboard. The data collected from the smartphone-based application were used as ground truth data to train the machine learning model. It comprises information on 138 trees, consisting of 73 healthy trees and 65 infected trees. The results proved that Horizontal transmit and Vertical receive (HV) component provides the highest accuracy which is 76.2% for identifying infected plants and 70.8% for identifying healthy plants. In conclusion, this study examines the capacity to detect infected plants using ALOS2 PALSAR2 SAR imagery and enhance the visualization of results, facilitating a clearer understanding of disease dynamics within large-scale plantation areas.