Landslides are one of the most widespread, yet difficult to predict, geohazards in the world. Theyrepresent a major threat to society, which is why it is so important to understand and study them.Landslide detection is one of the approaches that help scientists to gain a better knowledge of landslides,how they form and the characteristics they hold. In this work, a new landslide detection technique istested in the Klarälven Region, Sweden. The method involves using multi-temporal DEMs generatedfrom LiDAR point clouds from 2010 and 2020. Once two respective DEMs are generated from theoriginal point cloud data sets, a DEM of Difference (DoD) is obtained by subtracting the 2010 DEMfrom the 2020 DEM. The DoD highlights areas of vertical change over the ten-year period. In order todetect potential landslides, a Level of detection (LoD) threshold is applied, as well as a second upperthreshold filtering out elevation difference resulting from anthropogenic activities. Lastly, a spatialthreshold is applied to eliminate any remaining potential errors clusters that are too small to belandslides. Once this stage is completed, the study area is surveyed, focusing on certain areacharacterised by particular slope angles and soil types most likely to exhibit mass movement. Thishelped in the analysis and interpretation of the data. In order to discuss the results as fully as possible,four types of region were studied in more detail using four representative examples. Since a few doubtfulcases were found, potential sources of error are also being explored. Despite minor inaccuracies, themethod appears to be a relevant technique for detecting differences in altitude, even if it also considersunnatural mass movements.