Macroinvertebrate biomass estimation is essential for answering ecological questions in aquatic ecosystems. Length-mass regression is the most common method used, as it is precise and fast compared to other methods (e.g., destructively drying specimens and measuring mass). However, these regressions are data depauperate because regression production is challenged by several factors. The traditional method of hand measuring individuals is time-intensive and costly, and precise measurements of individual length are essential to accurately predict mass. Additionally, the specific coefficients of regressions vary widely across taxon, and having regression data derived from specific taxa is crucial for getting accurate measurements. As such, the use of semi-automatic image analysis techniques is a promising avenue for increasing the efficiency and accuracy of aquatic macroinvertebrate biomass data. One such technique is the ZooScan measurement tool, which uses an industrial waterproof scanner to image organisms and artificial intelligence software to separate and measure individuals semi-automatically. The ZooScan has been used extensively for biomass estimates of marine zooplankton to the effect of improved efficiency and accuracy; freshwater macroinvertebrates present an underutilized application with similar implications. One potential benefit is the availability of novel metrics such as body area, which may be a better predictor of biomass as it captures morphological traits that body length may not. This project aimed to test the ZooScan tool as a novel image analysis technique against traditional hand measurement techniques by producing length-mass regressions for three underrepresented aquatic invertebrate families. The relationship between dry mass and body metric was assessed by fitting a power function to the data, and regression coefficients were compiled. Published compilations of length-mass regressions for aquatic macroinvertebrates exist but are spatially limited. There is evidence to indicate that published regressions do not hold up as distributions of genera shift within families across spatial gradients as well, demonstrating a need for regionally targeted length-mass regression compilations. Regional compilations of regression data, while feasible with traditional hand-measurement techniques, are made significantly more achievable with the onset and availability of automatic measurement techniques improving the efficiency and accuracy of biomass estimation of invertebrates in aquatic systems.