Taking precise, objective and rapid measurements of the midline shift of the brain is especially important in patients with traumatic brain injuries, as it is a decisive factor in making therapeutic decisions in a timely manner.

At DotProduct we develop a solution that quickly and with high precision reports the magnitude of the midline shift to treating physicians using brain CT scans and classical machine learning and computer vision techniques.

Within the framework of this project we have developed models and simulators with a high degree of automation and parameterization that may soon be incorporated into the fire management system used by the Uruguayan Society of Forestry Producers. These models, includes a modified version of FPI (fire potential index) and a fire simulator based on Cell2Fire but completly developed by us to fully control it behavor.

The aim of this system (under development) is to allow video capture of a child sitting on the lap of an adult. The video is processed to perform feature extractions and the different reactions to the stimuli are studied. Subsequently, these characteristics are used to discriminate children between the ones within the autism spectrum and and the neurotypical ones using artificial intelligence.

Experience a revolution in personal fashion with our cutting-edge outfit recommendation platform completly centered around the user. Powered by sophisticated machine learning algorithms, our technology is designed to understand your unique style preferences and needs. With personalized suggestions that cater to your individual fashion profile, you’ll find the perfect ensemble for any occasion. For businesses, our platform represents a breakthrough in digital marketing – connecting with your desired audience and boosting sales conversions has never been easier. Unleash the power of tailored style with us today!

The project was developed for CONAPROLE and focuses on automating the estimation of forage productivity in plots dedicated to cattle feed for dairy farms, using satellite imagery from Sentinel-2. Through the acquisition and processing of remote sensing data, key vegetation indices are automatically calculated and integrated into productivity models, to assess the biomass available in each plot. The data pipeline and automated analysis were developed in Python, enabling efficient, large-scale monitoring and timely decision-making for optimizing management. By integrating satellite data with ground-based information, the system provides accurate and up-to-date forage productivity estimates, significantly improving the planning process in operations.