GoldSpot's proprietary Artificial Intelligence (AI) and geological interpretation highlights Lithium-Tantalum potential at Critical Elements' New Block 1-6 and 7 claims within the Nemiscau greenstone belt
Engagement with Critical Elements to uncover EV battery material showcases GoldSpot's ability to work with leaders across all commodities and deposit types to identify new mineral exploration targets
A total of 19 high to moderate prospectivity Lithium-Tantalum targets were identified, with Nickel-Copper and Gold potential also revealed
Toronto, Ontario--(Newsfile Corp. - September 7, 2021) - GoldSpot Discoveries Corp. /zigman2/quotes/210443510/delayed CA:SPOT -2.97% (otcqx:SPOFF) ("GoldSpot" or the "Company"), a leading technology services company leveraging machine learning to transform the mineral discovery process and Critical Elements Lithium Corporation /zigman2/quotes/204783849/delayed CA:CRE +10.84% (otcqx:CRECF) /zigman2/quotes/206100393/delayed DE:F12 +5.50% ("Critical Elements") are pleased to announce the results of a property-wide comprehensive data review, compilation, and target generation on Critical Element's New Block 1-6 and 7 claims within the prolific Nemiscau greenstone belt in James Bay, Québec.
GoldSpot works with leading exploration and mining clients across all commodities and deposit types, using cutting-edge technology and geoscientific expertise to mitigate exploration risks and significantly increase the efficiency and success rate of mineral exploration across resources.
Vincent Dubé-Bourgeois, CEO of GoldSpot Discoveries commented: "I'm thrilled to announce the results of our investigation and analysis of Critical Element's claims sourced from our team's extensive digital extraction of assessment files, government data and academic studies. This dataset provided outcrop/sample description, bedrock geology, geochemical analyses, and geophysical surveys which generated lithium-tantalum, copper-nickel, and gold focused targets, using our geological and machine learning methods. We look forward to working with the Critical Elements exploration team to validate these targets and further advance the claims."
Jean-Sébastien Lavallée, CEO of Critical Elements commented: "We are very pleased with the results of the combined AI targeting and the outcrop detection conducted by GoldSpot. These cutting-edge approaches enabled us to quickly generate several promising targets. These tools are extremely useful to reduce exploration cost and time, in particular the large portfolio of 700 square kilometers owned by Critical Elements."
Figure 1: Location of Critical Elements' projects, Eeyou Istchee, James Bay, Québec
To view an enhanced version of this graphic, please visit:
This study hinged on digital extraction from an exhaustive collection of compiled data, including assessment files, government data and academic studies. This dataset provided outcrop/sample description, bedrock geology, geochemical analyses, and geophysical surveys. Original data cleaned and combined to create a comprehensive data set for geological interpretation and machine learning processes.
The compilation of discrete outcrop observations allowed a reliable update to existing geologic maps, resulting in a refined pegmatite map. A total of 42 pegmatite bodies were added to the current geological map, highlighting previously unknown potential for economic lithium-tantalum mineralization.
An up-to-date structural interpretation was created based on a high-resolution aeromagnetic survey commissioned by Critical Elements. This survey revealed structurally complex patterns, including large-scale folds and major ENE-trending ductile fault zones.
GoldSpot Target Generation
GoldSpot generated lithium-tantalum, copper-nickel, and gold focuses targets, using a "Smart approach" of knowledge- and AI data-driven methods.
Processes: The AI data analysis trains machine learning algorithms to predict the presence of lithium-tantalum (model 1), copper-nickel (model 2), and gold (model 3), using all variables (features), both numeric and interpreted on a 10 x 10 m grid cell datacube. Once the model performs to a satisfactory level, results produced include:
Performance: The best prediction model for the lithium-tantalum model was obtained using a Random Forest classifier for which performance metrics were above 80% precision. The updated geology and structural interpretation were the dominant contributors to the targeting model.