Core Insights - Caris Life Sciences has developed and validated a new version of Caris GPSai™, a deep learning AI tool that enhances diagnostic accuracy for cancers of unknown primary (CUP) and misclassified tumors [1][2][3] Group 1: Technology and Methodology - Caris GPSai™ utilizes deep learning and whole exome and whole transcriptome sequencing (WES/WTS) to improve the precision of tumor origin predictions and identify potential misdiagnoses [2][4] - The tool was trained on WES/WTS data from over 200,000 cases and can classify tumors into 90 categories, achieving 95.0% accuracy in non-CUP cases and reporting tissue of origin in 84.0% of CUP cases [3][4] Group 2: Clinical Impact - In clinical use over eight months, GPSai changed the diagnosis for 704 patients, with 86.1% of these cases impacting treatment eligibility based on Level 1 clinical evidence [4][5] - 53.6% of surveyed physicians reported altering treatment plans due to the insights provided by GPSai, highlighting its significant influence on patient care [4][5] Group 3: Company Overview - Caris Life Sciences is a leading AI TechBio company focused on precision medicine, utilizing advanced AI and machine learning to analyze complex molecular data for improved healthcare solutions [6][7] - The company aims to transform healthcare through comprehensive molecular profiling and has established a large-scale clinico-genomic database to support its initiatives [6][7]