WALTHAM, Mass.--(BUSINESS WIRE)--BostonGene, a pioneer in artificial intelligence for tumor and immune biology, has announced the results of a landmark, independent, blinded multi-vendor study validating its AI platform for precision HER2 scoring in breast cancer. The study, published in the journal Modern Pathology, evaluated the performance of 10 different artificial intelligence models in assessing human epidermal growth factor receptor 2 (HER2) expression across whole-slide images from breast cancer samples. The research demonstrated that BostonGene’s AI foundation model achieved the highest level of agreement with expert pathologists among all participating systems. The results underscore the platform’s accuracy, consistency, and reliability in detecting HER2 status—a critical biomarker for guiding treatment decisions in breast cancer patients. The study was designed to be rigorous and unbiased, with pathologists and AI models analyzing the same set of slides without knowledge of the clinical outcomes. BostonGene’s model not only matched expert-level precision but also outperformed several other AI systems in terms of reproducibility and sensitivity in identifying both positive and negative HER2 cases. These findings reinforce BostonGene’s position as a leader in AI-driven digital pathology, particularly in complex biomarker assessment where accuracy directly impacts patient care. The company’s foundation model is trained on vast, diverse datasets and is designed to adapt across different imaging platforms and laboratory protocols, making it highly scalable for clinical use. With the growing adoption of AI in pathology, this validation adds strong evidence to the value of integrating advanced machine learning into routine diagnostic workflows. BostonGene’s technology is already being used in clinical and research settings to support more precise, data-driven decisions in oncology. The publication marks a significant milestone in the validation of AI tools in real-world pathology applications, and highlights the potential for AI to reduce diagnostic variability, improve turnaround times, and enhance the consistency of cancer biomarker testing.