Development of an AI-powered Prediction of BRAF Mutation & Patient Response to Immunotherapy from Digital Pathology H&E Slides in Metastatic Melanoma

The goal of this webinar is to introduce the Bio-AI Health Predict X platform and demonstrate the feasibility of using H&E slides for the detection of molecular markers as well as for the prediction of patient response to therapy.

Bio-AI Health will demonstrate: 

  • How pathology annotations were used to develop a melanoma-specific tumor detection model 
  • How that integrates with the discovery of molecular biomarkers in tissue


Bio-AI Health proposes a novel Artificial Intelligence (AI) application to advance precision medicine and correlate biomarker detection to patient response to therapy. The use of pathology slides for the detection of key molecular signatures and prediction of patient response to therapy is an emerging field with enormous value propositions. The ability to leverage existing tissue samples and not burden patients with additional tests has enormous value in the clinic. 

In this webinar, Bio-AI Health in collaboration with Dr. Paolo Ascierto and his clinical laboratory at the National Tumor Institute “Fondazione G. Pascale” in Naples, Italy will demonstrate how AI and Deep learning was used on H&E samples to predict the BRAF molecular status of patients as well as for the prediction of immunotherapy. 

Advances in Next-Generation Sequencing (NGS) have facilitated patient recruitment to therapy by identifying mutations that drive certain cancers and allow for specific drug treatment. However, its use is limited due to the amount of tissue needed, high cost associated with tests, complex workflow, and low test availability in all markets. Pathology slides on the other hand are readily available as most patients identified with cancer will have undergone a biopsy and have the H&E slide generated. 

Bio-AI Health’s proprietary Predict-X platform technology integrates a broad range of disparate data sets from translational studies, clinical trials, and real-world evidence databases to improve patient selection criteria & increase success rates for clinical programs. 

Bio-AI’s goal is to reduce the complexity of patient profiling data for use in clinical practice and provide researchers and clinicians with valuable and actionable insights.  Bio-AI adopts a flexible “plug & play” development framework that enables the rapid generation of new predictive AI models to be tailored to the precise needs of pharma and biotech partners