Selecting the right targeted therapy or clinical trial for each cancer patient is one of modern oncology’s greatest challenges. Each tumor carries a unique combination of four to five actionable driver alterations, drawn from millions of possible molecular profiles across hundreds of cancer genes. Traditional evidence-generation models built on randomized trials for each unique profile cannot keep pace with this level of complexity.
In this presentation, Dr. Istvan Petak will demonstrate how computational reasoning, a next-generation deterministic form of artificial intelligence, can address this combinatorial explosion in cancer genomics. By integrating multidimensional molecular data with clinical and real-world evidence, computational reasoning systems can identify rational therapy options and clinical trial matches in real time, with explainable logic and reproducible results.
Through illustrative case examples, Dr. Petak will show how this approach transforms the precision oncology workflow from variant interpretation to treatment recommendation, enabling scalable, evidence-based personalization that augments both laboratory and clinical decision-making.
Learning Objectives:
1. Understanding the concept of precision oncology
2. Understanding the dimensionality problem of precision oncology
3. Understanding the use of different types of AI in precision oncology
