Despite improvements in diagnostics and treatments response rates to treatments remain low, usually 4-30%. A main reason is that our current clinical tools are based on population studies and measure the average response instead of the responses of individual patients. DigOnc aims to change that by developing a clinical decision support system (CDSS) that can faithfully diagnose and predict treatment responses of individual patients. This CDSS is driven by a novel computational modelling core that uses artificial intelligence (AI) to integrate different types of data (clinical, molecular, imaging) on a patient and then extracts dynamic mechanistic models for predictive personalized simulations of disease progression and treatment responses. This core is scalable and adjustable to address different clinical questions through plugins supplying information on different clinical scenarios – like a digital oncologist. 

To demonstrate the utility and versatility of the DigOnc tool we will investigate three use cases for (i) choosing the best personalized treatment strategy for non-small cell lung cancer (NSCLC); (ii) predicting the extent of axillary lymph node involvement in breast cancer after positive sentinel lymph node biopsy; and (iii) identifying the best neoadjuvant therapy for HER2 positive breast cancer patients. These questions represent an unmet clinical need and allow for prospective validation studies within the lifetime of the grant. DigOnc will be designed for use in clinical practice with a user friendly interface and compliant to medical device regulations. The design will also take into account patient input and healthcare provider requirements.