 Imagine, you are all doctors and you have to recommend the best treatment for this cancer patient. Today, we have a technology that can recommend movies. Sadly, we don't have a technology that can recommend treatments. Ladies and gentlemen, I am a machine learning scientist and I want to help doctors to choose the right treatment for cancer patients. In cancer, few genes go mad, which means genes change function. As a result, cells acquire hostile intelligence, causing these cells to replicate enormously. To stop cancer, we need to kill our own cells and we can do this intelligently by personalizing treatments that target only the mad genes in the patient. But the key question is, how would a doctor know which genes have gone mad? I think the answer lies in the big data that are emerging in cancer nowadays. However, these data are incredibly huge and they come from multiple sources. Unfortunately, a doctor alone cannot analyze their complex relationships and I believe artificial intelligence can solve this. Just like it solved movie recommendation in Netflix. Therefore, my vision is to augment the doctor's expertise with AI powered by big data. Unlike other AI approaches, my novel AI solution combines multiple data sources to learn their complex interactions. Specifically, it searches for statistical dependencies across multiple genomic patterns and treatment outcomes to identify the mad genes. This data-driven knowledge extracted from multiple sources can better help the doctors to think beyond current treatment choices. The game-changing idea of combining multiple sources seems ambitious, but I have proof that it boosts treatment efficacies. Together with colleagues, I have developed the best integrative AI method that won two international competition on personalized treatment prediction outperforming 50 teams globally and were published in top scientific fora. This unprecedented success already gives the confidence to take the AI research to the clinics. So, with the Scholar Award, I will further optimize the AI method to clinical settings and test them on the patient's data which are uniquely available to me from the largest cancer hospital in Finland. What's awesome is that my solution will help the doctor to recommend the most effective treatment for cancer patients. To sum up, cancer is an intelligent killing machine. Let's defeat it with artificial intelligence. Thank you. So now over to the jury to ask the next questions. Perfect. Actually, well, I have millions of questions, but I only ask one. I mean, we know that there has been a lot of effort in trying to use artificial intelligence, especially in cancer treatment. Why your solution will be better than the others? Since we know that there have been some, well, not failure, but some of these were already not very successful. So why this solution is better? Thank you for the question. So our AI solution combines information from multiple data sources. So that we have shown already by winning the dream challenges, which were these challenges held, that integrating multiple information in an intelligent way improves the predictions. Basically, in other words, it improves the treatment efficacies in the cancer. So this is one thing which we should find that this works. Yeah, I had a similar question. There are quite a few efforts. And I think the differentiating factor is who has the best data sources. And often the initial focus on the types of cancers seems to be a way to take off. Is there any initial focus or initial types of cancers you're optimizing for? You're optimizing. Yeah. Thank you for the question again. So basically my institute has one of the state-of-the-art facilities for cancer research. They are working in multiple cancers, but my specific focus is on blood cancer, specifically leukemia. Mohamed, I really like what you're doing. And I absolutely think that the doctors need help of AI. Thank you. So with AI, exactly like you talked about here, it's about data. So where will this data come from? Have you thought about that? Yes. Again, thank you for the nice question. So basically I'm working an institute of molecular medicine, FIM, which has the state-of-the-art facilities to collect data with a clinical partner, which is Helsinki University Cancer Hospital. So we are collecting data from there. And then from there, the patient data comes to me to analyze all these stuffs. Perfect. Thank you very much. Cool. Can everyone give us round of applause for Mohamed? Okay. Thank you. Thank you. Thank you so much.