Optimizing cancer therapy with machine learning
Professor Giuseppe Masucci from Karolinska Institutet’s Onco-Immunotherapy Innovation and Research Unit (OITIRU) developed an algorithm for defining a more accurate prognosis and personalized treatment plan for cancer patients. Genevia Technologies has thoroughly validated the algorithm’s performance, and is now their chosen partner in improving the mathematical model.
The algorithm was developed to identify an optimal combination of therapies for each patient by assessing clinical data. The aim is to eventually create a generalized, easy-to-use model that clinicians all over the world can use, with any type of cancer data.
– Before I contacted Genevia, I had developed a simple algorithm using clinical databases of colon cancer patients. A statistician’s help was needed to validate its accuracy, and then to corroborate the initial algorithm using more sophisticated methods, says professor Giuseppe Masucci.
The bottleneck of the project was overcome quickly, after finding the right partner.
– I had been waiting for one and a half years for the validation, but with Genevia, the work was done within six weeks after the very first meeting.
Professor Masucci describes his disbelief that after a handful of meetings with Genevia, he had acquired the long-desired model validation. Geneva’s team applied machine learning technologies in their model, which led to an outcome in agreement with professor Masucci’s algorithm.
– The project went better than expected. I was so surprised that the project manager Felipe Simão was able to understand my ideas so quickly.
The trust that was built in the first project made professor Masucci choose Genevia as their partner for the further development of the statistical model. The work is proceeding smoothly, and the communication has been effortless from day one.
– So far, there haven’t been any challenges with the collaboration. We have always received well-explained answers to all of our questions.
Both parties agree that the benefits of the project are mutual. Professor Masucci is excited about the possibility of enhancing cancer treatment with the aid of statistical approaches, while Genevia has the opportunity to learn more alongside the project.
– This project has been particularly interesting to work in as it’s real-world applications in cancer treatment have been clear from the start. It has been a pleasure to collaborate with professor Masucci to combine our machine learning expertise with his clinical experience and data to further develop a tool that can be of assistance in the daily routine of cancer clinicians, says the project manager Felipe Simão.
Together towards better solutions
A colleague at Karolinska Institute encouraged professor Masucci to contact Genevia in the first place, and he now happily spreads the good word around.
– The three key reasons why I would recommend Genevia to others are fast responding, easy communication and, of course, their high-quality knowledge, sums up professor Masucci.
– The best thing is to have a creative bioinformatics partner who understands me and can put their own ideas into the project. The partnership is a continual brainstorm of originality.