Cancer can’t be accepted as a normal part of society anymore, “it’s time for a ‘Vision Zero’ in oncology”. The complexity of cancer diseases demands intelligent solutions. Due to AI, advanced data engineering can improve processes in hospitals. In this article, focussing on breast cancer, the implementation of AI in cancer treatment processes and the key role of Kipoly will give more insights about upcoming developments.
The urgency is unarguable, every second person in Germany receives a cancer diagnosis at some point in their lives. Breast cancer is a leading cause of death among women worldwide and the most common cancer in Germany. However, with increased screening and the development of treatment methods, breast cancer mortality rates have improved over the last few decades. Postoperative adjuvant systemic therapies including chemotherapy, hormone therapy, and target agents have contributed to improving mortality rates in breast cancer patients.
Despite such improvements, every fourth woman dies of breast cancer. Women who have previously undergone breast cancer treatment are at higher risk of the disease than those without a history of breast cancer. In addition, patients with recurrent breast cancer have worse prognoses than those who have not recurred. “It’s time for a ‘Vision Zero’ in oncology”. A professor at the Charité, the initiator of the cancer symposium and the chairman of the initiative “Vision Zero 2020” agree on that and call for an intelligent digital collection and networking of medical data. Doctors should be able to easily access quantitative diverse data about the prehistory, course of disease and treatment of cancer patients.
AI excels at recognizing patterns in large volumes of data, extracting relationships between complex features in the data, and identifying characteristics in data (including images) that cannot be perceived by the human brain. This has already led to results in radiology, where clinicians use computers to process images rapidly, thus allowing radiologists to focus their time on aspects for which their technical judgment is critical. For example, last year CE approved the first AI-based software to process images rapidly and assist radiologists in detecting breast cancer in screening mammograms.
Kipoly focuses on the integration of AI technology in cancer care which could improve the accuracy and speed of diagnosis, aid clinical decision-making, and lead to better health outcomes. AI-guided clinical care has the potential to play an important role in reducing health disparities, particularly in low-resource settings.
The project Sina represents our efforts in applying AI to breast cancer research and includes the following aspects:
Merging continual learning and federated learning is the highlight of this project. Federated Learning enables us to train Machine Learning models on sensitive data in a privacy preserving way. Therefore, we can collaboratively train a model with sensitive data at different locations such as hospitals – consequently or simultaneously. With this technique numerous, previously unusable data sources can now be used for collaborative Machine Learning. Continual Learning is built on the idea of learning in a continuous and adaptive manner, gaining knowledge about the external world and enabling the autonomous and incremental development of ever more complex processes. In the medical field, improving AI-based decision making relies heavily on the capability of our system to adapt over time as every single patient is unique and requires precise examination, diagnosis, prognosis and monitoring. Now imagine a system being able to learn from multiple physicians in various clinics, treating patients with a similar disease and more importantly without transferring any data outside the clinic. We need a platform that provides tailored medical suits for cancer patients. That is the whole idea behind Kipoly!
Reach out to Reza Esfahanian per Email with questions or ideas for the next steps firstname.lastname@example.org