Big data analytics to improve patient care in Oncology
Cancer is one of the main causes of death in Europe. In 2012 an estimated 2.6 million new cases were diagnosed in the EU and approximately 1.3 million patients succumbed to their disease, according to the report Cancer Screening in the European Union. Breast cancer, together with tumours of the prostate, lung, and colon or rectum comprise over half of all newly detected cancers. Lung and colorectal cancers contribute most to causes of cancer death.
The economic burden of cancer care in the EU is also significant. It amounted more than 126 billion euro in 2009, with healthcare accounting for 51 billion euro (40 %), under an article published in Lancet Oncology. This economic burden includes the primary care of cancer patients but also costs for productivity losses due to early death (42.3 billion euro), lost working days (9.4 billion euro), and informal care cost (23.2 billion euro).
The highest economic cost is attributed to lung cancer (15%), followed by breast cancer (12%), colorectal cancer (10%) and prostate (7%). The healthcare costs of cancer were equivalent to €102 per citizen, although it varied substantially depending on the country: from €16 per person in Bulgaria to €184 per person in Luxembourg.
One of the research streams within the BigMedilytics project is dedicated to developing technologies to more effectively improve the management of cancer patients. It also aims to enhance the treatment of cancer. For this reason, investigating and understanding the impact of big data is key to this project.
The project theme ‘Oncology’ comprises three clinical cancer pilots: one for lung cancer, breast cancer, and prostate cancer, respectively. For all cancer pilots, a range of key performance indicators (KPI’s) have been defined; these KPI’s will be addressed during the project with the aim to enhance on multiple of them. Ultimately, this will lead to more cost-effective patient management with improved patient diagnostics and treatments.
Heterogeneous data, which are created by multiple medical disciplines, are the most relevant data for the pilots addressed. It will be collected in a structured manner and big data technologies such as machine learning and deep learning will be applied to create knowledge and insight from the data.
(Image by ijeab / Freepik)
