As a response to the infodemic of cancer related publications, we’re developing the world’s first real-time systematic review tool (R-CANCER) by combining cutting edge machine learning and a global collective of dedicated curators.
See how we have already applied this methodology to the COVID-19 use case:
The speed and volume of clinical evidence that supports the development of clinical guidance and best practice for early detection of cancer is overwhelming and almost impossible to manage. In 2020, 212,471 citable papers have been published on the topic of cancer on pubmed.gov.
Misinformation can be harmful to people’s physical and mental health; increase stigmatisation; threaten precious health gains, and lead to poor observance of public health measures.
Without the appropriate trust and correct information, emerging and current cancer diagnostics and treatments can be overlooked.
We're creating the foundation for fast searching and review of accurate information quality checked information, based on science and evidence and preventing the spread, and combating, misinformation.
We’re building the one of the worlds largest and most current oncology evidence sources. This data pipeline converges with a novel curation methodology that adopts a “human in the loop” methodology for the characterisation of quality, relevance and key evidence across a range of scientific literature sources.
By capturing curators’ critical appraisal methodology through the application of discrete labelling and rating of information, we will develop a foundational data science dataset of articles in the realm of pooled cancer information.
This methodology requires volunteers with a background in medicine and reviewing scientific content, to support analysis and assessment of the data that will drive R-CANCER.
We’re urgently seeking data curators to help in the fight against cancer.
A background in medicine
Experience reviewing scientific content
A desire to analyse and assess data
Participation in a game-changing project
Potential citation at a collaborator in select output publications
