A team from the MRC Biostatistics Unit at the University of Cambridge were among the winners in a global challenge to predict the evolution of Alzheimer’s disease.

The Alzheimer’s Disease Prediction Of Longitudinal Evolution (TADPOLE) Challenge has been running since June 2017, with participants from across the globe.

TADPOLE is a collaboration between the EuroPOND consortium, led by UCL, and the Alzheimer’s Disease Neuroimaging Initiative (ADNI).

Professor Daniel Alexander (UCL Computer Science), project lead of TADPOLE, said: “This has been a unique and inspiring challenge that has brought together a wide international community of computer scientists, statisticians, and clinical experts to advance the state of the art in a focussed challenge of predicting onset, development, and progression of AD. We hope the outcomes will make steps towards realising disease-modifying treatments in one of the biggest healthcare challenges we currently face”.

Participants had to use data provided willingly by anonymous patients in the ADNI study to predict the progression of AD. Specifically, participants had to predict clinical status (diagnosis), a score on a cognitive test, and a measure of brain shrinkage from MRI.

The University of Cambridge team comprised Dr Steven Hill, Dr James Howlett, Dr Robin Huang, Dr Steven Kiddle, Dr Anaïs Rouanet, Dr Brian Tom and Dr Simon White who collaborated with Dr Sach Mukherjee and Dr Bernd Taschler of the German Center for Neurodegenerative Diseases.

As reward for winning the overall best D3 prediction category the team were awarded £2500 in prize money.

The prizes were funded by generous donations from the Alzheimer’s Society (UK), Alzheimer’s Research UK, and the Alzheimer’s Association (USA).

For more details on the challenge, see https://tadpole.grand-challenge.org