Unpublished work by the ‘father’ of modern computing could enable medical professionals to fine-tune precision medicine, which aims to personalise treatments on a patient-to-patient basis.

Research published in the Statistical Methods in Medical Research journal suggests Turing’s mathematical techniques could test the accuracy and effectiveness of diagnostic tools.

Precision Medicine

Turing’s method, according to researchers, could be used to tackle two key obstacles, including; how likely a person is to develop a disease, and if treatment programmes will work for them.

This could help to improve current statistical tools, which were developed in the 1980s, that assess the accuracy of diagnostic tests. Currently, medical professionals are unable to gauge how useful a test could be in determining a person’s risk of developing a disease.

Turing’s technique, researchers said, could prove more effective than the current C-statistic method. The C-statistic is widely used for evaluating the performance of diagnostic tests.

Professor Paul McKeigue, a researcher at the University of Edinburgh, suggests that the C-statistic fails to adequately indicate how diagnostic tests would perform when used to ‘stratify people by risk’.

Weights of Evidence

In a blog post for the journal Statistical Methods in Medical Research, professor McKeigue writes: “I propose an alternative approach based on estimating the weight of evidence. The background to this work lies in unpublished studies of Alan Turing, who in 1941 at Bletchley Park investigated the distribution of weights of evidence to decide the best strategy for breaking the ENIGMA code.

“Turing discovered some key properties of this distribution, which were extended in 1968 by his former assistant, Jack Good, by then one of the most influential Bayesian statisticians of the 20th century.”

Working at Bletchley Park in 1941, Turing devised his method when working to break the German Enigma code.

Turing’s approach investigated the distribution of so-called ‘weights of evidence’ – which establish the likely outcomes in a given situation – to help him decide the best strategy for decrypting German code.

A “New Era” of Precision Medicine

Applying Turing’s principles could potentially aid the development of personalised treatments, McKeigue argues.

While Turing established that the ‘weights of evidence’ was expected to vary over repeated experiments, this technique could be applied to evaluate diagnostic tests for personalised treatments.

“Most existing diagnostic tests for identifying people at high risk of cancer or heart disease do not come anywhere near the standards we could hope to see,” he writes.

“The new era of precision medicine is emerging, and this method should make it easier for researchers and regulatory agencies to decide when a new diagnostic test should be used.”