A computer program has made an important discovery in the treatment of spinal cord

The doctors have just learned something new about the post-operative recovery of spinal cord injuries in the spine. The connection between the success of long-term recovery and high blood pressure during surgery. Seemingly minor news, but it can help to quickly get on their feet millions of patients. The most interesting thing is how to make this discovery. He made a computer program.

New software conducted a meta-analysis of long-forgotten neurological research - and by a топологического data analysis failed to reveal patterns that are interested professionals.

The material for the meta-analysis were the results of studies that have written off as useless 20 years ago. After all, while there was no advanced machine learning systems, which is now the company has provided Ayasdi for the University of California at San Francisco, where he leads a group of neurologists work and statisticians. The functionality of the program is described in a scientific article , published yesterday in the journal Nature Communications.

"What nonsense it seemed, was very valuable information», - говорит Adam Ferguson (Adam Ferguson), a leading researcher of the center of Brain and Spinal Injury in University of California in San Francisco, one of the authors of scientific articles. How high is the value of this information, yet to be explored as a result of clinical trials, but the story has raised a number of interesting questions. For example, does it make sense to conduct new clinical trials, if it is possible to re-analyze the results of the old. And more: whether to publish in open access raw data from "failed" studies, even if they did not result scientific paper or some discovery.

Spinal cord injury - one of the fields of medicine, which is almost not developed. In this area for 20 years there were no significant findings. Spinal injuries are too complex and therefore difficult to trace than in other systems. The researchers decided again to analyze an array of old archived data in search of some hidden relationships.

For this purpose, the topological analysis of the data - a technique that uses the concept of geometric topology to find hidden patterns in large data sets. The program found many laws, most of them were known to physicians, including the inefficiency of various drugs. However, the harmful effects of high pressure on long-term recovery of the patient - this no one knew.





Neuroscientists say they could hardly find a pattern by conventional methods, without the use of date-Meiningen. There are too many variables to be considered.

The inventor of the topological analysis of the data - Stanford mathematician Gunnar Karlsson (Gunnar Carlsson), co-author of a scientific work. He is now president of the company Ayasdi, which was founded for the simultaneous use of technology of topological data analysis and machine learning. Soft Ayasdi analyzes data without human involvement, does not require the nomination of theories or areas of work.

Before spinal cord topological data analysis technique made another medical discovery: she helped discover unique mutation of breast cancer a > at which there was a high survival rate.

Source: geektimes.ru/post/264050/