A team of scientists discovered that fMRI machines offer results with a 70% false positives risk. Almost 40,000 studies might have been affected by a mistake in the software of the device.
A team of Swedish and British researchers analyzed data validity for three of the most used software packages for brain scans. The results showed that bugs in the programs made previous fMRI studies to be highly unreliable.
The mistake was due to the fact that the scan analysis algorithm reduced the size of the data clusters and overestimated their significance.
fMRI is an imaging device that uses functional magnetic resonance to display views of the body structure and composition.
The functional neuroimaging procedure of the fMRI helps scientists to measure brain activity, as it detects the changes associated with blood flow.
The software divides the image scan of the brain into three-dimensional units of information called voxels. Later, the program does a search among the voxels to find the blood flow inside the brain.
The present study analyzed the results of fMRI scans performed on 499 healthy people all around the world. The tests were performed when the individuals were in resting state.
The researchers split the scans into 20 groups and compared them one against each other. A second data analysis involved the comparison among the three most used software packages for fMRI.
The accepted error margin between the image and the reality is of 5%. However, the authors found a 70% chance of false positives, which means that the program displayed data clusters that were not present in the actual brain activity of the person.
In other words, the fMRI scan showed activity in regions of the brain that were, in fact, not active.
The most concerning implication of the study is that thousands of research papers that used fMRI scans are now proven to be false, or at least irrelevant.
The fMRI device has been invented over 25 years ago, and the scientists wonder how was it possible that until now, no one had ever challenged the results of the scans.
While the news may be catastrophic for the scientific community, some experts argue that the software bug only affected the AFNI package and that the dysfunction appears just in the activation map.
Another rational argument against the importance of the error was that 70% false positives risk would not necessarily mean that 30% of the positive results are false, but only that there is a 70% chance that a positive result is false. This is a faint consolation.
Image Source: Wikipedia
Latest posts by Alan O’Leary (see all)
- Woman Found Alive After Missing for 42 Years - Oct 30, 2017
- October Will Welcome The Draconid Meteor Shower And The Orionids - Oct 6, 2017
- Scientists Are At A Loss After Unearthing A Porpoise Grave - Sep 22, 2017