Researchers at NewYork-Presbyterian/Weill Cornell managed to provide evidence that a patient with a severe brain injury could, in their own way, communicate accurately by using complex machine-learning techniques to decipher repeated advanced brain scans. The study published in this week's issue of the Archives of Neurology, shows the difficulty of determining whether or not a patient is able to communicate using only measured brain activity, even if they are able to generate reliable patterns of brain activation in response to instructed commands.
According to the researchers, patients in a barely conscious state, or those with locked-in syndrome, i.e. who have normal cognitive function but with severe motor impairment, who can follow commands yet have no motor response, may not produce clearly interpretable communications using the same patterns of brain activity.
The researchers state that even though less refined methods have proven to be successful, their new approach offers significant new insights into brain function and level of consciousness, whilst identifying mechanisms of variation in brain activity supporting cognitive function after injury.
Dr. Nicholas D. Schiff, professor of neurology and neuroscience and professor of public health at Weill Cornel Medical College, and a neurologist at NewYork-Presbyterian Hospital/Weill Cornell Medical Center explains:
"In these studies we have reanalyzed earlier published data that demonstrated an effort to communicate using brain activations alone that apparently failed but was nonetheless a clear effort to generate a response. Importantly, the reanalysis with new, more sensitive methods provides evidence that the problem with communication may reflect a mismatch of our expectations in designing the assessment, rather than a failure on the subject's part in an attempt to accurately communicate with us."
Leading researcher, Jonathan Bardin, a fourth-year neuroscience graduate student at Weill Cornell Medical College comments:
"Our study shows that multivariate, machine-learning methods can be useful in determining whether patients are attempting to communicate, specifically when applied to data that already show evidence of a signal in univariate, more standard methods of analysis. It is our clinical and ethical imperative to learn as much as possible about their ability to communicate. A simple bedside exam is not good enough."
Dr. Schiff adds:
"We need a set of methods that are both powerful and simple, and we are not there yet, as this study shows. We are using quite complex tasks to perhaps detect just the few of many patients who are conscious."
Varying abilities among patientsIn a continuation of the NewYork-Presbyterian/Weill Cornell study, researchers assessed methods in which fMRI can establish lines of communication with brain-injured patients to estimate the their pain levels and to understand other clinical parameters, as well as whether or not they could benefit from rehabilitation that would improve care and quality of life.
The study specifically follows up on a previous study featured in the journal Brain last February, which showed that findings of using fMRI to detect consciousness in six either barely conscious or locked-in patients resulted in broad and largely unpredictable differences in the patients' ability to respond to a simple command, such as "imagine swimming -- now stop".
The same command was subsequently used to answer simple yes/no or multiple-choice questions and the variation became apparent when compared with the bedside ability to interact using voice or gestures.
Whilst some patients who could not communicate by gestures or voice were unable to perform the mental tests, others unable to communicate by gestures or voice were able to sometimes respond to the researchers' questions by using mental imagery.
Interestingly, some patients who were able to communicate using gestures or voice could not perform the mental tasks.
According to the researchers, these findings indicate that currently there are no exams available that can precisely evaluate whether or not some severely brain-injured individuals have a higher-level functioning.
Commenting on the findings Dr. Schiff says:
"There are people whose personal autonomy is abridged because they don't have a good motor channel to express themselves despite, in some cases, having a clear mind and opinions and desires about themselves and the world. Not all minimally conscious patients are the same, and not all patients with locked-in syndrome are the same."
Methods Need To Be Sensitive and FlexibleThe biggest new finding is a reinterpretation of the observations made in a 25-year-old patient, who was the only one of six patients able to use fMRI signals for communication in the earlier study. Her results were confusing, as it appeared that she was consistently responding to the answer that was directly after the correct answer.
"It's often seen in patients like this -- she had a stroke that damaged her brain -- that there can be a cognitive delay in some area of the brain. FMRI is a readout of blood flow instead of actual neural activity, so these delays could be caused by an interruption of blood flow due to damage or could just mean they are working on the problem more slowly, and the answer looks wrong because it is given in the next response period."
To gain a better insight, Bardin used a newer technique that was developed on the origins of machine-learning research, to instruct a computer to assess the patient's multiple fMRI scans after she answered the two questions several times. This multivariate method used the same data as used in the first study, that just as the commonly used "univariate" evaluation, specifically examines the Supplementary Motor Area (SMA) brain function, which is active when healthy individuals imagine doing something, however, the multivariate evaluation investigates if any part of the brain shows a pattern of activity that is consistent from one scan to the next.
"When there is significant damage to the brain, it can rewire itself so that functions associated with SMA could be processed somewhere else."
They discovered using this complex approach that the patient had indeed consistently attempted to communicate answers to both questions, but at a delayed speed, however they stress that one method of analyzing fMRI scans is not preferable to the other for all patients. They highlight that univariate methods should always be performed first, as multivariate approaches can be particularly noise sensitive, which can lead to false positives if used on their own.
They continue saying that if a signal is revealed using the standard approach, physicians could use the multivariate method to gain a better understanding and maybe identify a patient's response when the univariate results are ambiguous.
"We did all these things to simply show that we think this patient was trying to communicate. You have to be very careful in your data analysis before saying anything strongly about what a patient can or cannot do."
Dr. Schiff adds:
"Rigid experimental paradigms like those used in the field can very well miss important information about these patients. This is all extremely complex and messy, but we should expect that. Given the injuries some of our patients suffer, their cognitive abilities are very difficult to detect behaviorally or through simplistic tests or scans."
Written by Petra Rattue