Artificial Intelligence app to predict meningioma survival

Researchers recently published a study that shows proof of concept for how artificial intelligence can help doctors and brain tumor patients predicting survival and help make better treatment decisions. This study is published in Nature partner journal Digital Medicine. They also developed an open-source smartphone app meningioma.app to allow clinicians and other researchers to interactively explore the predictive algorithms described in the paper.

In this study, researchers from The Neuro (Montreal Neurological Institute-Hospital) and the Montreal Children’s Hospital of the McGill University Health Centre trained machine learning algorithms on data from more than 62,000 patients with a meningioma. Their goal was to find statistical associations between malignancy, survival, and a series of basic clinical variables including tumor size, tumor location, and surgical procedure. While the study demonstrated that the models could effectively predict outcomes in individual patients, the researchers emphasized the need for further refinements using larger sets that include brain imaging and molecular data.

More posts about medical apps at this link.

New findings of cortical activity during delta wave sleep sheds further light into memory formation

Scientists at the Center for Interdisciplinary Research in Biology have shown that delta waves emitted while we sleep are not generalized periods of silence during which the cortex rests, as has been described for decades in the scientific literature. Instead, they isolate assemblies of neurons that play an essential role in long-term memory formation. These results were published in journal Science.

While we sleep, the hippocampus reactivates itself spontaneously by generating activity similar to that while we are awake. It sends information to the cortex, which reacts in turn. This exchange is often followed by a period of silence called a ‘delta wave’ then by a rhythmic activity called a ‘sleep spindle’. This is when the cortical circuits reorganize to form stable memories.

However, the role of delta waves in the formation of new memories is still a puzzle: why does a period of silence interrupt the sequence of information exchanges between the hippocampus and the cortex, and the functional reorganization of the cortex?

The authors here looked more closely at what happens during delta waves themselves. They discovered, surprisingly, that the cortex is not entirely silent but that a few neurons remain active and form assemblies, i.e. small, coactive sets that code information. This unexpected observation suggests that the small number of neurons that activate when all the others stay quiet can carry out important calculations while protected from possible disturbances.

And the discoveries from this work go even further! Spontaneous reactivations of the hippocampus determine which cortical neurons remain active during the delta waves and reveal transmission of information between the two cerebral structures. In addition, the assemblies activated during the delta waves are formed of neurons that have participated in learning a spatial memory task during the day. Together these elements suggest that these processes are involved in memory consolidation.

To demonstrate it, in rats the scientists caused artificial delta waves to isolate either neurons associated with reactivations in the hippocampus or random neurons. Result: when the right neurons were isolated, the rats managed to stabilize their memories and succeed at the spatial test the next day.

Genetic-based epilepsy risk scores to promote precision medicine

Researchers led by Cleveland Clinic have developed new genetic-based epilepsy risk scores which could help provide more personalized epilepsy diagnosis and treatment. This research is published in the journal Brain.

Researchers combined all known common genetic variants identified from several large genome wide association study cohorts, which included more than 12,000 people with epilepsy and 24,000 healthy controls, to calculate the polygenic risk scores in more than 8,000 people with epilepsy and 622,000 population controls.

By combining the effect sizes of thousands of common genetic variants, these scores can be used to determine an individual’s risk for epilepsy. Researchers showed that these scores can accurately distinguish on a cohort level between healthy patients and those with epilepsy, as well as between patients with generalized and focal epilepsies.