Best Practices: Anne Frank- An Exhibition of World Perspectives  |  Best Practices: Learning Sciences by Simulated Flying  |  Science Innovations: Bacteriophages may play a role in childhood stunting  |  Teacher Insights: How novelty speeds up learning  |  Science Innovations: A major milestone in genetic research   |  Teacher Insights: Older people love learning from persons not machines   |  Technology Inceptions: Wearables to revolutionise the early detection of diseases  |  Parent Interventions: Why should we re-double efforts to promote children’s digital resilience  |  Technology Inceptions: Curd making refrigerator  |  Health Monitor: Missing proteins in neurological disorders in children  |  Parent Interventions: Sperm quality influenced by diet  |  Technology Inceptions: Canon's New Flagship EOS-1D X Mark III   |  Rajagiri Round Table: The 57th RRT shares more creative ways of vacation  |  National Edu News: Introduction of two levels of Mathematics for examination  |  National Edu News: CBSE shares guidelines for parents   |  
February 19, 2018 Monday 09:57:30 AM IST

Brain rhythms are sex specific

Science Innovations

19th February, 2018: Researchers from University of Twente, The Netherlands and University of Zurich, Switzerland have proven that electric brain signals of males and females show specific identifiable patterns. On employing a deep learning tool on to the database of brain signals collected using electroencephalogram (EEG), it could distinguish between the brain waves of a male or female person to an accuracy greater than 80%, suggesting that they contain unique pattern. The brainwaves from a transgender brain were not included in this study. These differences often evade visual inspection, even by the trained eye of a neurologist, but are “seen” by a powerful computer equipped with faculty of ‘deep learning’.

Researchers set up a special learning computer, a so-called ‘convolutional neural network’. This is an artificial neural net with several layers, which could determine over nine million parameters simultaneously. The network was first trained itself based on inputs from known sources, male or female. As the specific characteristics of male and female brain were not known beforehand, they were not entered. The machine on its own identified specific pattern of male and female brain waves and developed an ability to recognize them. A trained compute could take as input the brainwaves from an unspecified source and predict which belongs to whom, with sufficient accuracy, suggesting that they contained unique but invisible patterns.

The major differences in brain rhythms between were mainly in the beta frequency range, a frequency range between 20 and 25 Hz. Beta-rhythms are correlated with cognition and with tasks that are emotionally charged. It is well-known that women are better at recognizing emotions and at expressing themselves than men. These observations are reinstated by the current deep learning of brainwaves that are proven to be gender-specific.


The efficacy of deep nets to substitute or to complement human-guided feature extraction and knowledge discovery is vindicated in this research. This discovery could lead to gender-specific medicine and paves way for a more personalized medicine.

Comments