PNS-GAN: Conditional Generation of Peripheral Nerve Signals in the Wavelet Domain via Adversarial Networks
| Autor: | Blake A. Richards, Olivier Tessier-Lariviere, Luke Y. Prince, Emil Hewage, Lorenz Wernisch, Guillaume Lajoie, Oliver Armitage, Pascal Fortier-Poisson |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
0301 basic medicine
business.industry Noise (signal processing) Computer science Pattern recognition Neural engineering Signal Convolution 03 medical and health sciences 030104 developmental biology 0302 clinical medicine Wavelet Recurrent neural network Modulation Artificial intelligence business 030217 neurology & neurosurgery Decoding methods |
| Zdroj: | NER |
| Popis: | Simulated datasets of neural recordings are a crucial tool in neural engineering for testing the ability of decoding algorithms to recover known ground-truth. In this work, we introduce PNS-GAN, a generative adversarial network capable of producing realistic nerve recordings conditioned on physiological biomarkers. PNS-GAN operates in the wavelet domain to preserve both the timing and frequency of neural events with high resolution. PNS-GAN generates sequences of scaleograms from noise using a recurrent neural network and 2D transposed convolution layers. PNS-GAN discriminates over stacks of scaleograms with a network of 3D convolution layers. We find that our generated signal reproduces a number of characteristics of the real signal, including similarity in a canonical time-series feature-space, and contains physiologically related neural events including respiration modulation and similar distributions of afferent and efferent signalling. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|