Influence of cellular structures of skin on fiber activation thresholds and computation cost. (9th November 2018)

• Record Type: Journal Article
• Authors: Salkim, Enver; Shiraz, Arsam; Demosthenous, Andreas
• Is Part Of: Biomedical physics & engineering express. Volume 5:Number 1(2019)
• Publication Date: 2018-11-09
• Subjects: cellular structures -- computational models -- frontal nerve -- migraine -- neuromodulation; Medical physics -- Periodicals; Biophysics -- Periodicals; Biomedical engineering -- Periodicals; Medical sciences -- Periodicals; 610.153
• DOI: 10.1088/2057-1976/aaeaad
• Languages: English
• ISSNs: 2057-1976

Abstract

Abstract: Electrical neuromodulation is widely used to treat and manage neurological disorders. Migraine, a socioeconomic burden, may be treated using this technique. Transcutaneous stimulation of frontal nerves by electrodes placed on the forehead is of interest as it exposes patients to lower levels of risk and side-effects compared with surgical and pharmaceutical solutions and may be readily delivered. The size, shape and placement of the electrodes can be optimised using computational models involving a volume conductor model of anatomical structures and electrodes as well as nerve fibre models. A detailed volume conductor incorporating cell level structures of skin can yield an accurate map of electrical potential distribution due to an electrode setting. However, such a model imposes a very substantial computational cost which may impede the design process. Computation cost can be significantly reduced if the skin microscopic structures are ignored. In this study, we compare the accuracy and computation cost with and without skin microscopic structures on the outcome of a device for transcutaneous frontal nerve stimulation. The performance is presented as the percentage activation of target nerve fibres in response to the level of stimulus current delivered via surface electrodes placed on the forehead. When cell level structures of skin are not incorporated, discretisation time is reduced from 21 h to 0.4 h and the number of finite elements used from 18 M to 1.4 M. Only 1% difference in stimulus current thresholds is observed.