A Python Code For Maximum Likelihood Estimation Of The Location And Scale Parameters Of The Truncated Normal Distribution

• Published in: Medical Technologies National Congress (Online) pp. 1 - 4
• Main Authors: Ogutcen, Melih Yilmaz, Kocaturk, Mehmet, Okatan, Murat
• Format: Conference Proceeding
• Language: English
• Published: IEEE 04.11.2021
• Subjects: Band-pass filters; Codes; Filtering algorithms; Gaussian distribution; Maximum likelihood estimation; Microelectrodes; python code; Real-time systems; truncated Normal distribution; truncation thresholds
• ISSN: 2687-7783
• DOI: 10.1109/TIPTEKNO53239.2021.9632955

Extracellular neural recordings obtained from chronically implanted microelectrode arrays are widely used in behavioral neurophysiology and invasive brain-machine interfaces. After the raw recordings are band-pass filtered within a frequency band suitable for spike detection, spikes are often detected by amplitude thresholding. Developing principled methods for computing amplitude thresholds is an active research area. 'Truncation thresholds' are a pair of amplitude thresholds that are computed using a recently proposed algorithm. As part of an effort that aims to integrate this algorithm into a real-time data acquisition and spike detection system, here we present a Python code for maximum likelihood estimation of the location and scale parameters of the truncated Normal distribution, which is one of the steps involved in the computation of truncation thresholds.