Improving spike sorting efficiency with separability index and spectral clustering

• Published in: Medical engineering & physics 2025-01, Vol.135, p.104265, Article 104265
• Main Authors: Ranjbar, Leila, Parsaei, Hossein, Movahedi, Mohammad Mehdi, Sharifzadeh Javidi, Sam
• Format: Article
• Language: English
• Subjects: Feature extraction; Nervous system, Action potential, Spike sorting; Principal component analysis, Separability index; Spectral clustering
• ISSN: 1350-4533
• DOI: 10.1016/j.medengphy.2024.104265

•This study introduced a novel separability Index that quantifies the difficulty of spike sorting across various signals, facilitating precise comparisons of spike sorting algorithms across diverse datasets.•The separability index facilitates effective pre-classification of signals, helping to reduce the time and resources needed for processing large datasets.•The effectiveness of spectral clustering for spike sorting was evaluated using various feature sets, including raw samples, first and second derivatives, and principal components analysis (PCA).•The proposed method demonstrates superior accuracy over two previously published spike sorting methods, with performance metrics consistently aligning with the results of the separability index. This study explores the effectiveness of spectral clustering for spike sorting and proposes a Separability Index to measure the difficulty of spike sorting for a signal. The accuracy of spectral clustering is evaluated using different feature sets, including raw samples, first and second derivatives, and principal components analysis (PCA), and compared to two previously published methods. The results obtained over a dataset including 16 signals show that raw samples, with an average accuracy of 73.84 %, are effective for spectral clustering-based spike sorting. The analysis demonstrates that the proposed Separability Index can be utilized to classify signals beforehand, reducing the cost and processing time of large datasets. Furthermore, the proposed index can reveal spike sorting difficulty, making it a valuable tool for comparing the performance of various spike sorting methods in depth. The proposed method has higher accuracy (up to 23 %) compared to two previously published methods, and its accuracy is aligned with the Separability Index (correlation coefficient = 0.71). Overall, this study contributes to the field of spike sorting and offers insights into leveraging spectral clustering for this task.