Scholarlink

<p>In this paper, we propose a novel application of contrast-based histogram analysis for detecting action potentials in biosignals, leveraging the difference histogram technique to quantify signal contrast. Our approach extracts a histogram of differences, capturing the information related to action potentials (spikes), and computes the contrast from this histogram. This contrast is combined with an adaptive threshold that is dynamically calculated based on the contrast signal from the biosignal, which effectively allows differentiation of action potentials from background noise. We evaluate the performance of our spike detection method using both synthetic and real datasets commonly used for spike detection and sorting algorithms in intracellular and extracellular recordings. Our results indicate that our Contrast Spike Detector (CSD) outperformed other spike detection methods. Specifically, our CSD achieves remarkable performance, with accuracy, precision, recall, and F1-score rates exceeding 99% in most cases. This research aims to advance online spike detection methodologies suitable for both software and hardware implementations.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Contrast-based histogram analysis enables adaptive action potential detection

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月28日

查看全文
  • Francisco J. Iñiguez-Lomeli,
  • Valentin Flores-Payan,
  • Lilia del Carmen Castillo-Villarruel,
  • Victor H. Jimenez-Arredondo,
  • Juan Carlos Gomez-Carranza,
  • Horacio Rostro-Gonzalez

摘要

In this paper, we propose a novel application of contrast-based histogram analysis for detecting action potentials in biosignals, leveraging the difference histogram technique to quantify signal contrast. Our approach extracts a histogram of differences, capturing the information related to action potentials (spikes), and computes the contrast from this histogram. This contrast is combined with an adaptive threshold that is dynamically calculated based on the contrast signal from the biosignal, which effectively allows differentiation of action potentials from background noise. We evaluate the performance of our spike detection method using both synthetic and real datasets commonly used for spike detection and sorting algorithms in intracellular and extracellular recordings. Our results indicate that our Contrast Spike Detector (CSD) outperformed other spike detection methods. Specifically, our CSD achieves remarkable performance, with accuracy, precision, recall, and F1-score rates exceeding 99% in most cases. This research aims to advance online spike detection methodologies suitable for both software and hardware implementations.