Real-time feature-space filtering method for detecting minute signals in scanning probe microscopy

2023 Fiscal Year Final Research Report

• Project/Area Number: 22K18968
• Research Category: Grant-in-Aid for Challenging Research (Exploratory)
• Allocation Type: Multi-year Fund
• Review Section: Medium-sized Section 29:Applied condensed matter physics and related fields
• Research Institution: Kanazawa University
• Principal Investigator: Shinji Watanabe 金沢大学, ナノ生命科学研究所, 准教授 (70455864)
• Project Period (FY): 2022-06-30 – 2024-03-31
• Keywords: 特徴空間フィルタ / 走査型プローブ顕微鏡

Outline of Final Research Achievements

In this study, we designed and implemented a feature space filter based on machine learning for Scanning Ion Conductance Microscopy (SICM). This filter significantly improved the signal-to-noise ratio (SNR) and data throughput of SICM measurements. By employing the filter to accurately read and classify signal and noise information, we successfully enhanced the SNR. We demonstrated that this method is particularly effective in low SNR conditions compared to existing filtering techniques. However, the current model requires complex parameter adjustments due to its high degree of freedom in classifier configuration. Moving forward, it will be necessary to develop algorithms that simplify these parameter adjustments.

Free Research Field

ナノサイエンス

Academic Significance and Societal Importance of the Research Achievements

高速なフィードバック処理かつ微小信号の扱いが必須なシステムにおいて、フィルタ性能の向上は大きな課題である。本研究は、走査型プローブ顕微鏡（SPM）をこのようなシステムとして取り上げ、本研究でデザインしたデジタルフィルタが有用であることを示すことに成功した。ノイズと信号の情報を物理計測のモデルを用いて分類する本研究のフィルタ設計手法は幅広く応用できるものであり、汎用性が高い技術を開発できたと考えている。