# Feature extraction for LSTM-based finger-joint angle estimation from sEMG

DevFeed: [Feature extraction for LSTM-based finger-joint angle estimation from sEMG](<https://devfeed.tech/articles/feature-extraction-for-lstm-based-finger-joint-angle-estimation-from-semg-114729.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.02736>)

Author: Gerda de Jager, Marelie H Davel, Ian Thomson

Published: 2026-11-03T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Feature Engineering](<https://devfeed.tech/topics/feature-engineering.md>), [Embedded Software Dev](<https://devfeed.tech/topics/embedded-software-dev.md>), [On-device AI](<https://devfeed.tech/topics/on-device-ai.md>)

Tags: [arxiv](<https://devfeed.tech/tags/arxiv.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [embedded](<https://devfeed.tech/tags/embedded.md>), [features](<https://devfeed.tech/tags/features.md>), [lstm](<https://devfeed.tech/tags/lstm.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [real-time](<https://devfeed.tech/tags/real-time.md>), [regression](<https://devfeed.tech/tags/regression.md>), [rms](<https://devfeed.tech/tags/rms.md>), [series](<https://devfeed.tech/tags/series.md>), [window](<https://devfeed.tech/tags/window.md>)

## AI overview

On the Ninapro DB2 dataset, LSTM finger-joint angle estimation performed best with a 300 ms signal window, 100 ms step, and either the Mean Absolute Value (MAV) or Root Mean Squared (RMS) feature; both settings produced a test PCC of about 0.779. The study evaluated six time- and frequency-domain features, windows from 300 to 1200 ms, and steps from 10 to 100 ms, measuring regression performance, inference time, and memory use for embedded-device feasibility. It reports that window size, step size, and feature choice interact, so tuning them together can improve accuracy and computational feasibility. The supplied text is the abstract, so it does not provide detailed results for other configurations.

## Source excerpt

arXiv:2611.02736v1 Announce Type: new Abstract: Continuous estimation of finger-joint angles from surface electromyography (sEMG) is a promising approach for controlling dexterous myoelectric prostheses. Deep time series models have been applied successfully in this context, typically relying on feature extraction rather than raw signals, given limited subject-specific data. Various time-domain and frequency-domain features can be used, extracted for a given window and step size. This study systematically investigates these three design choices for Long Short-Term Memory (LSTM) regression on the Ninapro DB2 dataset. Window sizes from 300 to 1200 ms and step sizes from 10 to 100 ms were evaluated in combination with six time- and frequency-domain features. Models were evaluated using regression performance metrics, inference time, and memory footprint to assess feasibility on embedded microcontroller hardware suitable for a prosthetic device. A 300 ms window with a 100 ms step, paired with either the Mean Absolute Value (MAV) or Root Mean Squared (RMS) feature, yielded the best performance (test PCC of 0.779 +/- 0.080 / 0.779 +/- 0.085). Window size, step size, and feature choice were found to interact non-trivially: jointly tuning them can improve both accuracy and computational feasibility for real-time, embedded myoelectric control.