WiMi Hologram Cloud Inc., a leading global Hologram Augmented Reality Technology provider, announced that a new motion artifact suppression and morphology optimization algorithm is developed for motion artifacts such as peaks, baseline mutations and slow drifts in fNIRS signal processing based on mathematical morphology and median filtering methods. The algorithm makes full use of mathematical morphology methods to analyze and optimize the signal morphological features, and combines the advantages of median filtering algorithms for improvement, in order to enhance the ability of accurate identification and effective correction of motion artifacts in fNIRS signals, and to provide strong support for the accurate interpretation of brain functional activities.
The core of the algorithm is the strategy of integrated motion artifact suppression and morphological optimization. First, by calculating the approximate gradient sliding standard deviation of the signal, WiMi’s motion artifact suppression and morphological optimization algorithm for fNIRS signals (fNIRS-MASMOA) is capable of detecting the presence of motion artifacts, and then applying specific processing methods for different types of artifacts and then applies specific processing methods for different types of artifacts. For peaks, the algorithm uses an improved median filtering technique to remove them efficiently, and a mathematical morphology approach to optimize the shape of the signal through morphological manipulation to make baseline mutations and slow drifts more consistent with the true characteristics of brain activity. Compared to existing methods, fNIRS-MASMOA demonstrates excellence in terms of mean square error, signal-to-noise ratio, squared Pearson correlation coefficient, and peak-to-peak error. This algorithm represents a milestone in providing researchers with a new and efficient tool to study brain activity more accurately.
The fNIRS-MASMOA mainly consists of motion artifact detection, directional median filter processing and mathematical morphology optimization correction:
Motion artifact detection: The algorithm first performs approximate gradient sliding standard deviation calculations on the original fNIRS signal to detect motion artifacts in the signal. This step aims to accurately identify types of motion artifacts such as peaks, baseline mutations and slow drifts.
Also Read: Origin AI Announces TruPresence: A New Era in Smart Living
Directed median filtering processing: Once the motion artifacts are identified, the algorithm applies directed median filtering processing for the peaks type of motion artifacts. This processing method utilizes the gradient information and local features of the signal to perform directional filtering on the peaks, effectively removing the interference of peaks on the signal analysis.
Mathematical morphology optimization correction: For motion artifacts of the baseline mutations and slow drift types, the algorithm uses mathematical morphology optimization methods for correction. This is the use of morphology to process the signal in order to eliminate the effects of baseline mutations and slow drifts on the signal morphology and features, so as to achieve accurate reconstruction and optimization of the signal.
The technical framework of WiMi’s fNIRS-MASMOA integrates the directional median filtering algorithm, mathematical morphology algorithm, and gradient analysis in signal processing to achieve accurate suppression and optimization of the original signals through the differential processing of different types of motion artifacts in the fNIRS signals. The core idea is to adopt specific processing strategies for targeted correction of different types of motion artifacts to ensure the accuracy and reliability of the fNIRS signal data, and to provide an accurate database for the subsequent analysis of brain functional activities. Its combination of directional median filtering and mathematical morphology correction gives full play to the advantages of the two methods, constructs a comprehensive processing framework, and provides a comprehensive and efficient solution to the problem of motion artifacts in fNIRS signals. By effectively suppressing and correcting the motion artifacts of fNIRS signals, the algorithm is able to improve the precision and reliability of brain functional activity analysis, providing a more reliable database for researchers and medical professionals.
WiMi‘s fNIRS-MASMOA not only provides a new technique for brain functional imaging research, but also provides a broader space for cross-research and application in related fields. It is expected to promote the expansion of the application of brain functional imaging technology in cognitive neuroscience, neuroengineering, neurofeedback and other fields, and bring new development opportunities for future brain science research and medical practice.
SOURCE: PRNewswire
