Fast Fourier Research Articles

SARDIMM: High-Speed Near-Memory Processing Architecture for Synthetic Aperture Radar Imaging

The range-Doppler algorithm (RDA), a key technique for generating synthetic aperture radar (SAR) images, offers high-resolution images but requires significant memory resources and involves complex signal processing. Moreover, the multitude of fast Fourier transform (FFT) and inverse fast Fourier transform (IFFT) operations in RDA necessitates high bandwidth and lacks data reuse, leading to bottlenecks. This paper introduces a synthetic aperture radar dual in-line memory module (SARDIMM), which executes RDA operations near memory via near-memory processing (NMP), thereby effectively reducing memory accesses, execution time, and energy consumption. The embedded NMP module in SARDIMM optionally supports a combination of FFT, IFFT, and matched filter operations of the RDA for range and azimuth compression. The operator within the NMP module accelerates the FFT by performing two radix-2 single butterfly operations in parallel. The NMP module was implemented and validated on a Xilinx UltraScale+ field-programmable gate array (FPGA) using Verilog-HDL. The acceleration performance of RDA for images of various sizes was evaluated through a simulator modified with gem5 and DRAMSim3 and achieved a 6.34–6.93× speedup and 41.9–48.2% energy savings.

Validating an Evoked Potential Platform for Electrocochleography During Cochlear Implantation.

To validate electrocochleography (ECochG) between an auditory evoked potential (AEP) machine and an established cochlear implant (CI) manufacturer ECochG system. Intraoperative validation study at a tertiary referral center. Patients included adults and children undergoing cochlear implantation. Intraoperative ECochG was measured with both the Intelligent Hearing Systems (IHS) Duet AEP machine and Cochlear Corporation (CC) ECochG platform. Recording electrodes captured extracochlear measurements through a standard facial recess. Tone-bursts were presented from 250 Hz to 2 kHz (~110 dB SPL). A fast Fourier transform (FFT) of ECochG waveforms at key frequencies was summed into a total response (ECochG-TR). Pearson's correlation was utilized to evaluate the relationship between IHS-ECochG-TR and CC-ECochG-TR after confirming normality. Thirty patients were enrolled with an average age of 67 years (SD 18.8). In the ear that was implanted, mean preoperative pure-tone average (PTA; 0.5, 1, 2, and 4 kHz) was 87.4 dB HL (SD 19.3) and mean preoperative word-recognition scores (WRS) was 17.0% correct (SD 19.1). There was strong correlation (r = 0.905, 95% confidence interval: 0.809 to 0.954) between IHS-ECochG-TR (median 2.30 μV, range 0.1-148.26) and CC-ECochG-TR (median 3.00 μV, range 0.1-239.63). Four patients underwent transtympanic ECochG with the IHS system for feasibility evaluation and achieved similar responses. Extracochlear ECochG has been predictive of CI speech perception performance. The IHS duet system is a valid measure of extracochlear ECochG for the CI population. Future work will utilize this system for measuring transtympanic ECochG to improve preoperative estimation of CI performance. 3 Laryngoscope, 2024.

Model‐based fault diagnosis for safety‐critical chemical reactors: An experimental study

AbstractThis article presents an experimental application of fault detection, isolation, and estimation in a chemical reactor system, introducing a functional observer‐based approach without the need for linear approximation. The residual signal generators, functioning as disturbance‐decoupled functional observers, provide fault size estimates and enable fault isolation through multiple generators operating independently. The experimental study focuses on the 3‐Picoline oxidation process, deriving a discrete‐time model, and constructing specific residual generators for coolant inlet temperature and feed concentration faults. Fault diagnosis employs Fast Fourier Transform (FFT) filtering and Generalized Likelihood Ratio (GLR), facilitating on‐the‐fly detection during the experiment. The effectiveness of fault detection, disturbance decoupling, and estimation is experimentally validated.

Destriping model for adaptive removal of arbitrary oriented stripes in remote sensing images

Abstract Destriping in remote sensing image processing remains a challenging problem, particularly when dealing with stripe noise of arbitrary orientations. Conventional methods struggle to eliminate oblique stripes, leaving a crucial gap in the production of higher-level remote sensing products. In response, we propose a novel destriping model, the Adaptive Stripe Noise Removal (ASNR) Method, designed to adapt to different orientations of stripe noise, aiming for accuracy, robustness, speed, and simplicity. The paper first addresses the conventional challenges in stripe removal, emphasizing the unintended loss of information during the process. To overcome this, we conduct a detailed study of stripe noise characteristics, employing traditional Fast Fourier Transform (FFT) for stripe orientation approximation. However, conventional techniques using spatial representations risk damaging detailed structures. To go beyond these limitations, the proposed method combines spectral processing technology with an image guidance mechanism. This approach aims to generate a guided image that retains both denoised features and important details. In the frequency domain, the method corrects the stripe image by estimating a guidance image. Experimental results, both qualitative and quantitative, demonstrate the superiority and stability of the proposed method in removing stripe noise and preserving image details without introducing artifacts. The novel approach fills a critical gap in destriping methods, offering a fast, accurate, and adaptable solution for arbitrary orientations of stripe noise in remote sensing images.

A novel approach to estimating harmonic/interharmonic parameters

The conventional fast Fourier transforms(FFT) approach is susceptible to noise and has a low-frequency resolution with short sampling windows. This is a bottleneck problem for accurately monitoring harmonic and interharmonic. A new method is proposed to detect the parameters of harmonic and interharmonic in this study. At the beginning, the cross-correlation function and FFT estimate the rough values of frequency and phase. Next, the error between the rough and true values is derived. Then, the incoherence of signal and noise is used to create the cost function, and the optimal initial value is computed from the rough values and the error. Finally, the initial value is used in cost function iteration, the so-called Newton–Raphson method, to calculate the frequency and phase of harmonic and interharmonic. The least squares method is used to compute the amplitudes of harmonic and interharmonic. The proposed method allows adaptively selecting the basis function for the cost function instead of using the traditional orthogonal basis. Therefore, it achieves the super-resolution evaluation of harmonic and interharmonic frequency. The simulation results demonstrate that the proposed method provides high accuracy of parameter estimation and strong noise immunity.

Optimizing real-time phase detection in diverse rhythmic biological signals for phase-specific neuromodulation.

Closed-loop, phase-specific neurostimulation is a powerful method to modulate ongoing brain activity for clinical and research applications. Phase-specific stimulation relies on estimating the phase of an ongoing oscillation in real time and issuing a control command at a target phase. Phase detection algorithms based on Fast Fourier transform (FFT) are widely used due to their computational efficiency and robustness. However, it is unclear how algorithm performance depends on the spectral properties of the input signal and how algorithm parameters can be optimized. We used offline simulation to evaluate the performance of three algorithms (endpoint-corrected Hilbert Transform, Hilbert Transform and phase mapping) on three rhythmic biological signals with distinct spectral properties (rodent hippocampal theta potential, human EEG alpha and human essential tremor). First, we found that algorithm performance was more strongly influenced by signal amplitude and frequency variation compared with signal to noise ratio. Second, our simulations showed that the size of the data window for phase estimation was critical for the performance of FFT-based algorithms, where the optimal data window corresponds to the period of the oscillation. We validated this prediction with real time phase detection of hippocampal theta oscillations in freely behaving rats performing spatial navigation. Our findings define the relationship between signal properties and algorithm performance and provide a convenient method for optimizing FFT-based phase detection algorithms.

Collision cross-section measurements of small molecules via transient decay profiles observed in Orbitrap mass analyzers.

Collision cross section (CCS) of organic compounds can be measured via Fourier transform-based mass spectrometry (MS) by modeling the decay rate of transient signals in the analyzer. Deriving CCS values of low-mass molecules (mass < 2000 Da and CCS < 500 Å2) with Orbitrap MS is challenging due to their high axial frequencies and small absolute variances in cross-sectional profiles. Here, we acquired mass spectra of progressively more complex low-mass analytes using commercial Orbitrap mass spectrometers. The transient signals were processed using Fast Fourier transform (FFT) and short-time Fourier transform (StFFT) to derive decay constants of multiple select ionic species from a single MS full-scan experiment. Decay constants were translated into CCS values using at least two internal standards in the same mass spectrum. Our results suggest target ionic species should have high S/N in order to derive CCS values with ≤0.5% uncertainty. Limitations in the precision of CCS measurements reflect local space charge effects that disturb ion motion in the analyzer. The derived CCS values of polymer like fragments of Ultramark 1621 and small molecules such as individual protonated amino acids can achieve average ±1% error with selection of internal standards across a wide mass range. Future studies need to optimize the strategy to select internal standards in order to improve the precision and accuracy of CCS measurements for small molecules via Orbitrap MS.

Optimized power and speed of Split-Radix, Radix-4 and Radix-2 FFT structures

Fast Fourier transform (FFT) is a fundamental building block for digital signal processing applications where high processing speed is crucial. Resource utilization in implementing FFT structures can be minimized by optimizing the performance of multipliers and adders used within the design. FFTs are also widely used in various machine learning algorithms. To achieve increased processor efficiency and reduced resource utilization, we propose a hardware design for Radix-2, Radix-4, and Split-Radix FFT architectures that utilizes a novel parallel prefix adder. This design offers lower power consumption, smaller chip area, and faster operation compared to existing architectures. Our performance analysis focuses on metrics such as power consumption, clock speed, and hardware complexity for Radix-2, Radix-4, and Split-Radix FFT algorithms implemented with the proposed adder. We compare these metrics using our proposed arithmetic structure against existing adder designs. The results indicate that the Split-Radix FFT architecture achieves lower power consumption and smaller chip area compared to Radix-4 and Radix-2 methods. Additionally, the Split-Radix FFT exhibits a higher clock speed. Therefore, based on these findings, the Split-Radix algorithm appears to be a compelling choice for implementation on field-programmable gate arrays due to its high speed and lower power consumption.

Role of periodic oscillating flow modulators on mixed convection in a long horizontal channel

Present study explores mixed convection characteristics in a long horizontal channel subjected to multiple periodically distributed flow modulators. The flow modulators are represented by oscillating blades placed along a centerline of the channel whose lower and upper walls are kept at constant high and low temperatures respectively. In replicating the blade oscillation, moving mesh approach has been adopted within Arbitrary Lagrangian–Eulerian (ALE) framework for a representative periodical unit. The corresponding non-dimensional governing mass, momentum and energy conservation equations have been solved through Galerkin finite element solver for a wide variations of modulator's dynamic condition (oscillating frequency and maximum angular displacement) for different fluids represented by Prandtl number. Heat transfer performance of the system has been demonstrated in terms of spatially-averaged transient as well as time-averaged Nusselt number while qualitative analysis of fluid flow and thermal field has been presented as streamline and isotherm plots. Present study indicates that the time averaged Nusselt number undergoes significant variation with blade oscillating frequency and maximum angular displacement depending on both the Prandtl number and Reynolds Number. Power spectrum analysis obtained through Fast Fourier Transformation (FFT) of the imposed blade frequency and induced thermal frequency reveals different correlation depending on the blade frequency. Blade friction power requirement has been found to increase at higher blade frequency as well as maximum angular displacement. However, contrary to power consumption, increase in frequency does not result in a significant rise in heat transfer. Consequently, specific heat transfer decreases at higher blade oscillating frequency and maximum angular displacement.

An experimental analysis of outdoor UAV localisation through diverse estimators and crowd-sensed data fusion

Motivated by the challenge of achieving precise 3D outdoor localisation for unmanned aerial vehicles (UAVs) in global navigation satellite system (GNSS)-denied environments, this paper introduces an innovative technique. Integrating crowd-sensed data fusion to counter inertial navigation system (INS) drift during GNSS signal outages, the proposed method exploits diverse estimators to enhance its efficacy. A micro lightweight frequency modulated continuous wave (FMCW) radar mounted on the UAV captures ground scatterer reflections, processed via fast Fourier transform (FFT) to generate a range-Doppler map. This map facilitates forward velocity estimation during GNSS signal loss. This approach employs adaptive thresholding, image binarisation, and connected components-based techniques for target detection from a computer vision standpoint. The derived radar-based velocity fuses with magnetometer, barometer, and inertial measurement unit (IMU) data using diverse estimators like extended Kalman filter (EKF) and particle filter (PF). Real-time flight data evaluation and simulated outage periods using EKF and PF validate the outdoor localisation system. Experimental analyses demonstrate substantial improvements, enhancing 3D positioning accuracy by 99.89% and 99.83% for the initial and subsequent flights, respectively, leveraging PF to fortify the standalone INS mode during GNSS signal loss. This approach significantly enhances UAV localisation precision, particularly in challenging GNSS-denied scenarios, showcasing the potential for real-world applications.

Schizophrenia diagnosis based on diverse epoch size resting-state EEG using machine learning.

Schizophrenia is a severe mental disorder that impairs a person's mental, social, and emotional faculties gradually. Detection in the early stages with an accurate diagnosis is crucial to remedying the patients. This study proposed a new method to classify schizophrenia disease in the rest state based on neurologic signals achieved from the brain by electroencephalography (EEG). The datasets used consisted of 28 subjects, 14 for each group, which are schizophrenia and healthy control. The data was collected from the scalps with 19 EEG channels using a 250 Hz frequency. Due to the brain signal variation, we have decomposed the EEG signals into five sub-bands using a band-pass filter, ensuring the best signal clarity and eliminating artifacts. This work was performed with several scenarios: First, traditional techniques were applied. Secondly, augmented data (additive white Gaussian noise and stretched signals) were utilized. Additionally, we assessed Minimum Redundancy Maximum Relevance (MRMR) as the features reduction method. All these data scenarios are applied with three different window sizes (epochs): 1, 2, and 5 s, utilizing six algorithms to extract features: Fast Fourier Transform (FFT), Approximate Entropy (ApEn), Log Energy entropy (LogEn), Shannon Entropy (ShnEn), and kurtosis. The L2-normalization method was applied to the derived features, positively affecting the results. In terms of classification, we applied four algorithms: K-nearest neighbor (KNN), support vector machine (SVM), quadratic discriminant analysis (QDA), and ensemble classifier (EC). From all the scenarios, our evaluation showed that SVM had remarkable results in all evaluation metrics with LogEn features utilizing a 1-s window size, impacting the diagnosis of Schizophrenia disease. This indicates that an accurate diagnosis of schizophrenia can be achieved through the right features and classification model selection. Finally, we contrasted our results to recently published works using the same and a different dataset, where our method showed a notable improvement.

Human hand gesture recognition using fast Fourier transform with coot optimization based on deep neural network

ABSTRACT Hand motion detection is particularly important for managing the movement of individuals who have limbs amputated. The existing algorithm is complex, time-consuming and difficult to achieve better accuracy. A DNN is suggested to recognize human hand movements in order to get over these problems.Initially, the raw input EMG signal is captured then the signal is pre-processed using high-pass Butterworth filter and low-pass filter which is utilized to eliminate the noise present in the signal. After that pre-processed EMG signal is segmented using sliding window which is used for solving the issue of overlapping. Then the features are extracted from the segmented signal using Fast Fourier Transform. Then selected the appropriate and optimal number of features from the feature subset using coot optimization algorithm. After that selected features are given as input for deep neural network classifier for recognizing the hand movements of human. The simulation analysis shows that the proposed method obtain 95% accuracy, 0.05% error, precision is 94%, and specificity is 92%.The simulation analysis shows that the developed approach attain better performance compared to other existing approaches. This prediction model helps in controlling the movement of amputee patients suffering from disable hand motion and improve their living standard.

Enhancing Ring Laser Gyroscope data analysis through Neural Networks for Earthquake detection and Phenomenon identification

This work focuses on developing neural networks to optimize the ring laser gyroscope data analysis, for the GINGER experiment. The first neural network aims to provide data with minimal delays suitable to detect earthquakes. Comparing this neural network with a tool implemented in Labview based on the Fast Fourier Transform, the neural network is twice as precise with the same delay. The second network learns to recognize various phenomena in the data, such as transients from laser dynamics like split modes or mode jumps. These capabilities are realized through the automated labeling of data using neural networks and the implementation of a mask that facilitates the selection of high-quality data. The curated dataset can then undergo traditional offline analysis. The third network is a map to transform earthquake signals seen by the GIGS seismometer into the signals seen by the co-located GINGERINO ring laser prototype, useful to study rotational models of earthquakes.

Enhancing Generalizability of a Machine Learning Model for Infrared Thermographic Defect Detection by Using 3D Numerical Modeling

The paper describes the implementation of 3D numerical simulation in machine learning models used in infrared thermographic nondestructive testing. The enhancement of generalizability of such models emerges as a decisive factor for producing trust-worthy test results. First, it is demonstrated that the models trained on datasets with fixed parameters yield limited defect detection capabilities. The concept of training datasets, which include subtle variations in material thickness, thermal conductivity, as well as various combinations of material density and heat capacity, provides the best learning results and a noticeable ability to identify defects in all test datasets. Second, the model robustness in respect to noise is explored to demonstrate its ability to withstand additive and multiplicative random noise. Third, potentials of some known techniques of thermographic data processing, such as Thermographic Signal Reconstruction, Fast Fourier Transform and Temperature Contrast, are examined. In particular, the use of the Temperature Contrast data ensured sensitivity (True Positive Rate) better than 98% across all test datasets.

The influence of mineral inclusion on the effective strength of rock-like geomaterials

The influences of microstructure on the macroscopic mechanical behavior of a composite with a porous matrix reinforced by mineral inclusions are investigated in the present work by both numerical and theoretical methods. The mineral inclusions are embedded at the mesoscopic scale and much bigger than the pores which are located at the microscopic scale. In order to consider the properties of the studied rock-like geomaterials, such as the dissymmetry between tension and compression, the solid phase at the microscopic scale is assumed to obey to a Drucker–Prager criterion. Based on the studied microstructure, the Fast Fourier Transform (FFT) based numerical method is firstly adopted to investigate the macroscopic plastic yield stress of the studied composite. Different microstructure having different volume fraction of inclusion, micro-porosity and frictional coefficient of the solid phase are considered. Based on these numerical results, the existing theoretical yield criterion is estimated. One finds that it should be improved for the case of a microstructure having a high inclusion content. Then, a new macroscopic yield criterion is constructed in the present work by using the modified secant method. This criterion ameliorates fundamentally the one proposed in Shen et al. (2013), specially for the case of a deviatoric loading. It is then estimated and validated by comparing with the FFT based numerical results which were carried out in this work with different volume fractions of heterogeneous phase.

A unified FFT framework with coupled gradient damage model and phase-field method for the failure analysis of composites

A unified FFT framework coupling the Gradient Damage Model (GDM) and the Variational Phase-Field Model (PFM) for failure analysis of composites is proposed. GDM and PFM are used to simulate the failure of the matrix and the interphase, respectively. Based on the similarity of the damage evolution equations (elliptic partial differential equations) of GDM and PFM, a unified iterative solution and numerical implementation based on fast Fourier transform (FFT) are derived, and simultaneously applied to a computational model. In addition, the high computational efficiency of the FFT method is used to solve the mechanical equilibrium problem. The complete form of the PFM is introduced to avoid erroneous damage diffusion at the different phase interfaces of the composites, and a stress decomposition-based PFM is formulated. 2D and 3D computational examples are established to investigate the mesh insensitivity of the damage models used and the erroneous damage diffusion at interfaces. The applicability of the stress decomposition-based PFM at interphase is discussed in the context of single-edge notched specimens under tension and fiber-reinforced composites with interfacial phases under transverse tension. Finally, the performance of the unified computational framework is evaluated using two simple numerical examples, and the mechanical behavior of unidirectional CF/PEEK composites under transverse tensile loading is simulated based on the unified framework. The results are compared with experiments to evaluate the computational accuracy and practical applicability of the proposed framework.

AutoSamp: Autoencoding k-space Sampling via Variational Information Maximization for 3D MRI.

Accelerated MRI protocols routinely involve a predefined sampling pattern that undersamples the k-space. Finding an optimal pattern can enhance the reconstruction quality, however this optimization is a challenging task. To address this challenge, we introduce a novel deep learning framework, AutoSamp, based on variational information maximization that enables joint optimization of sampling pattern and reconstruction of MRI scans. We represent the encoder as a non-uniform Fast Fourier Transform that allows continuous optimization of k-space sample locations on a non-Cartesian plane, and the decoder as a deep reconstruction network. Experiments on public 3D acquired MRI datasets show improved reconstruction quality of the proposed AutoSamp method over the prevailing variable density and variable density Poisson disc sampling for both compressed sensing and deep learning reconstructions. We demonstrate that our data-driven sampling optimization method achieves 4.4dB, 2.0dB, 0.75dB, 0.7dB PSNR improvements over reconstruction with Poisson Disc masks for acceleration factors of R = 5, 10, 15, 25, respectively. Prospectively accelerated acquisitions with 3D FSE sequences using our optimized sampling patterns exhibit improved image quality and sharpness. Furthermore, we analyze the characteristics of the learned sampling patterns with respect to changes in acceleration factor, measurement noise, underlying anatomy, and coil sensitivities. We show that all these factors contribute to the optimization result by affecting the sampling density, k-space coverage and point spread functions of the learned sampling patterns.

New Insights on the Information Content of the Normalized Difference Vegetation Index Sentinel-2 Time Series for Assessing Vegetation Dynamics

The Sentinel-2 NDVI time series information content from 2017 to 2023 at a 10 m spatial resolution was evaluated based on the NDVI temporal dependency in five scenarios in central Spain. First, time series were interpolated and then filtered using the Savitzky–Golay, Fast Fourier Transform, Whittaker, and Maximum Value filters. Temporal dependency was assessed using the Q-Ljung-Box and Fisher’s Kappa tests, and similarity between raw and filtered time series was assessed using Correlation Coefficient and Root Mean Square Error. An Interpolating Efficiency Indicator (IEI) was proposed to summarize the number and temporal distribution of low-quality observations. Type of climate, atmospheric disturbances, land cover dynamics, and management were the main sources of variability in five scenarios: (1) rainfed wheat and barley presented high short-term variability due to clouds (lower IEI in winter and spring) during the growing cycle and high interannual variability due to precipitation; (2) maize showed stable summer cycles (high IEI) and low interannual variability due to irrigation; (3) irrigated alfalfa was cut five to six times during summer, resulting in specific intra-annual variability; (4) beech forest showed a strong and stable summer cycle, despite the short-term variability due to clouds (low IEI); and (5) evergreen pine forest had a highly variable growing cycle due to fast responses to temperature and precipitation through the year and medium IEI values. Interpolation after removing non-valid observations resulted in an increase in temporal dependency (Q-test), particularly a short term in areas with low IEI values. The information improvement made it possible to identify hidden periodicities and trends using the Fisher’s Kappa test. The SG filter showed high similarity values and weak influence on dynamics, while the MVF showed an overestimation of the NDVI values.

QRCODE: Massively parallelized real-time time-dependent density functional theory for periodic systems

We present a new software module, QRCODE (Quantum Research for Calculating Optically Driven Excitations), for massively parallelized real-time time-dependent density functional theory (RT-TDDFT) calculations of periodic systems in the open-source Qbox software package. Our approach utilizes a custom implementation of a fast Fourier transformation scheme that significantly reduces inter-node message passing interface (MPI) communication of the major computational kernel and shows impressive scaling up to 16,344 CPU cores. In addition to improving computational performance, QRCODE contains a suite of various time propagators for accurate RT-TDDFT calculations. As benchmark applications of QRCODE, we calculate the current density and optical absorption spectra of hexagonal boron nitride (h-BN) and photo-driven reaction dynamics of the ozone-oxygen reaction. We also calculate the second and higher harmonic generation of monolayer and multi-layer boron nitride structures as examples of large material systems. Our optimized implementation of RT-TDDFT in QRCODE enables large-scale calculations of real-time electron dynamics of chemical and material systems with enhanced computational performance and impressive scaling across several thousand CPU cores.

A Machine Learning Prediction Model for Myelitis and Multiple Sclerosis Based on Fourier Transform Features from MRI Images

Myelitis is a neurodegenerative disease positioned in the spinal cord, with multiple sclerosis (MS) being a common subtype. Radiological indicators enable the diagnosis of these diseases. This study proposes a classification framework to detect myelitis, MS, and healthy control (HC) groups using magnetic resonance imaging (MRI) images. The feature extraction step involves applying the fast Fourier transform (FFT) to MRI images. FFT is important because it converts spatial data into the frequency domain, making it easier to identify patterns and abnormalities that indicate these diseases. Then, statistical features (mean, minimum, maximum, standard deviation, skewness, kurtosis, and total energy) are extracted from this frequency information. These features are then used to train support vector machine (SVM), k-nearest neighbor (KNN), and decision tree algorithms. In multi-class classification (myelitis vs. MS vs. HC), the proposed method achieves a classification accuracy of 99.31% with SVM, with average precision, recall, and F1-score values of 99.27%, 99.21%, and 99.24%, respectively, indicating effective classification across all classes. In the binary class classification (HC vs. MS, MS vs. myelitis, HC vs. myelitis), the SVM achieves an outstanding classification accuracy of 99.36%, 99.71%, and 100% respectively. This study highlights the efficiency of FFT-based feature extraction in forming detection patterns for classifying HC, MS, and myelitis classes.