High-pass Filter Research Articles

Development of Second Order High Frequency Filter for Partial Discharge Measurement

Partial discharge (PD) can occur on any high-voltage equipment. This study is particularly concerned with the PD that occurs on underground cables. Researchers frequently struggle to isolate interference noise from the PD signal when doing PD measurements. Therefore, to reduce the interference noise during PD measurements, Sallen-Key high-pass filters with a 500 kHz, or a little bit higher cut-off frequency are designed. This filter is designed with a designated cut-off frequency value because the frequency range of the PD signal is between 1 MHz to 1 GHz. Based on the results of experimental testing for actual filters, the designed filter is a high-pass filter with a cut-off frequency of 500KHz. When the generator function's input frequency values are set between 500 kHz and 2.4 MHz, the signal will pass the filter and provide output values for both frequencies and amplitude. However, at frequencies lower than 500 kHz, the signal is beginning to be filtered and the output signal's amplitude is getting close to zero. Based on the PD testing results, there is interference on the PD signal shown on the oscilloscope when a 1kV supply is applied to an underground cable without using a filter. But, when the proposed filter is used in PD testing, the ripple voltage is reduced from 123 mV to 67 mV. This proves that when the designed filter is used in PD testing, it can reduce interference noise on the PD signal.

Image Color Balance with Laplacian and Gaussian Pyramid (CBLGP) Algorithm

In recent times, technology has rapidly and significantly evolved across all sectors. Digital image processing stands out as a modern technology aimed at achieving clear images. However, digitized images often encounter issues of low quality, such as unclear or underwater images that require enhancement for better visibility. These problems stem from factors like deficient focusing, lighting, and various constraints leading to low contrast, shading, and artifacts. Underwater and satellite images consistently face less-than-ideal conditions due to environmental factors like light refraction in water, particle scattering, and dust in aquatic environments. Similarly, challenges in space, such as poor illumination and lack of contrast, further complicate image analysis. Overcoming these obstacles is crucial for extracting valuable information, necessitating advanced processing techniques. This paper introduces an enhanced Gaussian/Laplace color balance-fusion algorithm designed to improve image visibility. Modifications to certain equations result in sharper and clearer images. The algorithm begins by determining the white balance of the input RGB color image and subsequently enhances its intensity. Edge improvement is carried out separately using a depth filter. The weights for each image are then determined and combined to form a Laplace Pyramid. A color restoration technique is applied to process the resulting image, producing the final enhanced image. While existing methods for image contrast enhancement typically focus on image features, they often neglect user characteristics. This paper explores the application of image sharpening, a prominent image enhancement technique, in clearing underwater or low-quality images using the proposed algorithm. Keywords: EDSHE, High pass filter, White patch ratix, Laplacian pyramid, Color restoration.

Low-voltage solution-processed NaxCu1−xI thin-film transistors for mimicking synaptic plasticity

In this article, NaxCu1−xI thin-film transistors gated by chitosan for low-voltage operation have been created by doping CuI with different Na concentrations (x = 0, 0.05, 0.1). It is found that the devices exhibit optimal performance when x is 0.05: a large current on/off ratio of 1.62 × 105, a steep subthreshold slope of 17.72 mV/dec, a saturation field-effect mobility of 0.51 cm2/V s, and a threshold voltage of 1.10 V. The operating voltage of the devices is reduced to below 2 V due to the electric-double-layer (EDL) effect. At a frequency of 10 Hz, a maximum specific capacitance of 1.36 μF/cm2 can be obtained in the chitosan. The effects of bias stress and laser on Na0.05Cu0.95I thin-film transistors (TFTs) have been examined at the end of the article, and the results revealed that Na0.05Cu0.95I TFTs possess good stability. As the bias stress gets longer and the laser power increases, the transfer curves of TFTs shift positively. Also, artificial synaptic behaviors and functions have been simulated experimentally. It can be divided into single-pulse, double-pulse, and multiple-pulse. According to the experimental results, features such as short-term plasticity, long-term plasticity, paired-pulse facilitation, high-pass filtering, pulse logic, and spatial summation have been achieved. The electrostatic modulation of EDL due to proton transverse migration is vital for this simulation. The realization of low-voltage synaptic Na0.05Cu0.95I TFTs prepared by solution method with pulse logic and spatial summation functions is crucial for application of portable biosensors and neuromorphic systems.

Reinforcement-Learning-Based Virtual Inertia Controller for Frequency Support in Islanded Microgrids

As the world grapples with the energy crisis, integrating renewable energy sources into the power grid has become increasingly crucial. Microgrids have emerged as a vital solution to this challenge. However, the reliance on renewable energy sources in microgrids often leads to low inertia. Renewable energy sources interfaced with the network through interlinking converters lack the inertia of conventional synchronous generators, and hence, need to provide frequency support through virtual inertia techniques. This paper presents a new control algorithm that utilizes the reinforcement learning agents Twin Delayed Deep Deterministic Policy Gradient (TD3) and Deep Deterministic Policy Gradient (DDPG) to support the frequency in low-inertia microgrids. The RL agents are trained using the system-linearized model and then extended to the nonlinear model to reduce the computational burden. The proposed system consists of an AC–DC microgrid comprising a renewable energy source on the DC microgrid, along with constant and resistive loads. On the AC microgrid side, a synchronous generator is utilized to represent the low inertia of the grid, which is accompanied by dynamic and static loads. The model of the system is developed and verified using Matlab/Simulink and the reinforcement learning toolbox. The system performance with the proposed AI-based methods is compared to conventional low-pass and high-pass filter (LPF and HPF) controllers.

Simple spike patterns and synaptic mechanisms encoding sensory and motor signals in Purkinje cells and the cerebellar nuclei

Whisker stimulation in awake mice evokes transient suppression of simple spike probability in crus I/II Purkinje cells. Here, we investigated how simple spike suppression arises synaptically, what it encodes, and how it affects cerebellar output. Invitro, monosynaptic parallel fiber (PF)-excitatory postsynaptic currents (EPSCs) facilitated strongly, whereas disynaptic inhibitory postsynaptic currents (IPSCs) remained stable, maximizing relative inhibitory strength at the onset of PF activity. Short-term plasticity thus favors the inhibition of Purkinje spikes before PFs facilitate. Invivo, whisker stimulation evoked a 2-6ms synchronous spike suppression, just 6-8ms (∼4 synaptic delays) after sensory onset, whereas active whisker movements elicited broadly timed spike rate increases that did not modulate sensory-evoked suppression. Firing in the cerebellar nuclei (CbN) inversely correlated with disinhibition from sensory-evoked simple spike suppressions but was decoupled from slow, non-synchronous movement-associated elevations of Purkinje firing rates. Synchrony thus allows the CbN to high-pass filter Purkinje inputs, facilitating sensory-evoked cerebellar outputs that can drive movements.

A Complete Pipeline for Heart Rate Extraction from Infant ECGs

Infant electrocardiograms (ECGs) and heart rates (HRs) are very useful biosignals for psychological research and clinical work, but can be hard to analyse properly, particularly longform (≥5 min) recordings taken in naturalistic environments. Infant HRs are typically much faster than adult HRs, and so some of the underlying frequency assumptions made about adult ECGs may not hold for infants. However, the bulk of publicly available ECG approaches focus on adult data. Here, existing open source ECG approaches are tested on infant datasets. The best-performing open source method is then modified to maximise its performance on infant data (e.g., including a 15 Hz high-pass filter, adding local peak correction). The HR signal is then subsequently analysed, developing an approach for cleaning data with separate sets of parameters for the analysis of cleaner and noisier HRs. A Signal Quality Index (SQI) for HR is also developed, providing insights into where a signal is recoverable and where it is not, allowing for more confidence in the analysis performed on naturalistic recordings. The tools developed and reported in this paper provide a base for the future analysis of infant ECGs and related biophysical characteristics. Of particular importance, the proposed solutions outlined here can be efficiently applied to real-world, large datasets.

High-performance IGZO/In2O3 NW/IGZO phototransistor with heterojunctions architecture for image processing and neuromorphic computing

The development of high-performance neuromorphic phototransistors is of paramount importance for image perception and depth memory learning. Here, based on metal-oxide heterojunction architecture, artificial synaptic phototransistors with synaptic plasticity have been achieved, demonstrating an artificial synapse that integrates central and optic nerve functions. Thanks to the sensitive light-detection properties, the optical power consumption of such photonic artificial synapses can be as low as 22 pico-joules, which is extremely competitive compared with other pure metal oxide photoelectric synapses ever reported. What is more, owing to its good short-term (STP) and tunable amplitude-frequency characteristics, the as-constructed device can function as a biomimetic high-pass filter for picture edge detection. Dual-mode synaptic modulation has been performed, combining photonic pulse with gate voltage stimulus. After photoelectric-synergistic modulation, the high synaptic weights enable the device to simulate complex neural learning rules for neuromorphic applications, including gesture recognition, image perception in the visual system, and classically conditioned reflexes. These results suggest that the current oxide-based heterojunction architecture displays potential application in future multifunction neuromorphic devices and systems.

A Novel Method for Identifying Resonance Frequency Band in Weak Bearing Fault Diagnosis of Electric Driving System

Because of the uncertainty of the structure and environment of the electric driving system (EDS), the fault signature of the rotating mechanism is complicated. A novel method based on Hilbert transform with modified fast kurtogram (HTMFK), which is used for identifying the bearing faults in the EDS, is proposed. The modified principle and algorithm flow of the proposed method are derived. A high pass filter based on the frequency band identified by HTMFK is constructed and applied to fault diagnosis. Simulation signals demonstrate the ability of demodulating signals and identifying the fault resonance band. The bearing fault bench experiment of EDS is carried out in a semianechoic chamber. The corresponding fault tests are conducted according to different operating conditions. The applicability of HTMFK is verified by comparing the square envelope spectrums. Compared with other methods, the proposed method identifies the fault resonance frequency band more effectively and expands the application range of bearing fault diagnosis in EDS.

A Cryogenic Wideband (2.5–14 GHz) Receiver System for the Arecibo Observatory 12 m Telescope

AbstractIn this paper we present details of the construction of a wideband, cryogenic receiver and its successful commissioning on the Arecibo Observatory 12 m telescope. The cryogenic receiver works in the 2.5–14 GHz frequency range. We upgraded the current narrow band, room temperature receivers of the telescope with the new wideband receiver. The current receiver is built around a Quadruple‐Ridged Flared Horn (QRFH) developed by Akgiray et al. (2013, https://doi.org/10.1109/tap.2012.2229953). To mitigate strong radio frequency interference (RFI) below 2.7 GHz, we installed a highpass filter before the first stage low noise amplifier (LNA). The QRFH, highpass filter, noise coupler and LNA are located inside a cryostat and are cooled to 15 K. The measured receiver temperature is 25 K (median value) over 2.5–14 GHz. The system temperature measured at zenith is about 40 K near 3.1 and 8.6 GHz and the zenith antenna gains are 0.025 and 0.018 K/Jy at the two frequencies respectively. We recommend the following improvements to the telescope system: (a) Upgrade the highpass filter to achieve better RFI rejection near 2.5 GHz; (b) Improve aperture efficiency at 8.6 GHz; (c) Upgrade the intermediate frequency system to increase the upper frequency of operation from 12 to 14 GHz.

Unveiling the Role of Side Chain for Improving Nonvolatile Characteristics of Conjugated Polymers-Based Artificial Synapse.

Interest has grown in services that consume a significant amount of energy, such as large language models (LLMs), and research is being conducted worldwide on synaptic devices for neuromorphic hardware. However, various complex processes are problematic for the implementation of synaptic properties. Here, synaptic characteristics are implemented through a novel method, namely side chain control of conjugated polymers. The developed devices exhibit the characteristics of the biological brain, especially spike-timing-dependent plasticity (STDP), high-pass filtering, and long-term potentiation/depression (LTP/D). Moreover, the fabricated synaptic devices show enhanced nonvolatile characteristics, such as long retention time (≈102 s), high ratio of Gmax/Gmin, high linearity, and reliable cyclic endurance (≈103 pulses). This study presents a new pathway for next-generation neuromorphic computing by modulating conjugated polymers with side chain control, thereby achieving high-performance synaptic properties.

Impact of enhancement filters of a CMOS system on halo artifact expression at the bone-to-implant interface.

To assess the impact of enhancement filters on the formation of halo artifacts in radiographs of dental implants obtained with a complementary metal oxide semiconductor (CMOS) system. Digital radiographs of dental implants placed in dry human mandibles were processed with the Noise Reduction smoothing filter, as well as the Sharpen 1, Sharpen 4, and Sharpen UM high-pass filters available in the CLINIVIEW™ software (Instrumentarium Dental, Tuusula, Finland). Subjective analysis involved evaluating the left, right, and apical surfaces of each implant for the presence of much, few, or no halo. The objective analysis involved measurement of the halo area using the Trainable Weka Segmentation plugin (ImageJ, National Institutes of Health, Bethesda, MD, USA). Data were analyzed using Friedman's test (subjective analysis) and ANOVA (objective analysis) (α = 5%). In the subjective evaluation, the Sharpen 4 filter produced more radiographs with much halo present, and in the objective evaluation, a bigger halo area when compared to the original images and the Noise Reduction filter for all surfaces (p < 0.05). When evaluating dental implants, priority should be given to original images and those enhanced with smoothing filters since they exhibit fewer halo artifacts. Post-processing tools, such as enhancement filters, may improve the image quality and assist some diagnostic tasks. However, little is known regarding the impact of enhancement filters in halo formation on CMOS systems, which have been increasingly used in dental offices.

Fault Diagnosis of Linear Guide Rails Based on SSTG Combined with CA-DenseNet

Monitoring the status of linear guide rails is essential because they are important components in linear motion mechanical production. Thus, this paper proposes a new method of conducting the fault diagnosis of linear guide rails. First, synchrosqueezing transform (SST) combined with Gaussian high-pass filter, termed as SSTG, is proposed to process vibration signals of linear guide rails and obtain time-frequency images, thus helping realize fault feature visual enhancement. Next, the coordinate attention (CA) mechanism is introduced to promote the DenseNet model and obtain the CA-DenseNet deep learning framework, thus realizing accurate fault classification. Comparison experiments with other methods reveal that the proposed method has a high classification accuracy of up to 95.0%. The experimental results further demonstrate the effectiveness and robustness of the proposed method for the fault diagnosis of linear guide rails.

Voltage-Mode Multifunction and Universal Biquads with Two CFOAs

In this paper, a new voltage-mode (VM) multifunction filter including only two grounded capacitors, three resistors, and two current feedback operational amplifiers is introduced. This filter with one high input impedance and two low output impedances can simultaneously provide band-pass filter (BPF) and low-pass filter (LPF) responses with an adjustable gain. A universal VM filter topology with three inputs and one output generating high-pass filter (HPF), BPF, LPF, notch filter, and all-pass filter (APF) responses is obtained from the multifunction one. It has the same number of active and passive elements as the multifunction and a low output impedance. Sensitivities of both introduced filters are low enough. However, the APF of the introduced universal filter has a passive element-matching problem. As an example, parasitic impedance effects are considered for the LPF of the universal filter, while non-ideal gain analyses are given for the HPF. To see how the introduced universal filter works, countless simulations are performed through the SPICE (Simulation Program with Integrated Circuit Emphasis) program in which 0.13 [Formula: see text]m IBM CMOS technology parameters are utilized. DC supply voltages are selected as ± 0.9 V. Also, several experiments using commercially available AD844s with ± 10 V supply voltages are carried out to see the performance.

Toward quantitative intrafractional monitoring in paraspinal SBRT using a proprietary software application: clinical implementation and patient results

Objective. We report on paraspinal motion and the clinical implementation of our proprietary software that leverages Varian’s intrafraction motion review (IMR) capability for quantitative tracking of the spine during paraspinal SBRT. The work is based on our prior development and analysis on phantoms. Approach. To address complexities in patient anatomy, digitally reconstructed radiographs (DRR’s) that highlight only the spine or hardware were constructed as tracking reference. Moreover, a high-pass filter and first-pass coarse search were implemented to enhance registration accuracy and stability. For evaluation, 84 paraspinal SBRT patients with sites spanning across the entire vertebral column were enrolled with prescriptions ranging from 24 to 40 Gy in one to five fractions. Treatments were planned and delivered with 9 IMRT beams roughly equally distributed posteriorly. IMR was triggered every 200 or 500 MU for each beam. During treatment, the software grabbed the IMR image, registered it with the corresponding DRR, and displayed the motion result in near real-time on auto-pilot mode. Four independent experts completed offline manual registrations as ground truth for tracking accuracy evaluation. Main results. Our software detected ≥1.5 mm and ≥2 mm motions among 17.1% and 6.6% of 1371 patient images, respectively, in either lateral or longitudinal direction. In the validation set of 637 patient images, 91.9% of the tracking errors compared to manual registration fell within ±0.5 mm in either direction. Given a motion threshold of 2 mm, the software accomplished a 98.7% specificity and a 93.9% sensitivity in deciding whether to interrupt treatment for patient re-setup. Significance. Significant intrafractional motion exists in certain paraspinal SBRT patients, supporting the need for quantitative motion monitoring during treatment. Our improved software achieves high motion tracking accuracy clinically and provides reliable guidance for treatment intervention. It offers a practical solution to ensure accurate delivery of paraspinal SBRT on a conventional Linac platform.

A low-cost dual bandpass planar filter for WiMAX and mobile communications

This study proposes a new design of low-cost dual-bandpass filter for worldwide interoperability and microwave access (WiMAX) band at &lt;br /&gt; 3.50 GHz and mobile communications band at 1.19 GHz. A high pass filter and a stopband filter make up the new dual-bandpass filter structure. Different theoretical studies were carried out for the design of the proposed filter. This filter’s base is a RO5880 substrate with a dielectric permittivity constant of 2.2, loss tangent of 0.0009 and 1.6 mm thickness. High mashing density was used to validate the various simulated structures while accounting for two numerical methods: the moment technique and the finite integration method. The final circuit's overall dimensions are 60×178,3 mm&lt;sup&gt;2&lt;/sup&gt;.

Real-time anomaly detection of short-term traffic disruptions in urban areas through adaptive isolation forest

The escalating congestion impacts of short-term traffic disruptions, such as double parking or short-duration work zones, are gaining increased attention. This study introduces an enhanced isolation forest-based unsupervised anomaly detection algorithm to detect short-term traffic disruptions in urban areas. To enable the detection in real-time, a sliding window approach is introduced to allow the streaming data to be read and processed according to the window size. An adaptive threshold that automatically adjusts itself to overcome the limitations of the miss rate on local anomalies is shown to further enhance the model predictions. The proposed algorithm is empirically validated on four study sites in Manhattan, New York City. The experimental results demonstrate that the proposed unsupervised algorithm can effectively detect different types of traffic anomalies including accidents, work zones/road closures, traffic jams, double parking events and police activities. On all sites, the average detection rate is 81.1% for traffic jams and 89.6% for police activities, respectively. For three out of the four sites, the detection rate ranges from 71.4% to 100% for accidents, work zones and double parking. An optimizer using high-pass filter is also presented to further improve off-line detection. The primary advantages of this proposed computationally-efficient method are that its only required data input is travel time information and that it does not need labeled data for training, which make it highly deployable for real-time operational applications and can be easily adopted by other cities.

Optimal filters for ERP research II: Recommended settings for seven common ERP components.

In research with event-related potentials (ERPs), aggressive filters can substantially improve the signal-to-noise ratio and maximize statistical power, but they can also produce significant waveform distortion. Although this tradeoff has been well documented, the field lacks recommendations for filter cutoffs that quantitatively address both of these competing considerations. To fill this gap, we quantified the effects of a broad range of low-pass filter and high-pass filter cutoffs for seven common ERP components (P3b, N400, N170, N2pc, mismatch negativity, error-related negativity, and lateralized readiness potential) recorded from a set of neurotypical young adults. We also examined four common scoring methods (mean amplitude, peak amplitude, peak latency, and 50% area latency). For each combination of component and scoring methods, we quantified the effects of filtering on data quality (noise level and signal-to-noise ratio) and waveform distortion. This led to recommendations for optimal low-pass and high-pass filter cutoffs. We repeated the analyses after adding artificial noise to provide recommendations for data sets with moderately greater noise levels. For researchers who are analyzing data with similar ERP components, noise levels, and participant populations, using the recommended filter settings should lead to improved data quality and statistical power without creating problematic waveform distortion.

A Multi Representation Deep Learning Approach for Epileptic Seizure Detection

Epileptic seizures, unpredictable in nature and potentially dangerous during activities like driving, pose significant risks to individual and public safety. Traditional diagnostic methods, which involve labour-intensive manual feature extraction from Electroencephalography (EEG) data, are being supplanted by automated deep learning frameworks. This paper introduces an automated epileptic seizure detection framework utilizing deep learning to bypass manual feature extraction. Our framework incorporates detailed pre-processing techniques: normalization via L2 normalization, filtering with an 80 Hz and 0,5 Hz Butterworth low-pass and high-pass filter, and a 50 Hz IIR Notch filter, channel selection based on standard deviation calculations and Mutual Information algorithm, and frequency domain transformation using FFT or STFT with Hann windows and 50% hop. We evaluated on two datasets: the first comprising 4 canines and 8 patients with 2.299 ictal, 23.445 interictal, and 32.915 test data, 400-5000Hz sampling rate across 16-72 channels; the second dataset, intended for testing, 733 icatal, 4.314 interictal, and 1908 test data, each 10 minutes long, recorded at 400Hz across 16 channels. Three deep learning architectures were assessed: CNN, LSTM, and a hybrid CNN-LSTM model-stems from their demonstrated efficacy in handling the complex nature of EEG data. Each model offers unique strengths, with the CNN excelling in spatial feature extraction, LSTM in temporal dynamics, and the hybrid model combining these advantages. The CNN model, comprising 31 layers, yielded highest accuracy, achieving 91% on the first dataset (precision 92%, recall 91%, F1-score 91%) and 82% on the second dataset using a 30-second threshold. This threshold was chosen for its clinical relevance. The research advances epileptic seizure detection using deep learning, indicating a promising direction for future medical technology. Future work will focus on expanding dataset diversity and refining methodologies to build upon these foundational results.

Harmonics Cancelation in 3-Phase 4-Wire System using Modified Reference Frame

In this paper, A modified synchronous reference frame (SRF) or (d-q theory) with Self-Tuning Filter (STF) for the three-phase four wires system is presented to control of 4-legs shunt active power filter (SAPF) under balanced supply voltages. The main goal of the paper is to apply this control method using unit vector generation with STF instead of PLL circuit. In utility power system, it has been observed that harmonics build a major role in reducing the power quality. These harmonics caused by nonlinear loads. Many consumers appliances demand quality of power continuously for their operation. The performance of the end user equipment is heavily dependent on the quality of power supplied to it. This power is affected by various external and internal factors. They are like voltage and frequency variations, faults, outages etc. The solution to overcome these problems is to filter out these harmonics. For this purpose, there are many filters' topologies and control methods are proposed and studied in the literature to eliminate harmonics currents. In our study, 4-legs shunt active power filter is proposed and controlled by d-q method with modified unit vector generation. The SAPF is proposed to solve problems of line harmonic currents and neutral line currents of three phase 4-wire power distribution system. The use of STF in the modified control method can optimize the performance of this topology regarding to classical one (high pass and low pass filters). The proposed control strategy is simulated in MATLAB Simulink and the results are presented.

Enhancing Nigrosome-1 Sign Identification via Interpretable AI using True Susceptibility Weighted Imaging.

Nigrosome 1 (N1), the largest nigrosome region in the ventrolateral area of the substantia nigra pars compacta, is identifiable by the "N1 sign" in long echo time gradient echo MRI. The N1 sign's absence is a vital Parkinson's disease (PD) diagnostic marker. However, it is challenging to visualize and assess the N1 sign in clinical practice. To automatically detect the presence or absence of the N1 sign from true susceptibility weighted imaging by using deep-learning method. Prospective. 453 subjects, including 225 PD patients, 120 healthy controls (HCs), and 108 patients with other movement disorders, were prospectively recruited including 227 males and 226 females. They were divided into training, validation, and test cohorts of 289, 73, and 91 cases, respectively. 3D gradient echo SWI sequence at 3T; 3D multiecho strategically acquired gradient echo imaging at 3T; NM-sensitive 3D gradient echo sequence with MTC pulse at 3T. A neuroradiologist with 5 years of experience manually delineated substantia nigra regions. Two raters with 2 and 36 years of experience assessed the N1 sign on true susceptibility weighted imaging (tSWI), QSM with high-pass filter, and magnitude data combined with MTC data. We proposed NINet, a neural model, for automatic N1 sign identification in tSWI images. We compared the performance of NINet to the subjective reference standard using Receiver Operating Characteristic analyses, and a decision curve analysis assessed identification accuracy. NINet achieved an area under the curve (AUC) of 0.87 (CI: 0.76-0.89) in N1 sign identification, surpassing other models and neuroradiologists. NINet localized the putative N1 sign within tSWI images with 67.3% accuracy. Our proposed NINet model's capability to determine the presence or absence of the N1 sign, along with its localization, holds promise for enhancing diagnostic accuracy when evaluating PD using MR images. 2 TECHNICAL EFFICACY: Stage 1.