Fine Control Research Articles

Landslide Hazard Assessment for Wanzhou Considering the Correlation of Rainfall and Surface Deformation

The landslide hazard assessment plays a crucial role in landslide risk mitigation and land use planning. The result of landslide hazard assessment corrected by surface deformation, obtained through time-series InSAR, has usually proven to have good application capabilities. However, the issue lies in the uncertainty of InSAR results, where some deformations cannot be calculated, and some are not true deformations. This uncertainty of InSAR results will lead to errors in landslide hazard assessment. Here, we attempt to evaluate landslide hazards by considering combined rainfall and surface deformation. The main objective of this research was to mitigate the impact of bias and explore the accurate landslide hazard assessment method. A total of 201 landslides and 11 geo-environment factors were utilized for landslide susceptibility assessment by support vector machine (SVM) model in Wanzhou District, Three Gorges Reservoir Area (TGRA). The preliminary hazard is obtained by analyzing the statistical data of landslides and rainfall. Based on the SAR image data of Sentinel-1A satellites from September 2019 to October 2021, the SBAS-InSAR method was used to analyze surface deformation. The correlation between surface deformation and rainfall was analyzed, and the deformation factor variables were applied to landslide hazard assessment. The research results demonstrate that the error caused by the uncertainty of InSAR results can be effectively avoided by analyzing the relationship between rainfall and surface deformation. Our results can effectively adjust and correct the hazard results and eliminate the errors in the general hazard assessment. Our proposed method can be used to assess the landslide hazard in more detail and provide a reference for fine risk management and control.

Modulation Steering Motion by Quantitative Electrical Stimulation in Pigeon Robots.

The pigeon robot has attracted significant attention in the field of animal robotics thanks to its outstanding mobility and adaptive capability in complex environments. However, research on pigeon robots is currently facing bottlenecks, and achieving fine control over the motion behavior of pigeon robots through brain-machine interfaces remains challenging. Here, we systematically quantify the relationship between electrical stimulation and stimulus-induced motion behaviors, and provide an analytical method to demonstrate the effectiveness of pigeon robots based on electrical stimulation. In this study, we investigated the influence of gradient voltage intensity (1.2-3.0 V) on the indoor steering motion control of pigeon robots. Additionally, we discussed the response time of electrical stimulation and the effective period of the brain-machine interface. The results indicate that pigeon robots typically exhibit noticeable behavioral responses at a 2.0 V voltage stimulus. Increasing the stimulation intensity significantly controls the steering angle and turning radius (p < 0.05), enabling precise control of pigeon robot steering motion through stimulation intensity regulation. When the threshold voltage is reached, the average response time of a pigeon robot to the electrical stimulation is 220 ms. This study quantifies the role of each stimulation parameter in controlling pigeon robot steering behavior, providing valuable reference information for the precise steering control of pigeon robots. Based on these findings, we offer a solution for achieving precise control of pigeon robot steering motion and contribute to solving the problem of encoding complex trajectory motion in pigeon robots.

Conditioning diffusion models via attributes and semantic masks for face generation

Deep generative models have shown impressive results in generating realistic images of faces. GANs managed to generate high-quality, high-fidelity images when conditioned on semantic masks, but they still lack the ability to diversify their output. Diffusion models partially solve this problem and are able to generate diverse samples given the same condition. This paper introduces a novel strategy for enhancing diffusion models through multi-conditioning, harnessing cross-attention mechanisms to utilize multiple feature sets, ultimately enabling the generation of high-quality and controllable images. The proposed method extends previous approaches by introducing conditioning on both attributes and semantic masks, ensuring finer control over the generated face images. In order to improve the training time and the generation quality, the impact of applying perceptual-focused loss weighting into the latent space instead of the pixel space is also investigated. The proposed solution has been evaluated on the CelebA-HQ dataset, and it can generate realistic and diverse samples while allowing for fine-grained control over multiple attributes and semantic regions. Experiments on the DeepFashion dataset have also been performed in order to analyze the capability of the proposed model to generalize to different domains. In addition, an ablation study has been conducted to evaluate the impact of different conditioning strategies on the quality and diversity of the generated images.

An Angle-Oriented Approach to Transferring Speech to Gesture for Highly Anthropomorphized Embodied Conversational Agents

Realistic co-speech gestures are important to anthropomorphize ECAs, as nonverbal behavior improves expressiveness of their speech greatly. However, the existing approaches to generating co-speech gestures with sufficient details (including fingers, etc.) in 3D scenarios are indeed rare. Additionally, they hardly address the problem of abnormal gestures, temporal–spatial coherence and diversity of gesture sequences comprehensively. To handle abnormal gesture issues, we put forward an angle conversion method to remove body part length from the original in-the-wild video dataset via transferring coordinates of human upper body key points into relative deflection angles and pitch angles. We also propose a neural network called HARP with encoder–decoder architecture to transfer MFCC featured speech audio into aforementioned angles on the basis of CNN and LSTM. The angles then can be rendered as corresponding co-speech gestures. Compared with the other latest approaches, the co-speech gestures generated by HARP are proved to be almost as good as the real person, i.e., they have strong temporal–spatial coherence, diversity, persuasiveness and credibility. Our approach puts finer control on co-speech gestures than most of the existing works by handling all key points of the human upper body. It is more feasible for industrial application, since HARP can be adaptive to any human upper body model. All related code and evidence videos of HARP can be accessed at https://github.com/drrobincroft/HARP .

Ultrauniform Plating of Lithium on 10-nm-Scale Ordered Carbon Grids for Long Lifespan Lithium Metal Batteries.

Tailorable lithium (Li) nucleation and uniform early-stage plating is essential for long-lifespan Li metal batteries. Among factors influencing the early plating of Li anode, the substrate is critical, but a fine control of the substrate structure on a scale of ≈10nm has been rarely achieved. Herein, a carbon consisting of ordered grids is prepared, as a model to investigate the effect of substrate structure on the Li nucleation. In contrast to the individual spherical Li nuclei formed on the flat graphene, an ultrauniform and nuclei-free Li plating is obtained on the ordered carbon with a grid size smaller than the thermodynamical critical radius of Li nucleation (≈26nm). Simultaneously, an inorganic-rich solid-electrolyte-interphase is promoted by the cross-sectional carbon layers of such ordered grids which are exposed to the electrolyte. Consequently, the carbon grids with a grid size of ≈10nm show a favorable cycling stability for more than 1100 cycles measured at 2mA cm-2 in a half cell. With LiNi0.8Co0.1Mn0.1O2 as cathode, the assembled full cell with a cathode capacity of 3 mAh cm-2 and a negative/positive ratio of 1.67 demonstrates a stable cycling for over 130 cycles with a capacity retention of 88%.

Controlled Stiffness of Direct-Write, Near-Field Electrospun Gelatin Fibers Generates Differences in Tenocyte Morphology and Gene Expression.

Tendinopathy is a leading cause of mobility issues. Currently, the cell-matrix interactions involved in the development of tendinopathy are not fully understood. In vitro tendon models provide a unique tool for addressing this knowledge gap as they permit fine control over biochemical, micromechanical, and structural aspects of the local environment to explore cell-matrix interactions. In this study, direct-write, near-field electrospinning of gelatin solution was implemented to fabricate micron-scale fibrous scaffolds that mimic native collagen fiber size and orientation. The stiffness of these fibrous scaffolds was found to be controllable between 1 MPa and 8 MPa using different crosslinking methods (EDC, DHT, DHT+EDC) or through altering the duration of crosslinking with EDC (1 h to 24 h). EDC crosslinking provided the greatest fiber stability, surviving up to 3 weeks in vitro. Differences in stiffness resulted in phenotypic changes for equine tenocytes with low stiffness fibers (∼1 MPa) promoting an elongated nuclear aspect ratio while those on high stiffness fibers (∼8 MPa) were rounded. High stiffness fibers resulted in the upregulation of matrix metalloproteinase (MMPs) and proteoglycans (possible indicators for tendinopathy) relative to low stiffness fibers. These results demonstrate the feasibility of direct-written gelatin scaffolds as tendon in vitro models and provide evidence that matrix mechanical properties may be crucial factors in cell-matrix interactions during tendinopathy formation.

The relationships between motor skills and executive functions in children with and without autism spectrum disorder.

To date, information on associations between motor skills and executive functions (EF) in autistic children is limited. The purpose of this study was to compare motor skills and EF performance between autistic children and typically developing (TD) children and to examine the relationships between motor skills and EF in these two groups. Forty-eight autistic children and 48 TD children aged 6 to 12 years were recruited for this study. Motor skills were measured with the Bruininks-Oseretsky Test of Motor Proficiency-2 (BOT-2). EF was assessed with the Stroop Color and Word Test, the Wisconsin Card Sorting Task (WCST), and the Test of Attentional Performance: Go/No-go test. Independent sample t-tests were used to compare the BOT-2 scores and EF measures between autistic children and TD children. Pearson product-moment correlation and regressions were conducted to assess the relationships between the BOT-2 scores and the EF measures for each group. Results showed that autistic children scored significantly lower than TD children on all four BOT-2 composite scores and a total motor composite. Autistic children also demonstrated significantly lower levels of performance on all EF measures than TD children. Further, autistic children showed more significant associations between motor skills and EF than TD children, particularly pronounced in the domains of fine manual control and manual coordination to cognitive flexibility, as well as manual coordination and inhibitory control. Continued development of motor skills and EF in autistic children is important. The relationships between motor skills and EF were significant among autistic children, suggesting future research on promoting EF through motor skill interventions in autistic children is required.

Laser shock-enabled optical–thermal–mechanical coupled welding method for silver nanowires

Silver nanowires (AgNWs) are recognized as highly promising materials for flexible and transparent electrode applications. However, existing material-processing methods fail to achieve uniform and reliable AgNWs junctions. In this study, we propose a new method using the laser shock effect combined with the laser heating effect, for creating AgNW junctions within thin films. We explored the welding mechanism of AgNWs through optic-thermal welding, laser shock-enabled mechanical welding, and laser-shock-enabled optical-thermal-mechanical (LS-OTM) experiments, as well as numerical simulations, and the results demonstrate that the innovative mechanism of the LS-OTM process lies in its utilization of laser shock to adjust the gap between the nanowire junctions, which in turn achieves a fine control of the thermal effect of the heating laser localised surface plasmon resonance, and the atomic diffusion in the solid state at intermediate temperature under the action of the impact force is the mechanism of the formation of high-quality junctions. We prepared flexible transparent conductive films and studied their transmittance, conductivity, and thermal properties, the results show that the flexible transparent conductive films prepared by LS-OTM welding method have excellent transmittance, conductivity, and thermal properties, this verifies the feasibility and effectiveness of this processing strategy. The LS-OTM method is a viable solution for manufacturing transparent, conductive films from AgNWs for emerging applications such as flexible heated films, flexible displays, and wearable medical devices.

Water Level Inversion Detection Method for Water Level Images without a Scale in Complex Environments

Accurately perceiving changes in water level information is key to achieving the fine control of water and flooding; however, the existing technology cannot achieve water level recognition in complex and harsh environments, such as at night; in haze, rain, or snow; or during obscuration by floating objects or shadows. Therefore, on the basis of a deep analysis of the characteristics of water level images in complex and harsh environments, in this study, we took full advantage of a deep learning network’s ability to characterise semantic features and carried out exploratory research on water level detection in no-water-ruler scenarios based on the two technical means of target detection and semantic segmentation. The related experiments illustrate that all the methods proposed in this study can effectively adapt to complex and harsh environments. The results of this study are valuable for applications in solving the difficulties of accurate water level detection and flood disaster early warnings in poor-visibility scenarios.

Cannabinoids regulate an insula circuit controlling water intake

The insular cortex, or insula, is a large brain region involved in the detection of thirst and the regulation of water intake. However, our understanding of the topographical, circuit, and molecular mechanisms for controlling water intake within the insula remains parcellated. We found that type-1 cannabinoid (CB1) receptors in the insular cortex cells participate in the regulation of water intake and deconstructed the circuit mechanisms of this control. Topographically, we revealed that the activity of excitatory neurons in both the anterior insula (aIC) and posterior insula (pIC) increases in response to water intake, yet only the specific removal of CB1 receptors in the pIC decreases water intake. Interestingly, we found that CB1 receptors are highly expressed in insula projections to the basolateral amygdala (BLA), while undetectable in the neighboring central part of the amygdala. Thus, we recorded the neurons of the aIC or pIC targeting the BLA (aIC-BLA and pIC-BLA) and found that they decreased their activity upon water drinking. Additionally, chemogenetic activation of pIC-BLA projection neurons decreased water intake. Finally, we uncovered CB1-dependent short-term synaptic plasticity (depolarization-induced suppression of excitation [DSE]) selectively in pIC-BLA, compared with aIC-BLA synapses. Altogether, our results support a model where CB1 receptor signaling promotes water intake by inhibiting the pIC-BLA pathway, thereby contributing to the fine top-down control of thirst responses.

Fault Detection and Isolation in Simulated Batch Operation of Fine Motion Control Rod Drives

Control rods and elements manage the power distribution in nuclear reactors through the motion of banks of rods distributed throughout the core. These positional changes are achieved through the actuation of fine motion control rod drive (FMCRD) mechanisms. In the BWRX-300 design by GE-Hitachi, this mechanism is electrically driven by a servomotor that allows for high-precision control of power outputs. Under operational transients, such as load-following, accurate and precise operation of these servomotors is necessary over long periods of time, so they are key maintenance targets to maintain availability and operational flexibility. Swiftly and precisely identifying faults in the drive mechanisms will support predictive maintenance and reduced costs. This paper used three different types of simulated faults to test the fault detectability of principal component analysis (PCA) when considering the simulated operations of banks of control rods and their associated servomotors. These faults were stator short-circuit faults, ball screw jam faults, and ball screw wear faults. Torque and position were monitored in the simulation. The position signal was insufficient to detect mechanical faults. Torque signals for each servomotor in the bank of rods undergoing multiple position demand changes were projected to a reduced dimensional space via PCA. Q and T2 statistics were employed for anomaly detection. Using this approach, all faults were detected, and the anomalies were isolated to the faulty FMCRD mechanism.

Subthalamic stimulation modulates context-dependent effects of beta bursts during fine motor control

Increasing evidence suggests a considerable role of pre-movement beta bursts for motor control and its impairment in Parkinson’s disease. However, whether beta bursts occur during precise and prolonged movements and if they affect fine motor control remains unclear. To investigate the role of within-movement beta bursts for fine motor control, we here combine invasive electrophysiological recordings and clinical deep brain stimulation in the subthalamic nucleus in 19 patients with Parkinson’s disease performing a context-varying task that comprised template-guided and free spiral drawing. We determined beta bursts in narrow frequency bands around patient-specific peaks and assessed burst amplitude, duration, and their immediate impact on drawing speed. We reveal that beta bursts occur during the execution of drawing movements with reduced duration and amplitude in comparison to rest. Exclusively when drawing freely, they parallel reductions in acceleration. Deep brain stimulation increases the acceleration around beta bursts in addition to a general increase in drawing velocity and improvements of clinical function. These results provide evidence for a diverse and task-specific role of subthalamic beta bursts for fine motor control in Parkinson’s disease; suggesting that pathological beta bursts act in a context dependent manner, which can be targeted by clinical deep brain stimulation.

STOPGAP: an open-source package for template matching, subtomogram alignment and classification.

Cryo-electron tomography (cryo-ET) enables molecular-resolution 3D imaging of complex biological specimens such as viral particles, cellular sections and, in some cases, whole cells. This enables the structural characterization of molecules in their near-native environments, without the need for purification or separation, thereby preserving biological information such as conformational states and spatial relationships between different molecular species. Subtomogram averaging is an image-processing workflow that allows users to leverage cryo-ET data to identify and localize target molecules, determine high-resolution structures of repeating molecular species and classify different conformational states. Here, STOPGAP, an open-source package for subtomogram averaging that is designed to provide users with fine control over each of these steps, is described. In providing detailed descriptions of the image-processing algorithms that STOPGAP uses, this manuscript is also intended to serve as a technical resource to users as well as for further community-driven software development.

Application performance of three common solid-state electrolytes in batteries

Due to the hidden benefits such as the possibilities to make energy density high, safety improved, and lifespan extended, solid-state batteries are a focal point in battery technology research. Current liquid electrolyte-based lithium-ion batteries, despite their maturity, have inherent issues like safety risks and limited energy density. While the advancement in solid-state battery technology can potentially overcome certain challenges, it also encounters its own issues including high contact interface impedance and ion transmission efficiency. For solid-state batteries to supplant traditional lithium-ion batteries in the future, these challenges must be addressed. The research provides an in-depth analysis of the three major types of solid-state electrolytes currently prevalent in the market, including oxide, sulfide, and polymer electrolytes. This is especially true for the electrochemical analysis of batteries after the electrolyte has been used. Each has its unique properties, synthesis methods, and limitations. Oxide electrolytes excel in high-temperature ionic conductivity but underperform at room temperature. Moreover, the synthesis of oxide solid electrolytes requires fine control over temperature and atmosphere, thus demanding high technical expertise. The sulfide category shows promise in room temperature conductivity, albeit with stability issues. Polymer electrolytes, however, are flexible and processable but have generally lower ionic conductivity. To enable large-scale applications, future research needs to focus on improving these electrolytes’ performance and developing cost-effective and efficient synthesis methods. The importance of research into solid-state lithium-ion batteries is tied to their capacity to transform the energy storage sector.

Polyoxometallate Cluster Induced High-Entropy Oxide Sub-1 nm Nanosheets as Photoelectrocatalysts for Zn-Air Batteries.

The lack of highly efficient and inexpensive catalysts severely hinders the large-scale application of Zn-air batteries (ZABs). High-entropy oxides (HEOs) exhibit unique structures and attractive properties; thus, they are promising to be used in ZABs. However, conventional high-temperature synthesis methods tend to obtain microscale HEOs with a lower exposure rate of active sites. Here, we report a facile solvothermal strategy for preparing two-dimensional (2D) HEO sub-1 nm nanosheets (SNSs) induced by polyoxometalate (POM) clusters. Taking advantage of the special 2D sub-1 nm structure and precise element regulation, these 2D HEOs-POM SNSs exhibit enhanced bifunctional oxygen evolution and oxygen reduction reaction activity under light irradiation. Further applying these 2D HEOs-POM SNSs to ZABs as cathode catalysts, the CoFeNiMnCuZnOx-phosphomolybdic acid SNSs-based ZABs deliver a low charge/discharge voltage gap of 0.25 V at 2 mA cm-2 under light irradiation. Meanwhile, it could maintain an ultralong-term stability for 1600 h at 2 mA cm-2 and 930 h at 10 mA cm-2. The 2D sub-1 nm structure and fine element control in HEOs provide opportunities to solve the problems of low intrinsic activity, limited active sites, and instability of air cathodes in ZABs.

Numerical investigation of heat transfer and temperature distribution in a Microwave-Heated Heli-Flow reactor and experimental validation

Microwave-heated flow systems (MWFSs), offering improved process control, enhanced product yields, and the potential for scalable and sustainable manufacturing routes, have gained significant popularity in continuously manufacturing nanomaterials and their hybrids. Temperature control in a microwave zone can maintain fine control over the manufactured materials' properties; however, the current state-of-the-art designs lack monitoring of the temperature distribution throughout the flow reactor. Herein, a unique model of numerical estimations of temperature distribution in a microwave-heated system featuring a helical flow milli-reactor, Heli-Flow, is described and validated by experimentally acquired temperature profiles. The proposed numerical model accurately predicts the temperature profile along the Heli-Flow and the steady-state outlet temperature at the MW cavity's exit. The helical design of the reactor promotes homogeneous heating and eliminates orientation-related discrepancies. The maximum outlet temperatures achieved at different microwave powers and FRs align well with expected trends. The modeled temperatures closely match the experimental and analytical measurements, indicating the effectiveness of the simulation approach. This research contributes to understanding temperature distribution in MWFSs and offers insights for optimizing nanomaterial synthesis and other applications.

Directional multiobjective optimization of metal complexes at the billion-system scale.

The discovery of transition metal complexes (TMCs) with optimal properties requires large ligand libraries and efficient multiobjective optimization algorithms. Here we provide the tmQMg-L library, containing 30k diverse and synthesizable ligands with robustly assigned charges and metal coordination modes. tmQMg-L enabled the generation of 1.37 million palladium TMCs, which were used to develop and benchmark the Pareto-Lighthouse multiobjective genetic algorithm (PL-MOGA). With fine control over aim and scope, this algorithm maximized both the polarizability and highest occupied molecular orbital-lowest unoccupied molecular orbital gap of the TMCs within selected regions of the Pareto front, without requiring prior knowledge on the objective limits. Instead of genetic operations on small ligand fragments, the PL-MOGA did whole-ligand mutation and crossover operations, which in chemical spaces containing billions of systems, yielded thousands of highly diverse TMCs in an interpretable manner.

Fine tuning of Nb-incorporated TiO2 thin films by atomic layer deposition and application as efficient electron transport layer in perovskite solar cells

Access to finely tuned thin films that can act as electron transport layer (ETL) and adapt to the absorber composition and whole cell fabrication process is key to achieve efficient perovskite-based solar cells. In this study, the growth of mixed niobium-titanium oxide (Nb-TiO2) thin films by atomic layer deposition and its use to extract photogenerated electrons is reported. Films were obtained at 200 °C from titanium (IV) i-propoxide, (t-butylimido)tris(diethylamido)niobium(V), and water by introducing Nb2O5 growth cycle in a TiO2 matrix. Process parameters (order of precursor introduction, cycle ratio) were optimized; the growth mechanism and the effective Nb incorporation were investigated by an in situ quartz crystal microbalance and x-ray photoelectron spectroscopy. The composition, morphology, structural, and optoelectronic properties of the as-deposited films were determined using a variety of characterization techniques. As a result, a fine control of the film properties (between TiO2 and Nb2O5 ones) could be achieved by tuning Nb content. To allow a successful implementation in solar devices, a comprehensive annealing study under several conditions (temperatures, various atmospheres) was conducted leading to an evolution of the optical properties due to a morphological change. Ultimately, the incorporation of these 15 nm-thick films in mesoscopic perovskite solar cells as ETL shows an improvement of the cell performances and of their stability with increasing Nb content, in comparison of both TiO2 and Nb2O5 pure compounds, reaching power conversion efficiency up to 18.3% and a stability above 80% of its nominal value after 138 h under illumination.

Serum biomarkers in patients with drug-resistant epilepsy: a proteomics-based analysis.

To investigate the serum biomarkers in patients with drug-resistant epilepsy (DRE). A total of 9 DRE patients and 9 controls were enrolled. Serum from DRE patients was prospectively collected and analyzed for potential serum biomarkers using TMT18-labeled proteomics. After fine quality control, bioinformatics analysis was conducted to find differentially expressed proteins. Pathway enrichment analysis identified some biological features shared by differential proteins. Protein-protein interaction (PPI) network analysis was further performed to discover the core proteins. A total of 117 serum differential proteins were found in our study, of which 44 were revised upwards and 73 downwards. The up-regulated proteins mainly include UGGT2, PDIA4, SEMG1, KIAA1191, CCT7 etc. and the down-regulated proteins mainly include ROR1, NIF3L1, ITIH4, CFP, COL11A2 etc. Pathway enrichment analysis identified that the upregulated proteins were mainly enriched in processes such as immune response, extracellular exosome, serine-type endopeptidase activity and complement and coagulation cascades, and the down-regulated proteins were enriched in signal transduction, extracellular exosome, zinc/calcium ion binding and metabolic pathways. PPI network analysis revealed that the core proteins nodes include PRDX6, CAT, PRDX2, SOD1, PARK7, GSR, TXN, ANXA1, HINT1, and S100A8 etc. The discovery of these differential proteins enriched our understanding of serum biomarkers in patients with DRE and potentially provides guidance for future targeted therapy.

Individual-finger motor imagery classification: a data-driven approach with Shapley-informed augmentation

Objective. Classifying motor imagery (MI) tasks that involve fine motor control of the individual five fingers presents unique challenges when utilizing electroencephalography (EEG) data. In this paper, we systematically assess the classification of MI functions for the individual five fingers using single-trial time-domain EEG signals. This assessment encompasses both within-subject and cross-subject scenarios, supported by data-driven analysis that provides statistical validation of the neural correlate that could potentially discriminate between the five fingers. Approach. We present Shapley-informed augmentation, an informed approach to enhance within-subject classification accuracy. This method is rooted in insights gained from our data-driven analysis, which revealed inconsistent temporal features encoding the five fingers MI across sessions of the same subject. To evaluate its impact, we compare within-subject classification performance both before and after implementing this augmentation technique. Main results. Both the data-driven approach and the model explainability analysis revealed that the parietal cortex contains neural information that helps discriminate the individual five fingers’ MI apart. Shapley-informed augmentation successfully improved classification accuracy in sessions severely affected by inconsistent temporal features. The accuracy for sessions impacted by inconsistency in their temporal features increased by an average of 26.3%±6.70 , thereby enabling a broader range of subjects to benefit from brain–computer interaction (BCI) applications involving five-fingers MI classification. Conversely, non-impacted sessions experienced only a negligible average accuracy decrease of 2.01±5.44% . The average classification accuracy achieved is around 60.0% (within-session), 50.0% (within-subject) and 40.0% (leave-one-subject-out). Significance. This research offers data-driven evidence of neural correlates that could discriminate between the individual five fingers MI and introduces a novel Shapley-informed augmentation method to address temporal variability of features, ultimately contributing to the development of personalized systems.