Multilayer Architecture Research Articles

TB-DROP: deep learning-based drug resistance prediction of Mycobacterium tuberculosis utilizing whole genome mutations

The most widely practiced strategy for constructing the deep learning (DL) prediction model for drug resistance of Mycobacterium tuberculosis (MTB) involves the adoption of ready-made and state-of-the-art architectures usually proposed for non-biological problems. However, the ultimate goal is to construct a customized model for predicting the drug resistance of MTB and eventually for the biological phenotypes based on genotypes. Here, we constructed a DL training framework to standardize and modularize each step during the training process using the latest tensorflow 2 API. A systematic and comprehensive evaluation of each module in the three currently representative models, including Convolutional Neural Network, Denoising Autoencoder, and Wide & Deep, which were adopted by CNNGWP, DeepAMR, and WDNN, respectively, was performed in this framework regarding module contributions in order to assemble a novel model with proper dedicated modules. Based on the whole-genome level mutations, a de novo learning method was developed to overcome the intrinsic limitations of previous models that rely on known drug resistance-associated loci. A customized DL model with the multilayer perceptron architecture was constructed and achieved a competitive performance (the mean sensitivity and specificity were 0.90 and 0.87, respectively) compared to previous ones. The new model developed was applied in an end-to-end user-friendly graphical tool named TB-DROP (TuBerculosis Drug Resistance Optimal Prediction: https://github.com/nottwy/TB-DROP), in which users only provide sequencing data and TB-DROP will complete analysis within several minutes for one sample. Our study contributes to both a new strategy of model construction and clinical application of deep learning-based drug-resistance prediction methods.

Deep learning for sentiment analysis on IMDB movie reviews using N-gram features

In the rapidly evolving digital landscape, the synergy of deep learning techniques and abundant datasets has opened new frontiers in various domains. This research delves into the film industry, specifically harnessing the potential of the International Movie DataBase (IMDB) dataset for sentiment analysis. Through a deep learning paradigm, we embark on sentiment classification of movie reviews, discerning between positive and negative sentiments. By navigating data preprocessing and N-gram feature extraction, we engineer a deep learning model comprising embedding, global average pooling, and multi-layer dense architectures. The experimental results underscore the model's prowess in sentiment analysis, emphasizing its capacity to empower informed decision-making within the film industry.

An image encryption algorithm based on multi-layered chaotic maps and its security analysis

This study introduces a novel symmetric image encryption algorithm that employs a multi-layered architecture of chaotic maps, incorporating both confusion and diffusion components. The selected chaotic maps, namely Aizawa, Ricker, Sine-Circle, and Chirikov, have been well-established as effective sources of chaos in the existing literature. Within the proposed algorithm, these chaotic maps are strategically utilised in distinct layers to enhance the security of the encrypted image. The algorithm demonstrates applicability to real-world scenarios, distinguished by its simplicity, high-speed performance, and low computational complexity. The efficacy of the method is underlined by its dependence on both original images and secret keys, rendering it resilient against brute force and differential attacks. Rigorous experimentation, encompassing correlation, histogram analysis, NBCR, NPCR, UACI, PSNR, and NIST values, establishes the proposed encryption algorithm's robustness and security against diverse attack modalities, including differential and statistical approaches. These findings position the algorithm as a compelling solution for safeguarding sensitive information in various practical applications.

Satellite-Terrestrial Integrated 6G: An Ultra-Dense LEO Networking Management Architecture

With the rapid development of low earth orbit (LEO) satellites, the ultra-dense LEO satellite-terrestrial integrated networking has become a promising paradigm to provide wide coverage, high capacity and flexible services for the sixth generation (6G) mobile communication networks. However, the natural heterogeneity, high dynamics, and large scale of ultradense LEO satellite networks pose new challenges to network control and management. To this end, in this article, we propose a medium earth orbit (MEO), low earth orbit (LEO) and satellite Earth stations (SESs) integrated multi-layered management architecture for the ultra-dense LEO satellite network, where global controllers and local controllers are introduced to reduce the network management complexity. With the aid of this hybrid multi-layered management framework, we can efficiently implement the network status control, mobility management, resource management, and service management in the satelliteterrestrial integrated network. To obtain proper global and local controllers with high management efficiency in the hybrid multilayered network management architecture, we further propose a novel and efficient grouping and clustering method for the ultradense LEO satellite network, where each group manager MEO satellite and cluster head (CH) LEO satellites are respectively considered as the global and local controllers of each LEO satellite group. Numerical simulations are carried out to evaluate the superiority and effectiveness of the proposed hybrid multilayered ultra-dense LEO networking management architecture.

Strong plasmon-mechanical coupling through standing acoustic waves and antenna enhancement, toward mass transduction

Plasmon-mechanical resonators are frequently employed in sensor development. Here, I report the coupling of plasmon and mechanical modes in a multilayer architecture. In the experiment, the multilayer comprised indium tin oxide (ITO) and silver (Ag) mirrors placed on the top and bottom of a SiO2 substrate. By activating a Fabry–Pérot hybrid mode and surface plasmon polariton in a Kretschmann configuration, I demonstrated the generation of a hybrid plasmon-mechanical mode by plasmonic dispersion and through strong coupling, which resulted in mechanical distortion at multiple frequencies. Adsorption of rhodamine G6 on Ag modifies the optomechanical spectrum; it causes eigenmodes to split, resulting in a coupling rate greater than the optical decay rate (κ∗<g). Besides, the excitonic modes were observed to interact with plasmon-mechanical modes, especially on-resonance, where the plasmon-mechanical modes exhibited the higher intensity. For example, the out-of-phase flexural mode (iii, g0=427×2πkHz) became active because of the coupling of the dye. Meanwhile, after dye absorption, the strong in-phase compressional mode (iv, g0=560×2πkHz) exhibited the widest splitting and an increase in the side peaks, which merged when the pump power was raised. The system was then analyzed to determine the ideal experimental settings for detecting mass by the oscillator in a controlled experiment. The mass sensor responsivity was evaluated to be F=1017Hzg−1. Advantages and disadvantages of this approach are presented.

Time-Series Forecasting Based on Multi-Layer Attention Architecture

Time-Series Forecasting Based on Multi-Layer Attention Architecture

Validation of a novel web-based application for hepatic arterial infusion floxuridine dosing.

15 Background: Hepatic arterial infusion (HAI) floxuridine (FUDR) dosing is a complex, multifactorial process with weight-based dosing dependent on the fold change of aspartate aminotransferase, alkaline phosphatase, and total bilirubin relative to reference values from preceding cycles and compounded over time. Given the complexity of dosing calculations and lack of experience with this treatment in community centers, HAI is typically administered only at tertiary oncology centers with limited access for rural disadvantaged patients. We developed an online tool to automate the dosing of FUDR to broaden safe and effective administration of HAI. Methods: The FUDR dosing algorithm developed at Memorial Sloan Kettering Cancer Center was digitalized as eDoseHAI, a novel multi-layer architecture web application tool featuring JavaScript (front-end), C# RESTful API (back-end) and Azure SQL server (data storage with Splunk data lake event registration). To validate this tool, manual dosing from patients who received ≥1 dose of HAI FUDR at our institution 6/2020-8/2023 was retrospectively compared to app-based dosing. Mismatch between app-based dosing and historical manual dosing was designated as dosing errors, excluding intentional adjustments based on physician discretion. Results: Internal validity was tested by entering records into eDoseHAI from 54 patients who received a total of 370 FUDR cycles (median 6; range 1-19). Ten patients (18.5%) were found to have unintentional dosing errors (e.g., delayed dose reductions) during treatment, impacting 52 cycles (14.1%). The median cycle at first error was 3 (range 2-6), impacting a median of 5 cycles (range 3-12). Dosing errors occurred in only 11.1% of the 45 patients supervised by a senior HAI physician, compared to 55.6% of the 9 patients supervised by three less experienced HAI physicians. Dosing accuracy improved over time; among the 14 patients on active HAI therapy, only 1 had an identified dosing error. eDoseHAI was able to reliably and accurately calculate dosing for all patients. Conclusions: HAI FUDR dosing is nuanced and complex, such that dosing errors occurred even with senior HAI oncologists. eDoseHAI is a web-based application tool designed to improve accuracy and thus safety of HAI dosing. We plan to further validate eDoseHAI using external data from an international HAI consortium, and to conduct a phase II clinical trial in which our team supervises FUDR dosing by community oncologists using eDoseHAI's supervisory mode with a goal to disseminate HAI therapy to disadvantaged rural patients who lack access.

Research on a Fault Diagnosis Method for the Braking Control System of an Electric Multiple Unit Based on Deep Learning Integration

A fault diagnosis method based on deep learning integration is proposed focusing on fault text data to effectively improve the efficiency of fault repair and the accuracy of fault localization in the braking control system of an electric multiple unit (EMU). First, the Borderline-SMOTE algorithm is employed to synthesize minority class samples at the boundary, addressing the data imbalance and optimizing the distribution of data within the fault text. Then, a multi-dimensional word representation is generated using the multi-layer bidirectional transformer architecture from the pre-training model, BERT. Next, BiLSTM captures bidirectional context semantics and, in combination with the attention mechanism, highlights key fault information. Finally, the LightGBM classifier is employed to reduce model complexity, enhance analysis efficiency, and increase the practicality of the method in engineering applications. An experimental analysis of fault data from the braking control system of the EMU indicates that the deep learning integration method can further improve diagnostic performance.

Advancing solar energy forecasting with modified ANN and light GBM learning algorithms

<abstract> <p>In the evolving field of solar energy, precise forecasting of Solar Irradiance (SI) stands as a pivotal challenge for the optimization of photovoltaic (PV) systems. Addressing the inadequacies in current forecasting techniques, we introduced advanced machine learning models, namely the Rectified Linear Unit Activation with Adaptive Moment Estimation Neural Network (RELAD-ANN) and the Linear Support Vector Machine with Individual Parameter Features (LSIPF). These models broke new ground by striking an unprecedented balance between computational efficiency and predictive accuracy, specifically engineered to overcome common pitfalls such as overfitting and data inconsistency. The RELAD-ANN model, with its multi-layer architecture, sets a new standard in detecting the nuanced dynamics between SI and meteorological variables. By integrating sophisticated regression methods like Support Vector Regression (SVR) and Lightweight Gradient Boosting Machines (Light GBM), our results illuminated the intricate relationship between SI and its influencing factors, marking a novel contribution to the domain of solar energy forecasting. With an R<sup>2</sup> of 0.935, MAE of 8.20, and MAPE of 3.48%, the model outshone other models, signifying its potential for accurate and reliable SI forecasting, when compared with existing models like Multi-Layer Perceptron, Long Short-Term Memory (LSTM), Multilayer-LSTM, Gated Recurrent Unit, and 1-dimensional Convolutional Neural Network, while the LSIPF model showed limitations in its predictive ability. Light GBM emerged as a robust approach in evaluating environmental influences on SI, outperforming the SVR model. Our findings contributed significantly to the optimization of solar energy systems and could be applied globally, offering a promising direction for renewable energy management and real-time forecasting.</p> </abstract>

MODELING THE INFLUENCE OF NANOPARTICLES AND GYROTACTIC MICROORGANISMS ON NATURAL CONVECTION IN A HEATED SQUARE CAVITY USING ARTIFICIAL NEURAL NETWORK (ANN)

The phenomenon of natural convection, which is widely used in nature and engineering applications, is a current issue that can be encountered in every field of daily life. In this study, the natural convection characteristics of a complex liquid containing nanoparticles and gyrotactic microorganisms in a heated square cavity were investigated using artificial neural network approach. The Nusselt number, average Sherwood number of nanoparticles, and average Sherwood number of microorganisms were considered as natural convection parameters and an artificial neural network model was developed to estimate these values. The Lewis number, Brownian motion parameter, thermophoresis parameter, and buoyancy ratio parameter values were defined as input parameters in the network model, which has a multilayer perceptron architecture developed with a total of 24 datasets. There were 10 neurons in the hidden layer of the network model, which has a Bayesian regularization training algorithm. The outputs obtained from the network model were compared with the target values; in addition, the prediction performance of the model was extensively analyzed using various performance parameters. It was seen that the predicted values obtained from the neural network and the target values were in an ideal harmony. On the other hand, the value of the coefficient of determination for the network model was 0.99999% and the mean deviation rates were lower than -0.03%. The results of the study showed that the developed neural network model can predict the natural convection parameters discussed with high accuracy.

Advantages and machine learning role in assessing the enterprises competitiveness in the digital economy

This work examines the possibilities and advantages using neural networks to assess the competitive enterprise’s ability in the digital economy. This issue is relevant, considering the progressing digital economy development and the data accumulation on the enterprise’s activities. Analysis of this data with machine learning methods will make it possible to effectively assess the enterprises competitiveness and make effective management decisions that will bring the company to a high level not only in different regions, but also in the country. To achieve this goal, it is necessary to identify the main factors that have the greatest impact on the business. Such factors include, for example, the introduction electronic document management, business processes robotization using the RPA platform, or their automation using programming languages. The most popular and convenient programming languages for solving similar problems are Python, C#, PowerShell. The latter offers a good approach for communicating with Windows OS and automating system administration tasks. Among the main factors influencing competitiveness, a total of 5 most significant factors were identified for the task: the financial position of the enterprise, the image of the organization, the quality of services, the personnel qualifications level and organization’s digitalization level. Competitiveness classes are proposed to determine the enterprise competitiveness. Three classes were defined for classification: Below_Norm, Normal, Above_Norm. To solve the problem, a multilayer neural network architecture has been developed, the implementation of which is implemented in Python. The use of such a solution makes it possible to achieve an accuracy of more than 90% when solving the classification task. To confirm the neural network functionality, a method for constructing a confusion matrix has been implemented. This matrix shows the percentage of objects that are accurately classified according to the incorrectly detected. Attached are the results of the neural network when solving the problem of classifying competitiveness. According to the data obtained, it was concluded that the values fed to the input classifier may be classified inaccurately, because they are on the boundaries of the results between the observed classes. The data for training neural networks was generated randomly and had a matrix size of 30x5. Since the purpose of the work is to argue for a machine learning approach to solve determining competitiveness problem, there was no need to use real data. The k-nearest neighbors method was chosen as the classification method, since this method shows high accuracy in solving such problems.

Crop yield forecasting using neural networks trained on the basis of agrometeorological and agrochemical data

In this study, a neural network model was developed and investigated for predicting crop yields based on data on weather conditions, the use of fertilizers and the content of basic nutrients in the soil (nitrogen, phosphorus and potassium). The research is based on the use of a multilayer perceptron architecture with Rely activation functions for hidden layers and linear activation for the output layer. The evaluation of the model quality was carried out using the mean square error (MSE), which was 0.5783 in the test sample, demonstrating high accuracy of predictions. Visualization of the results included analysis of scatter plots, residuals, histograms of residuals and comparison of distributions of actual and predicted values. The results obtained confirm the effectiveness of the proposed model for yield forecasting tasks, which makes it a valuable tool for optimizing agricultural production.

Artificial Intelligence and Blockchain Enabled Smart Healthcare System for Monitoring and Detection of COVID-19 in Biomedical Images.

Millions of individuals around the world have been impacted by the ongoing coronavirus outbreak, known as the COVID-19 pandemic. Blockchain, Artificial Intelligence (AI), and other cutting-edge digital and innovative technologies have all offered promising solutions in such situations. AI provides advanced and innovative techniques for classifying and detecting symptoms caused by the coronavirus. Additionally, Blockchain may be utilized in healthcare in a variety of ways thanks to its highly open, secure standards, which permit a significant drop in healthcare costs and opens up new ways for patients to access medical services. Likewise, these techniques and solutions facilitate medical experts in the early diagnosis of diseases and later in treatments and sustaining pharmaceutical manufacturing. Therefore, in this work, a smart blockchain and AI-enabled system is presented for the healthcare sector that helps to combat the coronavirus pandemic. To further incorporate Blockchain technology, a new deep learning-based architecture is designed to identify the virus in radiological images. As a result, the developed system may offer reliable data-gathering platforms and promising security solutions, guaranteeing the high quality of COVID-19 data analytics. We created a multi-layer sequential deep learning architecture using a benchmark data set. In order to make the suggested deep learning architecture for the analysis of radiological images more understandable and interpretable, we also implemented the Gradient-weighted Class Activation Mapping (Grad-CAM) based colour visualization approach to all of the tests. As a result, the architecture achieves a classification accuracy rate of 0.96, thus producing excellent results.

PEB-DDI: A Task-Specific Dual-View Substructural Learning Framework for Drug-Drug Interaction Prediction.

Adverse drug-drug interactions (DDIs) pose potential risks in polypharmacy due to unknown physicochemical incompatibilities between co-administered drugs. Recent studies have utilized multi-layer graph neural network architectures to model hierarchical molecular substructures of drugs, achieving excellent DDI prediction performance. While extant substructural frameworks effectively encode interactions from atom-level features, they overlook valuable chemical bond representations within molecular graphs. More critically, given the multifaceted nature of DDI prediction tasks involving both known and novel drug combinations, previous methods lack tailored strategies to address these distinct scenarios. The resulting lack of adaptability impedes further improvements to model performance. To tackle these challenges, we propose PEB-DDI, a DDI prediction learning framework with enhanced substructure extraction. First, the information of chemical bonds is integrated and synchronously updated with the atomic nodes. Then, different dual-view strategies are selected based on whether novel drugs are present in the prediction task. Particularly, we constructed Molecular fingerprint-Molecular graph view for transduc- tive task, and Bipartite graph-Molecular graph view for inductive task. Rigorous evaluations on benchmark datasets underscore PEB-DDI's superior performance. Notably, on DrugBank, it achieves an outstanding accuracy rate of 98.18% when predicting previously unknown interactions among approved drugs. Even when faced with novel drugs, PEB-DDI consistently exhibits outstanding generalization capabilities with an accuracy rate of 88.06%, attributing to the proper migrating of molecular basic structure learning.

DEEP SIDE-A DEEP LEARNING FRAMEWORK FOR DRUG SIDE EFFECT PREDICTION

Drug failures due to unforeseen adverse effects at clinical trials pose health risks for the participants and lead to substantial financial losses. Side effect prediction algorithms have the potential to guide the drug design process. LINCS L1000 dataset provides a vast resource of cell line gene expression data perturbed by different drugs and creates a knowledge base for context specific features. The state-of-the-art approach that aims at using context specific information relies on only the high quality experiments in LINCS L1000 and discards a large portion of the experiments. In this study, our goal is to boost the prediction performance by utilizing this data to its full extent. We experiment with 5 deep learning architectures. We find that a multimodal architecture produces the best predictive performance among multi-layer perceptron-based architectures when drug chemical structure (CS), and the full set of drug perturbed gene expression profiles (GEX) are used as modalities. Overall, we observe that the CS is more informative than the GEX. A convolutional neural network-based model that uses only SMILES string representation of the drugs achieves the best results and provides 13:0% macro-AUC and 3:1% micro-AUC improvements over the state-of-the-art. We also show that the model is able to predict side effect-drug pairs that are reported in the literature but was missing in the ground truth side effect dataset.

Multi-Layer Architecture for Enhancing Crop Quality with AI and IoT: A Structural Modelling Approach

Conventional crop management methods must be improved to address the increasing global food requirements. The exponential growth of the population exacerbates the issue at hand, the impacts of climate change and inadequate farming practices. This study analyzes the key determinants contributing to establishing a comprehensive framework for using Internet of Things (IoT) technology in the agricultural sector. The proposed Multi-Layer Architecture for Crop Quality (MLA-CQ) employs a modified version of the Total Interpretive Structural Modelling (mv-TISM) methodology to achieve this objective. This research used a mv-TISM approach to build and analyze the interrelationships among various factors that influence the adoption of IoT technology in the agriculture industry. This study introduces Artificial Intelligence (AI) by incorporating soft sensors into a remote sensing framework via deep learning. The initial data has undergone pre-processing procedures to identify and address missing values and perform data cleaning and noise reduction on the picture data obtained from farmland. Following the feature representation, a categorization procedure was performed employing an ensemble design. The suggested approach has been used to conduct experimental trials on various crops, resulting in a computing time reduction of 62%, accuracy of 95.2%, precision of 91.3 %, recall of 92.3%, and an F score of 93.1%.

Artificial Intelligence and Panendoscopy-Automatic Detection of Clinically Relevant Lesions in Multibrand Device-Assisted Enteroscopy.

Device-assisted enteroscopy (DAE) is capable of evaluating the entire gastrointestinal tract, identifying multiple lesions. Nevertheless, DAE's diagnostic yield is suboptimal. Convolutional neural networks (CNN) are multi-layer architecture artificial intelligence models suitable for image analysis, but there is a lack of studies about their application in DAE. Our group aimed to develop a multidevice CNN for panendoscopic detection of clinically relevant lesions during DAE. In total, 338 exams performed in two specialized centers were retrospectively evaluated, with 152 single-balloon enteroscopies (Fujifilm®, Porto, Portugal), 172 double-balloon enteroscopies (Olympus®, Porto, Portugal) and 14 motorized spiral enteroscopies (Olympus®, Porto, Portugal); then, 40,655 images were divided in a training dataset (90% of the images, n = 36,599) and testing dataset (10% of the images, n = 4066) used to evaluate the model. The CNN's output was compared to an expert consensus classification. The model was evaluated by its sensitivity, specificity, positive (PPV) and negative predictive values (NPV), accuracy and area under the precision recall curve (AUC-PR). The CNN had an 88.9% sensitivity, 98.9% specificity, 95.8% PPV, 97.1% NPV, 96.8% accuracy and an AUC-PR of 0.97. Our group developed the first multidevice CNN for panendoscopic detection of clinically relevant lesions during DAE. The development of accurate deep learning models is of utmost importance for increasing the diagnostic yield of DAE-based panendoscopy.

Soccer Training Optimization for Education: A Multi-Layer Architecture Simulating How Observers Understand Soccer Sceneries

Soccer Training Optimization for Education: A Multi-Layer Architecture Simulating How Observers Understand Soccer Sceneries

Exode des cerveaux et traumas culturels dans Avalon : Les secrets des émigrants heureux par Bogdan Suceavă (I)

Despite its impact on the education, research and health system, Romanian brain drain is still neglected in both the political agendas and cultural debates. Furthermore, the theme is almost absent within contemporary literature starting with the writings anchored in the transition period, which is precisely that of the first significant waves of labor and high-skilled emigration, but also that of some major social and cultural traumas. In this context, Avalon: The Secrets of the Happy Émigrés (2018) by Bogdan Suceavă fill up a main cultural gap as far as it individualize the social-historical traumatic experience of brain exodus by means of an autobiographical narrative which offers at the same time a realistic image of the so called “happiness” of émigré scholars, comprising their big efforts for achieving the highest performances in their field of research. Revealing the differences between the American and Romanian academic systems, which are condensed in the illustrative story of the hero-narrator, Avalon… incorporates, as Bildungsroman, a “cultural pedagogy” shaped especially under the guiding of one of the leading mathematicians on the world, Professor Bang Yen-Chen, but which is also grounded into the significant previous formative experience. In the same time, Avalon… offers a lesson of cosmopolitanism, which integrates, by means of literary memory, the cultural values of the post-communist East, especially those of Romanian recent pasts. Our analysis of the novel is focused on the “cultural-pedagogical” dimension of its multilayered architecture, in which the intertwined autobiographical, historical, and sociological levels are mobilized for converting the personal and collective traumas in a project of identity revision and rebuilding.

Designing hydrogen-free diamond like multilayer carbon coatings for superior mechanical and tribological performance

Diamond like carbon (DLC) coatings are extensively employed for their outstanding mechanical and tribological properties. To overcome the inherent high residual stress, therefore brittleness, DLC coatings with multilayer architecture were developed in the form of stacking alternate hard and soft layers to avoid premature failure under severe loading conditions. The current study was designed to investigate the impact of bilayer thickness (hard & soft) on wear of multilayer DLC particularly at high contact stress (2.0 GPa, 2.7 GPa, 3.0 GPa, 3.4 GPa). Seven DLC multilayer (1:1 bilayer ratio) samples with 1, 2, 5, 10, 20, 40, 80 bilayers were deposited on 440C steel and the overall coating thickness was mainatined at ∼1 µm. Moreover, the bilayer thickness effect was determined on mechanical, scratch adhesion and structural properties. Transmission electron microscope (TEM) was used to visualize discrete hard and soft layers in 80 bilayers (∼6 nm thin layer). G peak suppression and ID/IG increment were observed with reduction in bilayer thickness. Hardness, modulus, elastic strain to failure (H/E), plastic deformation resistance (H3/E2) and residual stresses showed an inverse relation with bilayer thickness. Furthermore, micro scratch adhesion decreases with layer thickness reduction. Only 100 nm and above bilayer thickness samples (1, 2, 5, 10 bilayers) could survive under high contact stress while the other three (20, 40, 80 bilayers) showed brittle failure using pin-on-disc tribometer. Additionally, 10 bilayers (100 nm thickness) produced the minimum wear (∼7.9 ×10−8 mm3/Nm) at 80 N.