Yafei Ning
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Yafei Ning has a total of 78 co-authors. Here are the top co-authors by number of shared publications:
- Hu Li
Shandong University
7 shared publications - Ziheng Li
Shandong Key Laboratory of Next-Generation Semiconductor Technology and Systems, School of Integrated Circuits, Shandong University, Jinan 250101, China
4 shared publications - Klaus Leifer
Ångström Laboratory, Department of Materials Science and Engineering, Uppsala University, 75121 Uppsala, Sweden
3 shared publications - Mingyang Wang
Institute of Fundamental and Frontier Sciences State Key Laboratory of Electronic Thin Films and Integrated Devices University of Electronic Science and Technology of China Chengdu China
3 shared publications - Syed Hassan Mujtaba Jafri
Department of Electrical Engineering, Mirpur University of Science and Technology (MUST), Mirpur 10250, Pakistan
3 shared publications - Xiaoqing Jiang
Shandong High Speed Information Group Co., Ltd., Jinan 250100, China
3 shared publications
Latest Publications
- Open Access
- 2 July 2026
- Published by MDPI AG in Electronics
- Vol. 15 (13) , 2906
A Novel High-Frequency Simulation Methodology for IBIS Models Utilizing Verilog-AMS Dynamic Parameter Compensation
Conventional I/O Buffer Information Specification (IBIS) models often suffer from reduced fidelity in high-speed signaling because their static table-lookup mechanism cannot accurately reproduce complex transient I/O-buffer dynamics. To address this limitation, this study proposes a Verilog-AMS-based dynamic parameter compensation method. First, the conventional IBIS model is reformulated into a three-layer architecture comprising a data interface layer, an intermediate variable computation layer, and a port response synthesis layer. Then, based on Kirchhoff’s current law (KCL), the monotonic dependence of the output voltage on the pull-up and pull-down driving factors, kpu and kpd, is analytically derived to provide a directional criterion for parameter correction. Building on this criterion, a pulse-width-driven compensation algorithm is developed by constructing a pulse-width-indexed dual-factor empirical adjustment matrix and detecting the pulse width of the input bitstream in real time during transient simulation. The detected pulse width is then used to dynamically update kpu and kpd, enabling the IBIS response to converge toward the transistor-level SPICE reference waveform. Three representative device models were evaluated at 666 Mbps and 1.302 Gbps using pseudo-random binary sequence excitation, and the model fidelity was quantified using the normalized mean square error (NMSE). The proposed method reduced the NMSE from −6.73 to −1.03 dB before compensation to −54.79 to −44.19 dB after compensation, demonstrating a substantial improvement in high-frequency IBIS modeling fidelity and confirming the robustness and adaptability of the pulse-width-aware dynamic compensation strategy under random high-speed excitation.
- 2 July 2026
- Published by MDPI AG in Electronics
- Vol. 15 (13) , 2906
Open AccessA Novel High-Frequency Simulation Methodology for IBIS Models Utilizing Verilog-AMS Dynamic Parameter Compensation
Conventional I/O Buffer Information Specification (IBIS) models often suffer from reduced fidelity in high-speed signaling because their static table-lookup mechanism cannot accurately reproduce complex transient I/O-buffer dynamics. To address this limitation, this study proposes a Verilog-AMS-based dynamic parameter compensation method. First, the conventional IBIS model is reformulated into a three-layer architecture comprising a data interface layer, an intermediate variable computation layer, and a port response synthesis layer. Then, based on Kirchhoff’s current law (KCL), the monotonic dependence of the output voltage on the pull-up and pull-down driving factors, kpu and kpd, is analytically derived to provide a directional criterion for parameter correction. Building on this criterion, a pulse-width-driven compensation algorithm is developed by constructing a pulse-width-indexed dual-factor empirical adjustment matrix and detecting the pulse width of the input bitstream in real time during transient simulation. The detected pulse width is then used to dynamically update kpu and kpd, enabling the IBIS response to converge toward the transistor-level SPICE reference waveform. Three representative device models were evaluated at 666 Mbps and 1.302 Gbps using pseudo-random binary sequence excitation, and the model fidelity was quantified using the normalized mean square error (NMSE). The proposed method reduced the NMSE from −6.73 to −1.03 dB before compensation to −54.79 to −44.19 dB after compensation, demonstrating a substantial improvement in high-frequency IBIS modeling fidelity and confirming the robustness and adaptability of the pulse-width-aware dynamic compensation strategy under random high-speed excitation.
- Open Access
- 19 May 2026
- Published by MDPI AG in Nanomaterials
- Vol. 16 (10) , 625
High-Performance Infrared Photodetectors Based on Graphene Nanoribbon Vertical Heterojunctions via Dissociated Double-Walled Carbon Nanotubes
Graphene nanoribbons (GNRs) inherit the exceptional carrier mobility of graphene while offering tunable bandgaps, making them promising for high-performance optoelectronics. Here, we report a high-performance near-infrared photodetector based on a p-GNR/Al2O3/n-Si vertical heterojunction, where GNR is directly produced by dissociating double-walled carbon nanotubes (DWCNTs). The 10 nm Al2O3 interlayer serves as an effective barrier and passivation layer, suppressing dark current and enhancing interfacial charge separation. Under 1064 nm illumination, the device delivers outstanding performance. At −6 V bias, the responsivity and detectivity reach 159.55 A/W and 2.01 × 1012 Jones, respectively. Notably, under zero-bias self-powered mode, it still achieves a high responsivity of 8.71 A/W, a detectivity of 1.15 × 1013 Jones, and a fast response time of 307.5 μs. These results fully validate the feasibility of GNR-based heterojunctions for high-performance optoelectronic devices and pave the way for their future integration into low-power, high-sensitivity photodetection systems and next-generation optoelectronic integrated circuits.
- 19 May 2026
- Published by MDPI AG in Nanomaterials
- Vol. 16 (10) , 625
Open AccessHigh-Performance Infrared Photodetectors Based on Graphene Nanoribbon Vertical Heterojunctions via Dissociated Double-Walled Carbon Nanotubes
Graphene nanoribbons (GNRs) inherit the exceptional carrier mobility of graphene while offering tunable bandgaps, making them promising for high-performance optoelectronics. Here, we report a high-performance near-infrared photodetector based on a p-GNR/Al2O3/n-Si vertical heterojunction, where GNR is directly produced by dissociating double-walled carbon nanotubes (DWCNTs). The 10 nm Al2O3 interlayer serves as an effective barrier and passivation layer, suppressing dark current and enhancing interfacial charge separation. Under 1064 nm illumination, the device delivers outstanding performance. At −6 V bias, the responsivity and detectivity reach 159.55 A/W and 2.01 × 1012 Jones, respectively. Notably, under zero-bias self-powered mode, it still achieves a high responsivity of 8.71 A/W, a detectivity of 1.15 × 1013 Jones, and a fast response time of 307.5 μs. These results fully validate the feasibility of GNR-based heterojunctions for high-performance optoelectronic devices and pave the way for their future integration into low-power, high-sensitivity photodetection systems and next-generation optoelectronic integrated circuits.
- Open Access
- 5 May 2026
- Published by MDPI AG in Electronics
- Vol. 15 (9) , 1954
Co-Design of Pipelining and Fixed-Point Quantization for SOVA-Turbo Codec IP Core
With the escalating demand for high reliability, low power consumption, and high throughput baseband processing capabilities in 5G and future 6G communication systems, the channel coding and decoding IP core serves as a critical module in baseband chips, where its hardware architecture and numerical system design exert a decisive influence on system performance. This paper proposes a SOVA-Turbo codec IP core architecture tailored for FPGA/SoC system integration. By reconstructing the Turbo iterative decoding data flow, a pipelined architecture is designed to facilitate the parallel operation of component decoders, as well as the storage and distribution of extrinsic information. Furthermore, addressing the numerical complexity of the SOVA algorithm in hardware implementation, a customized fixed-point representation and quantization co-design scheme is proposed to support hardware-efficient implementation while preserving the relative reliability relationships required for soft-decision decoding. Additionally, bit-level soft reliability information is generated at the decoder output to support iterative reliability evaluation under controlled noisy conditions. Simulation and implementation results demonstrate the feasibility of the proposed prototype in terms of timing closure, hardware cost, and decoder-side robustness under controlled perturbation.
- 5 May 2026
- Published by MDPI AG in Electronics
- Vol. 15 (9) , 1954
Open AccessCo-Design of Pipelining and Fixed-Point Quantization for SOVA-Turbo Codec IP Core
With the escalating demand for high reliability, low power consumption, and high throughput baseband processing capabilities in 5G and future 6G communication systems, the channel coding and decoding IP core serves as a critical module in baseband chips, where its hardware architecture and numerical system design exert a decisive influence on system performance. This paper proposes a SOVA-Turbo codec IP core architecture tailored for FPGA/SoC system integration. By reconstructing the Turbo iterative decoding data flow, a pipelined architecture is designed to facilitate the parallel operation of component decoders, as well as the storage and distribution of extrinsic information. Furthermore, addressing the numerical complexity of the SOVA algorithm in hardware implementation, a customized fixed-point representation and quantization co-design scheme is proposed to support hardware-efficient implementation while preserving the relative reliability relationships required for soft-decision decoding. Additionally, bit-level soft reliability information is generated at the decoder output to support iterative reliability evaluation under controlled noisy conditions. Simulation and implementation results demonstrate the feasibility of the proposed prototype in terms of timing closure, hardware cost, and decoder-side robustness under controlled perturbation.
- Open Access
Energy Dispersion Relationship and Hofstadter Butterfly of Triangle and Rectangular Moiré Patterns in Tight Binding States
Herein, the energy dispersion relationship and the density of states of triangular and rectangular moiré patterns are investigated using a tight binding model. Their characteristics of Hofstadter butterflies under different magnetic fields are also examined. The results indicate that, by analyzing different moiré superlattices, Hofstadter butterflies arising from different moiré pattern structures are obtained, exhibiting considerable fractal characteristics and self-similarities. Moreover, it is also observed that under an alternating magnetic field, the redistribution of electronic states leads to a significant change in the density of states curve, and the Van Hove peak changes with the increase in magnetic field intensity. This study enriches the understanding of the electronic behavior of moiré systems, but it also provides multiple potential application directions for future technological development.
Open AccessEnergy Dispersion Relationship and Hofstadter Butterfly of Triangle and Rectangular Moiré Patterns in Tight Binding States
Herein, the energy dispersion relationship and the density of states of triangular and rectangular moiré patterns are investigated using a tight binding model. Their characteristics of Hofstadter butterflies under different magnetic fields are also examined. The results indicate that, by analyzing different moiré superlattices, Hofstadter butterflies arising from different moiré pattern structures are obtained, exhibiting considerable fractal characteristics and self-similarities. Moreover, it is also observed that under an alternating magnetic field, the redistribution of electronic states leads to a significant change in the density of states curve, and the Van Hove peak changes with the increase in magnetic field intensity. This study enriches the understanding of the electronic behavior of moiré systems, but it also provides multiple potential application directions for future technological development.
- Open Access
- 10 May 2025
- Published by MDPI AG in Nanomaterials
- Vol. 15 (10) , 720
Doping Characteristics and Band Engineering of InSe for Advanced Photodetectors: A DFT Study
Two-dimensional materials have emerged as core components for next-generation optoelectronic devices due to their quantum confinement effects and tunable electronic properties. Indium selenide (InSe) demonstrates breakthrough photoelectric performance, with its remarkable light-responsive characteristics spanning from visible to near-infrared regions, offering application potential in high-speed imaging, optical communication, and biosensing. This study investigates the doping characteristics of InSe using first-principles calculations, focusing on the doping and adsorption behaviors of Argentum (Ag) and Bismuth (Bi) atoms in InSe and their effects on its electronic structure. The research reveals that Ag atoms preferentially adsorb at interlayer vacancies with a binding energy of −2.19 eV, forming polar covalent bonds. This reduces the band gap from the intrinsic 1.51 eV to 0.29–1.16 eV and induces an indirect-to-direct band gap transition. Bi atoms doped at the center of three Se atoms exhibit a binding energy of −2.06 eV, narrowing the band gap to 0.19 eV through strong ionic bonding, while inducing metallic transition at inter-In sites. The introduced intermediate energy levels significantly reduce electron transition barriers (by up to 60%) and enhance carrier separation efficiency. This study links doping sites, electronic structures, and photoelectric properties through computational simulations, offering a theoretical framework for designing high-performance InSe-based photodetectors. It opens new avenues for narrow-bandgap near-infrared detection and carrier transport optimization.
- 10 May 2025
- Published by MDPI AG in Nanomaterials
- Vol. 15 (10) , 720
Open AccessDoping Characteristics and Band Engineering of InSe for Advanced Photodetectors: A DFT Study
Two-dimensional materials have emerged as core components for next-generation optoelectronic devices due to their quantum confinement effects and tunable electronic properties. Indium selenide (InSe) demonstrates breakthrough photoelectric performance, with its remarkable light-responsive characteristics spanning from visible to near-infrared regions, offering application potential in high-speed imaging, optical communication, and biosensing. This study investigates the doping characteristics of InSe using first-principles calculations, focusing on the doping and adsorption behaviors of Argentum (Ag) and Bismuth (Bi) atoms in InSe and their effects on its electronic structure. The research reveals that Ag atoms preferentially adsorb at interlayer vacancies with a binding energy of −2.19 eV, forming polar covalent bonds. This reduces the band gap from the intrinsic 1.51 eV to 0.29–1.16 eV and induces an indirect-to-direct band gap transition. Bi atoms doped at the center of three Se atoms exhibit a binding energy of −2.06 eV, narrowing the band gap to 0.19 eV through strong ionic bonding, while inducing metallic transition at inter-In sites. The introduced intermediate energy levels significantly reduce electron transition barriers (by up to 60%) and enhance carrier separation efficiency. This study links doping sites, electronic structures, and photoelectric properties through computational simulations, offering a theoretical framework for designing high-performance InSe-based photodetectors. It opens new avenues for narrow-bandgap near-infrared detection and carrier transport optimization.
- 7 April 2025
- Published by AIP Publishing in Applied Physics Reviews
- Vol. 12 (2)
Superior self-powered infrared photodetector via semiconducting graphene-nanoribbons-based vertical heterojunctions
Self-powered photodetectors, operating without an external power source, have garnered extensive interest for infrared (IR) detection owing to their vast potential in low-power consumption sensor systems. Here, a short-wavelength infrared (SWIR) heterojunction photodetector utilizing semiconducting graphene nanoribbons has been achieved and demonstrated record-high performance. In the self-powered mode, the heterojunction photodetector presents a responsivity of 73.5 A/W and a detectivity of 7.5 × 1014 Jones, surpassing previously reported self-powered IR photodetectors by several orders of magnitude. The superior performance is primarily due to the enhancement of the electric field caused by the photogating effect at the heterointerface. The device also displays unparalleled performance at −5 V bias voltage, achieving a responsivity of 843 A/W, a detectivity of 1015 Jones, and an external quantum efficiency of 105%, all of which are record-breaking values for SWIR photodetectors to date. Therefore, our approach provides new insight and demonstrates great potential for future high-performance low-power consumption IR detection technology.
- 7 April 2025
- Published by AIP Publishing in Applied Physics Reviews
- Vol. 12 (2)
Superior self-powered infrared photodetector via semiconducting graphene-nanoribbons-based vertical heterojunctions
Self-powered photodetectors, operating without an external power source, have garnered extensive interest for infrared (IR) detection owing to their vast potential in low-power consumption sensor systems. Here, a short-wavelength infrared (SWIR) heterojunction photodetector utilizing semiconducting graphene nanoribbons has been achieved and demonstrated record-high performance. In the self-powered mode, the heterojunction photodetector presents a responsivity of 73.5 A/W and a detectivity of 7.5 × 1014 Jones, surpassing previously reported self-powered IR photodetectors by several orders of magnitude. The superior performance is primarily due to the enhancement of the electric field caused by the photogating effect at the heterointerface. The device also displays unparalleled performance at −5 V bias voltage, achieving a responsivity of 843 A/W, a detectivity of 1015 Jones, and an external quantum efficiency of 105%, all of which are record-breaking values for SWIR photodetectors to date. Therefore, our approach provides new insight and demonstrates great potential for future high-performance low-power consumption IR detection technology.
- Open Access
- 4 July 2024
- Published by MDPI AG in Electronics
- Vol. 13 (13) , 2637
Improved Implementation of Chua’s Circuit on an Active Inductor and Non-Autonomous System
Chua’s circuit is a well-established model for studying chaotic phenomena and is extensively implemented in fields like encrypted communication. However, a traditional Chua’s circuit has large volume, high component precision requirements and limited adjustable parameter range, which are not conducive to application. In order to solve these problems, we propose an improved implementation of Chua’s circuit on an active inductor and non-autonomous system. First, we adopt the strategy of using active inductors instead of traditional passive inductors, achieving the miniaturization of the circuit and improving the accuracy of inductance. In addition, we present the theory of substituting non-autonomous systems for classical autonomous systems to reduce the requirements for the accuracy of components and improve the robustness of the circuit. Lastly, we connect the extension resistor in parallel with Chua’s diode to optimize circuit structure, thereby increasing the range of the adjustable parameter. Based on the three improvements above, experiments have shown that the average maximum error tolerance of components of our improved design has been increased from 1.88% to 7.38% when generating a single vortex, and from 4.73% to 12.61% when generating a double vortex, compared with the traditional Chua’s circuit. The range of the adjustable parameter has been increased by 195.83% and 36.98%, respectively, when generating a single vortex and double vortex. In summary, our improved circuit is more practical than the traditional Chua’s circuit and has good application value.
- 4 July 2024
- Published by MDPI AG in Electronics
- Vol. 13 (13) , 2637
Open AccessImproved Implementation of Chua’s Circuit on an Active Inductor and Non-Autonomous System
Chua’s circuit is a well-established model for studying chaotic phenomena and is extensively implemented in fields like encrypted communication. However, a traditional Chua’s circuit has large volume, high component precision requirements and limited adjustable parameter range, which are not conducive to application. In order to solve these problems, we propose an improved implementation of Chua’s circuit on an active inductor and non-autonomous system. First, we adopt the strategy of using active inductors instead of traditional passive inductors, achieving the miniaturization of the circuit and improving the accuracy of inductance. In addition, we present the theory of substituting non-autonomous systems for classical autonomous systems to reduce the requirements for the accuracy of components and improve the robustness of the circuit. Lastly, we connect the extension resistor in parallel with Chua’s diode to optimize circuit structure, thereby increasing the range of the adjustable parameter. Based on the three improvements above, experiments have shown that the average maximum error tolerance of components of our improved design has been increased from 1.88% to 7.38% when generating a single vortex, and from 4.73% to 12.61% when generating a double vortex, compared with the traditional Chua’s circuit. The range of the adjustable parameter has been increased by 195.83% and 36.98%, respectively, when generating a single vortex and double vortex. In summary, our improved circuit is more practical than the traditional Chua’s circuit and has good application value.
- Open Access
- 26 March 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Photonics Journal
- Vol. 16 (3) , 1-9
High Performance Self-Powered Photodetectors Based on Graphene Nanoribbons/Al2O3/InGaZnO Heterojunctions
Self-powered photodetectors which operate without external power sources hold immense promise in future photodetection systems. To achieve high-performance self-powered optoelectronic devices, efficient electron-hole pair separation is critical to generate high photocurrents. In this work, we successfully synthesized semiconducting graphene nanoribbons (GNRs) with a direct bandgap of 1.80 eV and employed them to construct a high-performance GNR/Al 2 O 3 /IGZO heterostructure photodetector. The built-in electric field in the heterojunctions enables this photodetector to exhibit remarkable performance, showing a responsivity of up to 68 mA/W, a detectivity of 8.34 × 10 10 Jones, and rapid response times of 21/20 ms at zero bias. Furthermore, the photodetector features a wide spectral detection range of 405 to 1550 nm. These results highlight the promising potential of GNR/IGZO p-n heterojunction-based self-powered photodetectors in optoelectronic applications.
- 26 March 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Photonics Journal
- Vol. 16 (3) , 1-9
Open AccessHigh Performance Self-Powered Photodetectors Based on Graphene Nanoribbons/Al2O3/InGaZnO Heterojunctions
Self-powered photodetectors which operate without external power sources hold immense promise in future photodetection systems. To achieve high-performance self-powered optoelectronic devices, efficient electron-hole pair separation is critical to generate high photocurrents. In this work, we successfully synthesized semiconducting graphene nanoribbons (GNRs) with a direct bandgap of 1.80 eV and employed them to construct a high-performance GNR/Al 2 O 3 /IGZO heterostructure photodetector. The built-in electric field in the heterojunctions enables this photodetector to exhibit remarkable performance, showing a responsivity of up to 68 mA/W, a detectivity of 8.34 × 10 10 Jones, and rapid response times of 21/20 ms at zero bias. Furthermore, the photodetector features a wide spectral detection range of 405 to 1550 nm. These results highlight the promising potential of GNR/IGZO p-n heterojunction-based self-powered photodetectors in optoelectronic applications.
- Open Access
Advances in the Field of Graphene-Based Composites for Energy–Storage Applications
To meet the growing demand in energy, great efforts have been devoted to improving the performances of energy–storages. Graphene, a remarkable two-dimensional (2D) material, holds immense potential for improving energy–storage performance owing to its exceptional properties, such as a large-specific surface area, remarkable thermal conductivity, excellent mechanical strength, and high-electronic mobility. This review provides a comprehensive summary of recent research advancements in the application of graphene for energy–storage. Initially, the fundamental properties of graphene are introduced. Subsequently, the latest developments in graphene-based energy–storage, encompassing lithium-ion batteries, sodium-ion batteries, supercapacitors, potassium-ion batteries and aluminum-ion batteries, are summarized. Finally, the challenges associated with graphene-based energy–storage applications are discussed, and the development prospects for this field are outlined.
Open AccessAdvances in the Field of Graphene-Based Composites for Energy–Storage Applications
To meet the growing demand in energy, great efforts have been devoted to improving the performances of energy–storages. Graphene, a remarkable two-dimensional (2D) material, holds immense potential for improving energy–storage performance owing to its exceptional properties, such as a large-specific surface area, remarkable thermal conductivity, excellent mechanical strength, and high-electronic mobility. This review provides a comprehensive summary of recent research advancements in the application of graphene for energy–storage. Initially, the fundamental properties of graphene are introduced. Subsequently, the latest developments in graphene-based energy–storage, encompassing lithium-ion batteries, sodium-ion batteries, supercapacitors, potassium-ion batteries and aluminum-ion batteries, are summarized. Finally, the challenges associated with graphene-based energy–storage applications are discussed, and the development prospects for this field are outlined.
- Open Access
- 7 May 2022
- Published by Wiley in Disease Markers
- Vol. 2022, 1-14
Machine Learning-Based Radiomics for Prediction of Epidermal Growth Factor Receptor Mutations in Lung Adenocarcinoma
Identifying an epidermal growth factor receptor (EGFR) mutation is important because EGFR tyrosine kinase inhibitors are the first-line treatment of choice for patients with EGFR mutation-positive lung adenocarcinomas (LUAC). This study is aimed at developing and validating a radiomics-based machine learning (ML) approach to identify EGFR mutations in patients with LUAC. We retrospectively collected data from 201 patients with positive EGFR mutation LUAC (140 in the training cohort and 61 in the validation cohort). We extracted 1316 radiomics features from preprocessed CT images and selected 14 radiomics features and 1 clinical feature which were most relevant to mutations through filter method. Subsequently, we built models using 7 ML approaches and established the receiver operating characteristic (ROC) curve to assess the discriminating performance of these models. In terms of predicting EGFR mutation, the model derived from radiomics features and combined models (radiomics features and relevant clinical factors) had an AUC of 0.79 (95% confidence interval (CI): 0.77-0.82), 0.86 (0.87-0.88), respectively. Our study offers a radiomics-based ML model using filter methods to detect the EGFR mutation in patients with LUAC. This convenient and low-cost method may be of help to noninvasively identify patients before obtaining tumor sample for molecule testing.
- 7 May 2022
- Published by Wiley in Disease Markers
- Vol. 2022, 1-14
Open AccessMachine Learning-Based Radiomics for Prediction of Epidermal Growth Factor Receptor Mutations in Lung Adenocarcinoma
Identifying an epidermal growth factor receptor (EGFR) mutation is important because EGFR tyrosine kinase inhibitors are the first-line treatment of choice for patients with EGFR mutation-positive lung adenocarcinomas (LUAC). This study is aimed at developing and validating a radiomics-based machine learning (ML) approach to identify EGFR mutations in patients with LUAC. We retrospectively collected data from 201 patients with positive EGFR mutation LUAC (140 in the training cohort and 61 in the validation cohort). We extracted 1316 radiomics features from preprocessed CT images and selected 14 radiomics features and 1 clinical feature which were most relevant to mutations through filter method. Subsequently, we built models using 7 ML approaches and established the receiver operating characteristic (ROC) curve to assess the discriminating performance of these models. In terms of predicting EGFR mutation, the model derived from radiomics features and combined models (radiomics features and relevant clinical factors) had an AUC of 0.79 (95% confidence interval (CI): 0.77-0.82), 0.86 (0.87-0.88), respectively. Our study offers a radiomics-based ML model using filter methods to detect the EGFR mutation in patients with LUAC. This convenient and low-cost method may be of help to noninvasively identify patients before obtaining tumor sample for molecule testing.