Shuai Fang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Shuai Fang has a total of 20 co-authors. Here are the top co-authors by number of shared publications:
- Xuezhi Yang
Hefei University of Technology
2 shared publications ChangWen Chen
1 shared publication- Chengwei Zhou
Department of Thoracic Surgery, The First Affiliated Hospital of Ningbo University, Ningbo, China
1 shared publication Huihui Bai
Department of Biochemistry and Molecular Biology and Zhejiang Key Laboratory of Pathophysiology, School of Basic Medical Sciences, Health Science Center, Ningbo University, Ningbo, China
1 shared publication- Jie Zhang
First Affiliated Hospital of the University of Science and Technology of China, Hefei, China
1 shared publication - Jing Zhang
Beihang University
1 shared publication
Latest Publications
- Open Access
- 10 June 2026
- Published by MDPI AG in Mathematical and Computational Applications
- Vol. 31 (3) , 102
Bridging the Semantic Gap in Industry–Academia Collaboration: A Two-Stage RAG System for Intelligent Expert Recommendation
Aligning industrial technological demands with academic expertise is critical for effective technology transfer. However, existing Expert Recommendation Systems (ERS) are frequently hindered by a “semantic gap” arising from terminological discrepancies between industry and academia, alongside a reliance on rigid classification taxonomies. To address these limitations, this paper proposes an automated expert finding framework that integrates Large Language Models (LLMs) with a hierarchical Retrieval-Augmented Generation (RAG) mechanism. Initially, we employ LLMs for the unsupervised extraction of research domains and technical keywords from heterogeneous multi-source data. To mitigate terminological diversity, we introduce a vector clustering-based Semantic Normalization module. By mapping diverse keyword variants into unified “Concept Clusters,” this module reduces vocabulary sparsity by 98%. These organized clusters are structured into a “Semantic Tree” to support a hierarchical RAG strategy, enabling a coarse-to-fine retrieval process from broad disciplinary domains down to specific technical achievements. In this paper, RAG refers to a retrieval-augmented expert recommendation workflow, in which the retrieved achievements and expert evidence are used as grounded context for generating an explanatory recommendation report. Evaluations on a real-world dataset show that the framework achieves a Precision@5 of 78.4% and a Recall@10 of 81.2%, outperforming flat vector retrieval baselines by over 20% in precision. Furthermore, hierarchical domain pruning significantly reduces computational overhead, decreasing average query latency by a factor of three (to 115.8 ms). These results demonstrate that the proposed system effectively bridges the industry–academia semantic gap, providing a scalable and accurate solution for expert recommendation.
- 10 June 2026
- Published by MDPI AG in Mathematical and Computational Applications
- Vol. 31 (3) , 102
Open AccessBridging the Semantic Gap in Industry–Academia Collaboration: A Two-Stage RAG System for Intelligent Expert Recommendation
Aligning industrial technological demands with academic expertise is critical for effective technology transfer. However, existing Expert Recommendation Systems (ERS) are frequently hindered by a “semantic gap” arising from terminological discrepancies between industry and academia, alongside a reliance on rigid classification taxonomies. To address these limitations, this paper proposes an automated expert finding framework that integrates Large Language Models (LLMs) with a hierarchical Retrieval-Augmented Generation (RAG) mechanism. Initially, we employ LLMs for the unsupervised extraction of research domains and technical keywords from heterogeneous multi-source data. To mitigate terminological diversity, we introduce a vector clustering-based Semantic Normalization module. By mapping diverse keyword variants into unified “Concept Clusters,” this module reduces vocabulary sparsity by 98%. These organized clusters are structured into a “Semantic Tree” to support a hierarchical RAG strategy, enabling a coarse-to-fine retrieval process from broad disciplinary domains down to specific technical achievements. In this paper, RAG refers to a retrieval-augmented expert recommendation workflow, in which the retrieved achievements and expert evidence are used as grounded context for generating an explanatory recommendation report. Evaluations on a real-world dataset show that the framework achieves a Precision@5 of 78.4% and a Recall@10 of 81.2%, outperforming flat vector retrieval baselines by over 20% in precision. Furthermore, hierarchical domain pruning significantly reduces computational overhead, decreasing average query latency by a factor of three (to 115.8 ms). These results demonstrate that the proposed system effectively bridges the industry–academia semantic gap, providing a scalable and accurate solution for expert recommendation.
- Open Access
- 28 September 2024
- Published by Springer Nature in Molecular Cancer
- Vol. 23 (1) , 1-14
CircLIFRSA/miR-1305/PTEN axis attenuates malignant cellular processes in non-small cell lung cancer by regulating AKT phosphorylation
Non-small cell lung cancer (NSCLC) is typically diagnosed at advanced stages, which limits the effectiveness of therapeutic interventions. The present study aimed to explore the role of the newly identified circLIFRSA in the PTEN/AKT signaling pathway and its involvement in the malignant processes of NSCLC. CircLIFRSA expression was identified through microarray analysis, and its levels in NSCLC samples were quantified by RT-qPCR. The impact of circLIFRSA on cell growth, proliferation, apoptosis, and cell cycle were evaluated by MTT assay, colony formation assay, and flow cytometry. Additionally, Western blotting was employed to analyze the expression of PTEN and phosphorylated AKT (pAKT) in NSCLC cells. The expression of circLIFRSA was found to be significantly reduced in NSCLC cells and tissues. This downregulation correlated with various clinicopathological characteristics and indicated its potential as an early diagnostic biomarker for NSCLC. Importantly, circLIFRSA was shown to inhibit cell growth and proliferation while promoting apoptosis in NSCLC cells. Mechanically, circLIFRSA was found to attenuate the malignant processes of NSCLC cells via the miR-1305/PTEN axis and the suppression of AKT phosphorylation. These findings indicate that circLIFRSA/miR-1305/PTEN axis attenuates malignant processes by regulating AKT phosphorylation, and provide new insights into the potential of circLIFRSA as a biomarker for early diagnosis and as a promising therapeutic target in NSCLC.
- 28 September 2024
- Published by Springer Nature in Molecular Cancer
- Vol. 23 (1) , 1-14
Open AccessCircLIFRSA/miR-1305/PTEN axis attenuates malignant cellular processes in non-small cell lung cancer by regulating AKT phosphorylation
Non-small cell lung cancer (NSCLC) is typically diagnosed at advanced stages, which limits the effectiveness of therapeutic interventions. The present study aimed to explore the role of the newly identified circLIFRSA in the PTEN/AKT signaling pathway and its involvement in the malignant processes of NSCLC. CircLIFRSA expression was identified through microarray analysis, and its levels in NSCLC samples were quantified by RT-qPCR. The impact of circLIFRSA on cell growth, proliferation, apoptosis, and cell cycle were evaluated by MTT assay, colony formation assay, and flow cytometry. Additionally, Western blotting was employed to analyze the expression of PTEN and phosphorylated AKT (pAKT) in NSCLC cells. The expression of circLIFRSA was found to be significantly reduced in NSCLC cells and tissues. This downregulation correlated with various clinicopathological characteristics and indicated its potential as an early diagnostic biomarker for NSCLC. Importantly, circLIFRSA was shown to inhibit cell growth and proliferation while promoting apoptosis in NSCLC cells. Mechanically, circLIFRSA was found to attenuate the malignant processes of NSCLC cells via the miR-1305/PTEN axis and the suppression of AKT phosphorylation. These findings indicate that circLIFRSA/miR-1305/PTEN axis attenuates malignant processes by regulating AKT phosphorylation, and provide new insights into the potential of circLIFRSA as a biomarker for early diagnosis and as a promising therapeutic target in NSCLC.
- Open Access
- 12 October 2023
- Published by MDPI AG in Remote Sensing
- Vol. 15 (20) , 4936
Spatiotemporal Fusion Model of Remote Sensing Images Combining Single-Band and Multi-Band Prediction
In recent years, convolutional neural network (CNN)-based spatiotemporal fusion (STF) models for remote sensing images have made significant progress. However, existing STF models may suffer from two main drawbacks. Firstly, multi-band prediction often generates a hybrid feature representation that includes information from all bands. This blending of features can lead to the loss or blurring of high-frequency details, making it challenging to reconstruct multi-spectral remote sensing images with significant spectral differences between bands. Another challenge in many STF models is the limited preservation of spectral information during 2D convolution operations. Combining all input channels’ convolution results into a single-channel output feature map can lead to the degradation of spectral dimension information. To address these issues and to strike a balance between avoiding hybrid features and fully utilizing spectral information, we propose a remote sensing image STF model that combines single-band and multi-band prediction (SMSTFM). The SMSTFM initially performs single-band prediction, generating separate predicted images for each band, which are then stacked together to form a preliminary fused image. Subsequently, the multi-band prediction module leverages the spectral dimension information of the input images to further enhance the preliminary predictions. We employ the modern ConvNeXt convolutional module as the primary feature extraction component. During the multi-band prediction phase, we enhance the spatial and channel information captures by replacing the 2D convolutions within ConvNeXt with 3D convolutions. In the experimental section, we evaluate our proposed algorithm on two public datasets with 16x resolution differences and one dataset with a 3x resolution difference. The results demonstrate that our SMSTFM achieves state-of-the-art performance on these datasets and is proven effective and reasonable through ablation studies.
- 12 October 2023
- Published by MDPI AG in Remote Sensing
- Vol. 15 (20) , 4936
Open AccessSpatiotemporal Fusion Model of Remote Sensing Images Combining Single-Band and Multi-Band Prediction
In recent years, convolutional neural network (CNN)-based spatiotemporal fusion (STF) models for remote sensing images have made significant progress. However, existing STF models may suffer from two main drawbacks. Firstly, multi-band prediction often generates a hybrid feature representation that includes information from all bands. This blending of features can lead to the loss or blurring of high-frequency details, making it challenging to reconstruct multi-spectral remote sensing images with significant spectral differences between bands. Another challenge in many STF models is the limited preservation of spectral information during 2D convolution operations. Combining all input channels’ convolution results into a single-channel output feature map can lead to the degradation of spectral dimension information. To address these issues and to strike a balance between avoiding hybrid features and fully utilizing spectral information, we propose a remote sensing image STF model that combines single-band and multi-band prediction (SMSTFM). The SMSTFM initially performs single-band prediction, generating separate predicted images for each band, which are then stacked together to form a preliminary fused image. Subsequently, the multi-band prediction module leverages the spectral dimension information of the input images to further enhance the preliminary predictions. We employ the modern ConvNeXt convolutional module as the primary feature extraction component. During the multi-band prediction phase, we enhance the spatial and channel information captures by replacing the 2D convolutions within ConvNeXt with 3D convolutions. In the experimental section, we evaluate our proposed algorithm on two public datasets with 16x resolution differences and one dataset with a 3x resolution difference. The results demonstrate that our SMSTFM achieves state-of-the-art performance on these datasets and is proven effective and reasonable through ablation studies.
- 19 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 72 (00189456) , 1-13
Noncontact Blood Pressure Estimation Using BP-Related Cardiovascular Knowledge: An Uncalibrated Method Based on Consumer-Level Camera
Objective : The tiny change of skin color, caused by heartbeat, can be captured with consumer-level cameras by using the imaging photoplethysmography (iPPG) technique, offering a non-contact way of extracting pulse signals. The pulse signals have been demonstrated to contain information of human physiological characteristics and have been used for blood pressure (BP) estimation in recent years. According to BP-related cardiovascular knowledge, this paper presents a new method for BP estimation based on the iPPG pulse signals, featured by incorporating cardiovascular characteristics including heart rate (HR), stroke volume (SV), elasticity of vessel walls (EVW), and peripheral vascular resistance (PVR). Correlations between the systolic blood pressure (SBP), diastolic blood pressure (DBP), pulse pressure (PP), and the cardiovascular characteristics are extracted, which facilitates the selection of pulse features consistent with BP properties. Based on the selected features, two Bayesian neural network (BNN) models are constructed for the estimation of SBP and DBP respectively, where the machine learning (ML) uncertainty of the estimation is also evaluated. This method is uncalibrated that means it can work without additional information except for the videos from the camera. The proposed method has been tested on 220 patients with a history of cardiovascular diseases. Errors of the BP estimation are 9±13 (MAE±STD) mmHg for SBP, 7±10 (MAE±STD) mmHg for DBP, and the ML uncertainty of the estimation indicates reliability of the proposed method.
- 19 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 72 (00189456) , 1-13
Noncontact Blood Pressure Estimation Using BP-Related Cardiovascular Knowledge: An Uncalibrated Method Based on Consumer-Level Camera
Objective : The tiny change of skin color, caused by heartbeat, can be captured with consumer-level cameras by using the imaging photoplethysmography (iPPG) technique, offering a non-contact way of extracting pulse signals. The pulse signals have been demonstrated to contain information of human physiological characteristics and have been used for blood pressure (BP) estimation in recent years. According to BP-related cardiovascular knowledge, this paper presents a new method for BP estimation based on the iPPG pulse signals, featured by incorporating cardiovascular characteristics including heart rate (HR), stroke volume (SV), elasticity of vessel walls (EVW), and peripheral vascular resistance (PVR). Correlations between the systolic blood pressure (SBP), diastolic blood pressure (DBP), pulse pressure (PP), and the cardiovascular characteristics are extracted, which facilitates the selection of pulse features consistent with BP properties. Based on the selected features, two Bayesian neural network (BNN) models are constructed for the estimation of SBP and DBP respectively, where the machine learning (ML) uncertainty of the estimation is also evaluated. This method is uncalibrated that means it can work without additional information except for the videos from the camera. The proposed method has been tested on 220 patients with a history of cardiovascular diseases. Errors of the BP estimation are 9±13 (MAE±STD) mmHg for SBP, 7±10 (MAE±STD) mmHg for DBP, and the ML uncertainty of the estimation indicates reliability of the proposed method.
- Open Access
- 20 November 2022
- Published by MDPI AG in Remote Sensing
- Vol. 14 (22) , 5883
WDBSTF: A Weighted Dual-Branch Spatiotemporal Fusion Network Based on Complementarity between Super-Resolution and Change Prediction
Spatiotemporal fusion (STF) is a solution to generate satellite images with both high-spatial and high-temporal resolutions. The deep learning-based STF algorithms focus on spatial dimensions to build a super-resolution (SR) model or the temporal dimensions to build a change prediction (CP) model, or the task itself to build a data-driven end-to-end model. The multi-source images used for STF usually have large spatial scale gaps and temporal spans. The large spatial scale gaps lead to poor spatial details based on a SR model; the large temporal spans make it difficult to accurately reconstruct changing areas based on a CP model. We propose a weighted dual-branch spatiotemporal fusion network based on complementarity between super-resolution and change prediction (WDBSTF), which includes the SR branch and CP branch, and a weight module representing the complementarity of the two branches. The SR branch makes full use of edge information and high-resolution reference images to obtain high-quality spatial features for image reconstruction. The CP branch decomposes complex problems via a two-layer cascaded network, changes features from the difference image, and selects high-quality spatial features through the attention mechanism. The fusion result of the CP branch has rich image details, but the fusion accuracy in the changing area is low due to the lack of detail. The SR branch has consistent and excellent fusion performances in the changing and no-changing areas, but the image details are not rich enough compared with the CP branch due to the large amplification factor. Next, a weighted network was designed to combine the advantages of the two branches to produce improved fusion results. We evaluated the performance of the WDBSTF in three representative scenarios, and both visual and quantitative evaluations demonstrate the state-of-the-art performance of our algorithm. (On the LGC dataset, our method outperforms the suboptimal method by 2.577% on SSIM. On the AHB dataset, our method outperforms the suboptimal method by 1.684% on SSIM. On the CIA dataset, our method outperforms the suboptimal method by 5.55% on SAM).
- 20 November 2022
- Published by MDPI AG in Remote Sensing
- Vol. 14 (22) , 5883
Open AccessWDBSTF: A Weighted Dual-Branch Spatiotemporal Fusion Network Based on Complementarity between Super-Resolution and Change Prediction
Spatiotemporal fusion (STF) is a solution to generate satellite images with both high-spatial and high-temporal resolutions. The deep learning-based STF algorithms focus on spatial dimensions to build a super-resolution (SR) model or the temporal dimensions to build a change prediction (CP) model, or the task itself to build a data-driven end-to-end model. The multi-source images used for STF usually have large spatial scale gaps and temporal spans. The large spatial scale gaps lead to poor spatial details based on a SR model; the large temporal spans make it difficult to accurately reconstruct changing areas based on a CP model. We propose a weighted dual-branch spatiotemporal fusion network based on complementarity between super-resolution and change prediction (WDBSTF), which includes the SR branch and CP branch, and a weight module representing the complementarity of the two branches. The SR branch makes full use of edge information and high-resolution reference images to obtain high-quality spatial features for image reconstruction. The CP branch decomposes complex problems via a two-layer cascaded network, changes features from the difference image, and selects high-quality spatial features through the attention mechanism. The fusion result of the CP branch has rich image details, but the fusion accuracy in the changing area is low due to the lack of detail. The SR branch has consistent and excellent fusion performances in the changing and no-changing areas, but the image details are not rich enough compared with the CP branch due to the large amplification factor. Next, a weighted network was designed to combine the advantages of the two branches to produce improved fusion results. We evaluated the performance of the WDBSTF in three representative scenarios, and both visual and quantitative evaluations demonstrate the state-of-the-art performance of our algorithm. (On the LGC dataset, our method outperforms the suboptimal method by 2.577% on SSIM. On the AHB dataset, our method outperforms the suboptimal method by 1.684% on SSIM. On the CIA dataset, our method outperforms the suboptimal method by 5.55% on SAM).
- 5 December 2021
- Published by Springer Nature
- p. 224-234
Multi-label Classification of Hyperspectral Images Based on Label-Specific Feature Fusion
- 5 December 2021
- Published by Springer Nature
- p. 224-234
Multi-label Classification of Hyperspectral Images Based on Label-Specific Feature Fusion
- 5 August 2014
- Published by Optica Publishing Group in Optics Express
- Vol. 22 (16) , 19523-19537
Image dehazing using polarization effects of objects and airlight
Optics InfoBase is the Optical Society's online library for flagship journals, partnered and copublished journals, and recent proceedings from OSA conferences.
- 5 August 2014
- Published by Optica Publishing Group in Optics Express
- Vol. 22 (16) , 19523-19537
Image dehazing using polarization effects of objects and airlight
Optics InfoBase is the Optical Society's online library for flagship journals, partnered and copublished journals, and recent proceedings from OSA conferences.