Haijun Gao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Haijun Gao has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- Jiangtao Su
School of Electronic Information, Hangzhou Dianzi University, Hangzhou 310018, China
2 shared publications - Lingling Sun
Hangzhou Dianzi University
2 shared publications - Xiwei Huang
Hangzhou Dianzi University
2 shared publications - Dirong Chen
Key Laboratory of RF Circuits and systems, Ministry of Education, Hangzhou Dianzi University, Hangzhou 310018, China
1 shared publication - Guodong Su
Zhejiang Provincial Laboratory of Integrated Circuit Design, Hangzhou Dianzi University, Hangzhou 310018, China
1 shared publication - Jialin Cai
Hangzhou Dianzi University
1 shared publication
Latest Publications
- Open Access
UR-Net: An Integrated ResUNet and Attention Based Image Enhancement and Classification Network for Stain-Free White Blood Cells
The differential count of white blood cells (WBCs) can effectively provide disease information for patients. Existing stained microscopic WBC classification usually requires complex sample-preparation steps, and is easily affected by external conditions such as illumination. In contrast, the inconspicuous nuclei of stain-free WBCs also bring great challenges to WBC classification. As such, image enhancement, as one of the preprocessing methods of image classification, is essential in improving the image qualities of stain-free WBCs. However, traditional or existing convolutional neural network (CNN)-based image enhancement techniques are typically designed as standalone modules aimed at improving the perceptual quality of humans, without considering their impact on advanced computer vision tasks of classification. Therefore, this work proposes a novel model, UR-Net, which consists of an image enhancement network framed by ResUNet with an attention mechanism and a ResNet classification network. The enhancement model is integrated into the classification model for joint training to improve the classification performance for stain-free WBCs. The experimental results demonstrate that compared to the models without image enhancement and previous enhancement and classification models, our proposed model achieved a best classification performance of 83.34% on our stain-free WBC dataset.
Open AccessUR-Net: An Integrated ResUNet and Attention Based Image Enhancement and Classification Network for Stain-Free White Blood Cells
The differential count of white blood cells (WBCs) can effectively provide disease information for patients. Existing stained microscopic WBC classification usually requires complex sample-preparation steps, and is easily affected by external conditions such as illumination. In contrast, the inconspicuous nuclei of stain-free WBCs also bring great challenges to WBC classification. As such, image enhancement, as one of the preprocessing methods of image classification, is essential in improving the image qualities of stain-free WBCs. However, traditional or existing convolutional neural network (CNN)-based image enhancement techniques are typically designed as standalone modules aimed at improving the perceptual quality of humans, without considering their impact on advanced computer vision tasks of classification. Therefore, this work proposes a novel model, UR-Net, which consists of an image enhancement network framed by ResUNet with an attention mechanism and a ResNet classification network. The enhancement model is integrated into the classification model for joint training to improve the classification performance for stain-free WBCs. The experimental results demonstrate that compared to the models without image enhancement and previous enhancement and classification models, our proposed model achieved a best classification performance of 83.34% on our stain-free WBC dataset.
- 17 February 2022
- Published by Wiley in International Journal of Numerical Modelling: Electronic Networks, Devices and Fields
- Vol. 35 (4)
Analysis and modeling of the kink effect in S22 based on support vector machine for GaN HEMTs
In this work, the kink effect (KE), typically visible in S22, is analyzed and modeled. Two different modeling techniques: equivalent circuit modeling (ECM) method and machine learning method which based on support vector regression (SVR) technique are presented and compared, when applied to the S22 behavior of a Gallium Nitride (GaN) high electron mobility transistor (HEMT). The device under test (DUT) has a width of 8 × 125 μm, with a gate feature size of 0.25 μm. The proposed method identifies the effect that the bias voltage and extrinsic elements have on the S22 kink shape. Additionally, compared to ECM, the SVR model attains a superior fitting accuracy across the complete frequency band.
- 17 February 2022
- Published by Wiley in International Journal of Numerical Modelling: Electronic Networks, Devices and Fields
- Vol. 35 (4)
Analysis and modeling of the kink effect in S22 based on support vector machine for GaN HEMTs
In this work, the kink effect (KE), typically visible in S22, is analyzed and modeled. Two different modeling techniques: equivalent circuit modeling (ECM) method and machine learning method which based on support vector regression (SVR) technique are presented and compared, when applied to the S22 behavior of a Gallium Nitride (GaN) high electron mobility transistor (HEMT). The device under test (DUT) has a width of 8 × 125 μm, with a gate feature size of 0.25 μm. The proposed method identifies the effect that the bias voltage and extrinsic elements have on the S22 kink shape. Additionally, compared to ECM, the SVR model attains a superior fitting accuracy across the complete frequency band.
- Open Access
- 28 May 2020
- Published by MDPI AG in Electronics
- Vol. 9 (6) , 898
128.76–129.56 GHz Fundamental Voltage Control Oscillator in 65 nm CMOS
This paper presented a fundamental voltage control oscillator (VCO), and its operating frequency ranged from 128.76 GHz to 129.56 GHz. A differential low negative coupling shield switch (DLNCSS) inductor was proposed and analyzed, which helped to improve the tuning range and phase noise of the presented VCO. Meanwhile, an improved transformer-coupled resonant tank, which consisted of DLNCSS inductor and differential varactor using the common-cathode structure, was used to improve the phase noise of the proposed VCO further. This presented VCO was designed and fabricated by using 65 nm CMOS process, which occupied a compact area of 0.23 mm2, including all testing pads. The measurement results showed that this VCO consumed 11.2 mW from a 1.4 V supply and had the tuning range of 0.8 GHz. The output power ranged from −11.8 dBm to −10.6 dBm in its operating frequency band. The post-layout simulation showed that the phase noise at 1 MHz was better than −82 dBc/Hz.
- 28 May 2020
- Published by MDPI AG in Electronics
- Vol. 9 (6) , 898
Open Access128.76–129.56 GHz Fundamental Voltage Control Oscillator in 65 nm CMOS
This paper presented a fundamental voltage control oscillator (VCO), and its operating frequency ranged from 128.76 GHz to 129.56 GHz. A differential low negative coupling shield switch (DLNCSS) inductor was proposed and analyzed, which helped to improve the tuning range and phase noise of the presented VCO. Meanwhile, an improved transformer-coupled resonant tank, which consisted of DLNCSS inductor and differential varactor using the common-cathode structure, was used to improve the phase noise of the proposed VCO further. This presented VCO was designed and fabricated by using 65 nm CMOS process, which occupied a compact area of 0.23 mm2, including all testing pads. The measurement results showed that this VCO consumed 11.2 mW from a 1.4 V supply and had the tuning range of 0.8 GHz. The output power ranged from −11.8 dBm to −10.6 dBm in its operating frequency band. The post-layout simulation showed that the phase noise at 1 MHz was better than −82 dBc/Hz.
- 29 August 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Biomedical Circuits and Systems
- Vol. 11 (6) , 1478-1487
A Surface Acoustic Wave Pumped Lensless Microfluidic Imaging System for Flowing Cell Detection and Counting
The future point-of-care diagnostics requires miniaturizing the existing bulky and expensive bioanalysis instruments, where lab-on-CMOS-chip-based technology can provide a promising solution. In this paper, we presented a surface acoustic wave (SAW) pumped lensless microfluidic imaging system for flowing cell detection and counting. Different from the previous lensless systems, which employ external bulky syringe pump for cell driven, the developed system directly integrates the SAW pump on the CMOS image sensor chip to drive the cell-containing microfluid. Moreover, an efficient temporal-differencing-based motion detection algorithm is proposed for continuous flowing cell detection and counting. Experimental results show that the SAW pump can drive the cells to flow at different driven powers, and also can keep the channel temperature below 40 °C so as not to harm the cells. The human bone marrow stromal cells flowing in the microfluidic channel can be automatically detected and counted with a low statistical error rate of −6.53%. The developed system thereby is competitive for point-of-care cell detection and counting application.
- 29 August 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Biomedical Circuits and Systems
- Vol. 11 (6) , 1478-1487
A Surface Acoustic Wave Pumped Lensless Microfluidic Imaging System for Flowing Cell Detection and Counting
The future point-of-care diagnostics requires miniaturizing the existing bulky and expensive bioanalysis instruments, where lab-on-CMOS-chip-based technology can provide a promising solution. In this paper, we presented a surface acoustic wave (SAW) pumped lensless microfluidic imaging system for flowing cell detection and counting. Different from the previous lensless systems, which employ external bulky syringe pump for cell driven, the developed system directly integrates the SAW pump on the CMOS image sensor chip to drive the cell-containing microfluid. Moreover, an efficient temporal-differencing-based motion detection algorithm is proposed for continuous flowing cell detection and counting. Experimental results show that the SAW pump can drive the cells to flow at different driven powers, and also can keep the channel temperature below 40 °C so as not to harm the cells. The human bone marrow stromal cells flowing in the microfluidic channel can be automatically detected and counted with a low statistical error rate of −6.53%. The developed system thereby is competitive for point-of-care cell detection and counting application.