Jun Sun
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Jun Sun has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
Xiaojun Ji
Department of Cardiology, Wenzhou Central Hospital, Affiliated to Wenzhou Medical University
2 shared publicationsXinjun Miao
Department of Emergency, Wenzhou Central Hospital, Affiliated to Wenzhou Medical University
2 shared publicationsAbdulilah Mohammad Mayet
Electrical Engineering Department, King Khalid University
1 shared publication- Abdulilah Mohammad Mayet
Electrical Engineering Department, College of Engineering, King Khalid University, Abha 61411, Saudi Arabia
1 shared publication Bukao Ni
Department of Critical Care Medicine, Wenzhou Central Hospital, Affiliated to Wenzhou Medical University
1 shared publicationChaochao Wang
Provincial Key Laboratory of Multimodal Perceiving and Intelligent Systems, Jiaxing UniversitY
1 shared publication
Latest Publications
- Open Access
- 21 April 2026
- Published by Frontiers Media SA in Frontiers in Physiology
- Vol. 17, 1637455
An approach to machine learning-based non-invasive hemoglobin estimation using multi-wavelength PPG signal features
Introduction: The determination of hemoglobin (Hb) levels is pivotal in the clinical diagnosis and management of anemia and other blood disorders. Traditionally, Hb measurement requires blood samples, which is invasive and can cause patient discomfort and anxiety. There is a growing demand for non-invasive methods that can reduce patient stress and streamline the monitoring process. This study addresses this need by exploring the use of photoplethysmography (PPG) signals for non-invasive Hb measurement. Methods: Utilizing raw PPG data from 68 subjects, which include both red and infrared light signals, this research applies feature extraction techniques. Three key features—mean, kurtosis, and skewness—were extracted from each signal type, as applied to the dataset, resulting in a comprehensive dataset of six features per subject. These features, along with demographic data such as gender and age, were used as inputs to a Multilayer Perceptron (MLP) neural network. Results: The neural network was adept at predicting Hb levels, achieving a Mean Relative Error (MRE) of less than 2.46%. Discussion: The implications of this research are significant, offering a potential shift in how blood hemoglobin levels are measured in clinical settings. By leveraging feature extraction methods and artificial neural networks, this study not only validates the efficacy of PPG as a non-invasive diagnostic tool but also paves the way for future advancements in medical technology. The successful application of machine learning techniques in this context highlights a pathway towards more patient-friendly, efficient, and accessible health monitoring systems.
- 21 April 2026
- Published by Frontiers Media SA in Frontiers in Physiology
- Vol. 17, 1637455
Open AccessAn approach to machine learning-based non-invasive hemoglobin estimation using multi-wavelength PPG signal features
Introduction: The determination of hemoglobin (Hb) levels is pivotal in the clinical diagnosis and management of anemia and other blood disorders. Traditionally, Hb measurement requires blood samples, which is invasive and can cause patient discomfort and anxiety. There is a growing demand for non-invasive methods that can reduce patient stress and streamline the monitoring process. This study addresses this need by exploring the use of photoplethysmography (PPG) signals for non-invasive Hb measurement. Methods: Utilizing raw PPG data from 68 subjects, which include both red and infrared light signals, this research applies feature extraction techniques. Three key features—mean, kurtosis, and skewness—were extracted from each signal type, as applied to the dataset, resulting in a comprehensive dataset of six features per subject. These features, along with demographic data such as gender and age, were used as inputs to a Multilayer Perceptron (MLP) neural network. Results: The neural network was adept at predicting Hb levels, achieving a Mean Relative Error (MRE) of less than 2.46%. Discussion: The implications of this research are significant, offering a potential shift in how blood hemoglobin levels are measured in clinical settings. By leveraging feature extraction methods and artificial neural networks, this study not only validates the efficacy of PPG as a non-invasive diagnostic tool but also paves the way for future advancements in medical technology. The successful application of machine learning techniques in this context highlights a pathway towards more patient-friendly, efficient, and accessible health monitoring systems.
- Open Access
- 12 March 2025
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 1-21
FPGA implementation of a complete digital spiking silicon neuron for circuit design and network approach
When attempting to replicate the same biological spiking neuron model actions of the human brain, the spiking neuron model methodology and hardware realization design for the nervous system of the brain are crucial considerations. This work provides a modified neural modeling of complete Digital Spiking Silicon neuron model (DSSN4D). This model is capable for regenerating the basic attributes of the original model using a simplified power-2 based modeling technique. The suggested spiking neuron model is based on the fundamental power-2 based operations that can be implemented similar to the basic attributes of the main model. Removing the nonlinear parts of the main model (original one) and replacing them with modified ones leads to achieving a low-cost, low-error, and high-frequency digital system rather than the original modeling. A Xilinx Virtex-7 XC7VX690T FPGA board has been thought of and utilized for hardware realization and of the proposed model (this can validate the proposed system). The original and proposed models (in terms of neural activities) exhibit a significant degree of resemblance, according to hardware results. Additionally, greater frequency and low-cost conditions have been attained. Results of implementation indicate that overall savings are higher than for other papers and the original approach. Additionally, the new neural model’s frequency, which is roughly 502.184 MHz, is much greater than the original model’s frequency, which was 224 MHz. Also, results in hardware level shows that the proposed model takes a maximum 0.01% of the available resources of a Virtex-7 FPGA board.
- 12 March 2025
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 1-21
Open AccessFPGA implementation of a complete digital spiking silicon neuron for circuit design and network approach
When attempting to replicate the same biological spiking neuron model actions of the human brain, the spiking neuron model methodology and hardware realization design for the nervous system of the brain are crucial considerations. This work provides a modified neural modeling of complete Digital Spiking Silicon neuron model (DSSN4D). This model is capable for regenerating the basic attributes of the original model using a simplified power-2 based modeling technique. The suggested spiking neuron model is based on the fundamental power-2 based operations that can be implemented similar to the basic attributes of the main model. Removing the nonlinear parts of the main model (original one) and replacing them with modified ones leads to achieving a low-cost, low-error, and high-frequency digital system rather than the original modeling. A Xilinx Virtex-7 XC7VX690T FPGA board has been thought of and utilized for hardware realization and of the proposed model (this can validate the proposed system). The original and proposed models (in terms of neural activities) exhibit a significant degree of resemblance, according to hardware results. Additionally, greater frequency and low-cost conditions have been attained. Results of implementation indicate that overall savings are higher than for other papers and the original approach. Additionally, the new neural model’s frequency, which is roughly 502.184 MHz, is much greater than the original model’s frequency, which was 224 MHz. Also, results in hardware level shows that the proposed model takes a maximum 0.01% of the available resources of a Virtex-7 FPGA board.