Shipeng Han

Dr. Shipeng Han


Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
20 October 2024 in Bioengineering
Verified
Journal Article
Open access
PDF file
MHMin Hu
Min Hu
YZYaorong Zhang
Yaorong Zhang
HXHuijun Xue
Huijun Xue
HLHao Lv
Hao Lv
SHShipeng Han
Shipeng Han
Accurate segmentation of thyroid nodules in ultrasound images is crucial for the diagnosis of thyroid cancer and preoperative planning. However, the segmentation of thyroid nodules is challenging due to their irregular shape, blurred boundary, and uneven echo texture. To address these challenges, a novel Mamba- and ResNet-based dual-branch network (MRDB) is proposed. Specifically, the visual state space block (VSSB) from Mamba and ResNet-34 are utilized to construct a dual encoder for extracting global semantics and local details, and establishing multi-dimensional feature connections. Meanwhile, an upsampling–convolution strategy is employed in the left decoder focusing on image size and detail reconstruction. A convolution–upsampling strategy is used in the right decoder to emphasize gradual feature refinement and recovery. To facilitate the interaction between local details and global context within the encoder and decoder, cross-skip connection is introduced. Additionally, a novel hybrid loss function is proposed to improve the boundary segmentation performance of thyroid nodules. Experimental results show that MRDB outperforms the state-of-the-art approaches with DSC of 90.02% and 80.6% on two public thyroid nodule datasets, TN3K and TNUI-2021, respectively. Furthermore, experiments on a third external dataset, DDTI, demonstrate that our method improves the DSC by 10.8% compared to baseline and exhibits good generalization to clinical small-scale thyroid nodule datasets. The proposed MRDB can effectively improve thyroid nodule segmentation accuracy and has great potential for clinical applications.

15 November 2024 in Sensors
Verified
Journal Article
Open access
PDF file
MHMin Hu
Min Hu
YZYaorong Zhang
Yaorong Zhang
TJTeng Jiao
Teng Jiao
HXHuijun Xue
Huijun Xue
XWXue Wu
Xue Wu
HLHao Lv
Hao Lv
Small-scale pedestrian detection is one of the challenges in general object detection. Factors such as complex backgrounds, long distances, and low-light conditions make the image features of small-scale pedestrians less distinct, further increasing the difficulty of detection. To address these challenges, an Enhanced Feature-Fusion YOLO network (EFF-YOLO) for small-scale pedestrian detection is proposed. Specifically, this method employs a backbone based on the FasterNet block within YOLOv8n, which is designed to enhance the extraction of spatial features while reducing redundant operation. Furthermore, the gather-and-distribute (GD) mechanism is integrated into the neck of the network to realize the aggregation and distribution of global information and multi-level features. This not only strengthens the faint features of small-scale pedestrians but also effectively suppresses complex background information, thereby improving the accuracy of small-scale pedestrians. Experimental results indicate that EFF-YOLO achieves detection accuracies of 72.5%, 72.3%, and 91% on the three public datasets COCO-person, CityPersons, and LLVIP, respectively. Moreover, the proposed method reaches a detection speed of 50.7 fps for 1920 × 1080-pixel video streams on the edge device Jetson Orin NX, marking a 15.2% improvement over the baseline network. Thus, the proposed EFF-YOLO method not only boasts high detection accuracy but also demonstrates excellent real-time performance on edge devices.

25 April 2025 in Sensors
Verified
Journal Article
Open access
PDF file
SHShipeng Han
Shipeng Han
CCChaowen Chai
Chaowen Chai
MHMin Hu
Min Hu
YWYanni Wang
Yanni Wang
TJTeng Jiao
Teng Jiao
HLHao Lv
Hao Lv
Multimodal pedestrian detection has garnered significant attention due to its potential applications in complex scenarios. The complementarity characteristics between infrared and visible modalities can enhance detection performance. However, the design of cross-modal fusion mechanisms and the in-depth exploration of inter-modal complementarity still pose challenges. To address this, we propose TCCDNet, a novel network integrating cross-modal complementarity. Specifically, the efficient multi-scale attention C2f (EMAC) is designed for the backbone, which combines the C2f structure with an efficient multi-scale attention mechanism to achieve feature weighting and fusion, thereby enhancing the model’s feature extraction capacity. Subsequently, the cross-modal complementarity (CMC) module is proposed, which enhances feature discriminability and object localization accuracy through a synergistic mechanism combining channel attention and spatial attention. Additionally, a deep semantic fusion module (DSFM) based on a cross-attention mechanism is incorporated to achieve deep semantic feature fusion. The experimental results demonstrate that TCCDNet achieves a MR−2 of 7.87% on the KAIST dataset, representing a 3.83% reduction compared to YOLOv8. For the other two multimodal pedestrian detection datasets, TCCDNet attains mAP50 scores of 83.8% for FLIR ADAS and 97.3% for LLVIP, outperforming the baseline by 3.6% and 1.9% respectively. These results fully validate the effectiveness and advancement of the proposed method.

13 May 2026 in Bioengineering
Verified
Journal Article
Open access
PDF file
MYMingyi Yang
Mingyi Yang
NYNianzeng Yuan
Nianzeng Yuan
SHShipeng Han
Shipeng Han
TGTianjiao Guo
Tianjiao Guo
WLWen Luo
Wen Luo
HLHao Lv
Hao Lv
The Focused Assessment with Sonography in Trauma (FAST) enables rapid point-of-care screening for internal hemorrhage by detecting free fluid, but its accuracy is highly operator-dependent and prone to missed diagnoses in emergency settings. While deep learning–based AI assistance can address these limitations, most existing models rely on computationally intensive networks, restricting deployment on resource-limited bedside devices. Thus, developing a lightweight architecture for free fluid detection is essential for clinical translation. We propose a lightweight model based on YOLOX, incorporating a dual-stream fusion (DSF) backbone to preserve spatial details while reducing computation, and a global fusion feedback (GFF) neck to enhance efficient multi-scale feature fusion. We also built a dedicated dataset using ultrasound images from rabbits with active liver hemorrhage to better mimic in vivo sonographic features of free fluid. Compared with mainstream detectors, our method achieves the lowest FLOPs and parameter count while maintaining superior precision, recall, and F1-score. Ablation studies validate that DSF improves accuracy and reduces complexity, and GFF further lowers computational costs with minimal performance tradeoff. The proposed approach enables fast, accurate free fluid detection on constrained bedside devices, advancing intelligent point-of-care trauma assessment.

1 January 2025 in Journal of Electronic Imaging
Verified
Journal Article
Closed access
PDF file
XWXue Wu
Xue Wu
MHMin Hu
Min Hu
YZYaorong Zhang
Yaorong Zhang
JLJianguo Luo
Jianguo Luo
YWYanni Wang
Yanni Wang
SHShipeng Han
Shipeng Han
Trauma hemorrhage is the leading cause of death among combat casualties. Traditional human rescue methods are often constrained by ongoing firefights and terrain factors, which prevent rescuers from reaching the front lines to provide emergency care within the critical one-hour window. In recent years, with the rapid development of drones and robots, utilizing their machine vision to identify the location of body trauma hemorrhages can provide a priori information to subsequent rescue teams, promoting faster and more targeted medical interventions. Therefore, we construct for the first time a dataset comprising 5313 visible light images of body trauma hemorrhage and propose a keypoint-guided trauma hemorrhage detection network (KGDet). Specifically, the algorithm initially employs the YOLOv5 to identify bleeding areas. It then utilizes the HRNet to detect the coordinates of 77 contour keypoints on the human body. Finally, by calculating the IoU between the trauma hemorrhage areas and the body part contour regions, the algorithm accurately matches trauma hemorrhage areas to their corresponding body parts, effectively identifying the specific location of trauma hemorrhage in injured individuals. Experimental results demonstrate that the algorithm can accurately identify trauma hemorrhage sites.