Studies On Imaging Detection Of Lightweight Incisional Hernia Mesh Based On ABUS Technique

Incisional hernia is an abdominal wall defect that occurs after abdominal surgery and is caused by notch infection,poor healing and so on.Once incisional hernia develops,it cannot heal on its own,and surgery is the only effective treatment.Currently,most treatments for incisional hernias involve placing a mesh into the abdominal wall to repair the abdominal wall defect.Heavyweight（HW）mesh contains more polypropylene,while lightweight（LW）mesh contains less,has lighter weight,thinner thickness and larger mesh spacing.Therefore,the use of LW mesh to repair incisional hernia can improve histocompatibility,reduce the amount of foreign body residue and inflammation.Due to people’s pursuit of minimizing the amount of foreign bodies left in the body,HW mesh was gradually replaced by the LW mesh.Therefore,in terms of incisional hernia repair materials,the LW mesh is an ideal material and has a broad application prospect.Implanting a mesh in the abdominal wall to repair an incisional hernia is a very effective method,but the implantation of a foreign body in the body can also cause a number of mesh-related complications,which requires the doctor to evaluate the diagnosis of the mesh.Currently,three medical imaging modes of computed tomography（CT）,magnetic resonance imaging（MRI）and two-dimensional handheld ultrasound（HHUS）are mainly used to detect hernia mesh in patients.However,for the LW mesh,it is not visible on CT because it is equally attenuated relative to surrounding tissue.The imaging significance of MRI and HHUS for LW mesh was also poor.Therefore,it is very important to study on imaging detection of LW mesh.In order to facilitate the subsequent algorithm to select and process the region of interest in 3D ultrasound image,a data sampling method of 3-D ultrasound images is proposed in this paper to improve the operating efficiency of subsequent image processing algorithm.The experimental results show that the proposed method can extract the representative data from the original 3D image,the optimal homogeneous cube represents the block with the highest uniformity,and the typical heterogeneous cube contains the most details and represents the block with the highest difference.The two cubes not only effectively retain the features of the original image,but also reduce the image data to 1.758 percent of the original volume.For further understanding of the texture features of lightweight mesh and laying a foundation for subsequent automatic detection of lightweight mesh,this paper proposes a lightweight mesh imaging detection method based on automated 3D breast ultrasound（ABUS）technology.The two-dimensional and three-dimensional gray level cooccurrence matrix（GLCM）was used to analyze the texture feature of the region of interest（ROI）,that is,the lightweight mesh.The results of in vitro and in vivo experiments show that the mesh texture provided by the ABUS in the coronal plane is very effective for the imaging detection of lightweight hernia mesh.In order to realize the automatic detection of lightweight hernia mesh in ABUS images,this paper proposes a lightweight hernia mesh detection algorithm based on YOLOv5 model transfer.The experimental results show that the detection accuracy is as high as 95.5%,and the detection speed also reaches the average level of 1.16 seconds per volume data,which is greatly improved compared with the 50.456 seconds spent by manual inspection.It is of great significance to promote the use of LW incisional hernia mesh and assist doctors to quickly locate the LW incisional hernia mesh area.