Segmentation of prostate region in magnetic resonance images based on improved V-Net / 生物医学工程学杂志

Magnetic resonance (MR) imaging is an important tool for prostate cancer diagnosis, and accurate segmentation of MR prostate regions by computer-aided diagnostic techniques is important for the diagnosis of prostate cancer. In this paper, we propose an improved end-to-end three-dimensional image segmentation network using a deep learning approach to the traditional V-Net network (V-Net) network in order to provide more accurate image segmentation results. Firstly, we fused the soft attention mechanism into the traditional V-Net's jump connection, and combined short jump connection and small convolutional kernel to further improve the network segmentation accuracy. Then the prostate region was segmented using the Prostate MR Image Segmentation 2012 (PROMISE 12) challenge dataset, and the model was evaluated using the dice similarity coefficient (DSC) and Hausdorff distance (HD). The DSC and HD values of the segmented model could reach 0.903 and 3.912 mm, respectively. The experimental results show that the algorithm in this paper can provide more accurate three-dimensional segmentation results, which can accurately and efficiently segment prostate MR images and provide a reliable basis for clinical diagnosis and treatment.

Meticulous Management of the Whole Life Cycle of Medical Equipment in Hospital / 中国医疗器械杂志

OBJECTIVE@#Aiming at the different characteristics of the various stages of medical equipment life cycle in hospital, research on the targeted and meticulous management mode.@*METHODS@#Divides the whole life cycle of medical equipment in hospital into four phases, which are the selection demonstration period, purchase acceptance period, maintenance period, and retirement disposal period, and comparison with human fetal period, infant stage, adult stage and old age.@*RESULTS@#With the meticulous management mode, the service quality of medical equipment in hospital has been improved, and the service benefits have been enhanced.@*CONCLUSIONS@#According to the respective characteristics of different stages, the corresponding meticulous management mode is implemented to make the management more scientific and standardized, and the operation is safer and more reliable, which escorts the whole life cycle of medical equipment in hospital.

Effects of Different Geometric Parameters on Flexibility of Z-Shaped Stent-Grafts / 医用生物力学

Objective To analyze the influence of different geometric parameters on flexibility of the commonly used Z-shaped stent-grafts for treating thoracic aortic aneurysm, as well as the primary and secondary order of such influence. Methods The three-dimensional models of the stent-grafts with different strut numbers, wire diameters, crest height, bending radius were established by SolidWorks and imported to ANSYS software for finite element analysis．The 60° rotation of X-axis was applied to the stent-grafts, and the flexibility of the stent-grafts was evaluated according to 3 evaluation parameters (Von Mises stress, reaction force and bending torque)． Results After bending of the stent-grafts, the maximum stress was concentrated on the inside of the bend；reducing the wire diameter, crest height and strut number of the stent-grafts, the flexibility of the stent-grafts would increase; increasing the bending radius， the flexibility of the stent grafts would increase; the effect of the wire diameter and strut number on flexibility of the stent-grafts was stronger than that from the bending radius and crest height． Conclusions The strut number, wire diameter, crest height, bending radius had a significant impact on flexibility of the stent-grafts. The research findings can provide theoretical references for the selection and optimal design of the stent-grafts in clinic, and have a positive influence on reducing the incidence rate of complications such as new entry.

Breast cancer risk prediction model based on improved local ternary pattern algorithm / 中国医学影像技术

Objective To explore the value of new and fused conventional texture features extracted from mammograms using improved local ternary patterns (LTP) in predicting risk of breast cancer.Methods Mammograms were segmented.Based on improved LTP,the new and conventional texture features were extracted from segmented mammograms of bilateral breasts.Then the features of bilateral breasts were merged.The high dimensional characteristics were reduced with principal component analysis (PCA).Finally,the new texture features were classified with k-nearest neighbor (KNN),and the fusion features were clustered with logistic alternating decision tree (LADTree) algorithm.Results The area under ROC curve (AUC) of new texture features for predicting breast cancer was 0.732 4 ±0.042 8,and the sensitivity,specificity and prediction accuracy was 72.04％ (67/93),74.51％ (76/102) and 73.33％ (143/195),respectively.Furthermore,AUC of fusion features was 0.865 5± 0.014 8,the sensitivity,specificity and prediction accuracy was 84.95％ (79/93),88.23％ (90/102) and 86.67％ (169/195),respectively.Conclusion The new texture features based on improved LTP have high prediction accuracy for breast cancer,and the prediction efficacy can be improved after fusion with conventional features.

Prediction of near-term breast cancer risk based on virtual optical density image / 中国医学影像技术

Objective To investigate the value of improving the prediction accuracy of near-term risk for developing breast cancer by transforming the original mammography image and fusing the different types of image features using the algorithm of machine learning.Methods The craniocaudal (CC) full-field digital mammography (FFDM) of 185 women were downloaded from the clinical database at the university of Pittsburgh medical center.Firstly,the original gray images were segmented and transformed into virtual optical density images.Then the asymmetry features were separately extracted from original gray images and virtual optical density images.Two decision tree classifiers of the first stage were trained based on the features extracted from two types of image.And the scores output from the two classifiers were used as input to train the second stage of one decision tree classifier.Leave-one-case-out method was used to validate the prediction performance of near-term risk of breast cancer.Results Using two-stage decision tree fusion method to predict breast cancer,the area under the ROC curve (AUC) was 0.9612±0.0132.And the sensitivity,specificity and prediction accuracy were 96.63％(86/89),91.67％(88/96) and 94.05％(174/185).Conclusion The features extracted from virtual optical density image have higher discriminatory power of predicting breast cancer.Fusing the two kinds of image features twice by two-stage decision tree method can help to improve the prediction accuracy of near-term risk of breast cancer.

Design and In Vitro Experimental Study of an Endoscopic Multiple-clip Applier / 生物医学工程学杂志

Considering the problems such as reposition limited, easily detached and singly fired of the existing clip products, we developed an endoscopic multiple-clip applier which can apply 4 clips fired successively at a time. Th instrument also equipped with an independent grasper which can be used to clamp target tissues. In order to explor its feasibility and effectiveness of endoluminal closure of gastric perforation, 22 pig stomachs were making a 1 cm full-thickness incision from outside and closed by multiple-clip applier (n = 12) in vitro. Outcome was measured by bursting pressure and compared with negative control (n = 5) and hand suture (n = 5). We set a threshold pressure value (10 mm Hg) for a secure closure. Except 2 cases of invalid data, the mean bursting pressures of negative control, multiple-clip applier, hand suture were (1.5 ± 0.3) mm Hg, (46.0 ± 7.1) mm Hg, and (72.5 ± 7.7) mm Hg, respectively. The results showed that bursting pressure of multiple-clip applier was significantly higher than that of negative control (P < 0.05) and threshold value. Multiple-clip applier can be served as an effective and safe device to perform the endoluminal closure of gastric perforation.

Effects of MiR-194 on the metastasis of human osteosarcoma cell line U2-OS by recombinant lentivirus vector / 中国肿瘤临床

Objective:This study aimed to construct a lentiviral expression vector for microRNA-194 and investigate its effect on the metastasis of human osteosarcoma cell line U2-OS. Methods:Pri-and mature miR-194 amplified by PCR were inserted into the plenty-GFP vector and identified by restriction endonuclease digestion and nucleotide sequencing. The osteosarcoma cell line U2-OS was transfected with the lentivirus. Then, the stable transfected cells were used in Transwell and wound healing assay. Results:Restric-tion analysis and sequencing showed that the recombinant lentiviral expression vector was constructed correctly. The titers of obtained overexpression and suppression expression recombinant lentivirus were 1.5*108 and 4*108 TU/ml. Cell metastasis ability was signifi-cantly different in different experimental groups (P<0.01). Conclusion:The lentiviral expression vector for microRNA-194 was suc-cessfully constructed. MicroRNA-194 could influence the metastasis of the osteosarcoma cell line U2-OS;thus, it could be further ex-plored as a potential target in osteosarcoma therapy.

XRII image distortion correction for C-arm-based surgical navigation system / 生物医学工程学杂志

XRII image distortion correction is a key technique in a C-arm-based surgical navigation system. In this paper, we present a new correction method, which integrates the method of moving least squares(MLS) and the method of multilevel B-spline approximation (MBA). Experiments show that the proposed method leads to good outcome of image correction and it is more accurate than the exsiting correction methods.

C-arm based surgical navigation system / 中国组织工程研究

C-arm based surgical navigation is a typical computer assisted surgery technique; it takes XR II images as carriers,and helps surgeons with high quality operation by demonstrate operating entity virtually. In this paper, we focused on some issues concerning C-arm based surgical navigation technique, including the system operation theory, key techniques, consisting parts,main application, advantages and shortcomings. On this basis, we also discussed the development trend of this kind of surgical navigation technique.

Object identification and data extraction in XRII images for C-arm based surgical navigation / 中国组织工程研究

It is a key technique of C-arm based surgical navigation to recognize circular objects and extract their geometric data from XRII images. Previous methods possess low detecting & extracting accuracy and low reliability. In this paper, we proposed a hybrid object detecting algorithm. Firstly, an improved Circle Hough Transform (CHT) was used to obtain the accumulative space, and the section of the space was used to acquire a binarized image. Secondly, connected component analysis method was used to recognize circular objects and extract their areas and center coordinates. In the improved CHT, mask and integral operator were redefined. In the connected component analysis, a new circle measurement was used. Results of the study showed that the proposed algorithm possesses high detecting ratio, high detecting accuracy, and reliability.

An improved design of calibration target used in C-arm-based surgical navigation system / 生物医学工程学杂志

Seeing that there are several defects in the calibration target currently used in C-arm-based surgical navigation system, we propose an improved design of calibration target based on the theory of generalized mapping. The method of design is introduced, the analysis is made, and the results are reported. The newly designed calibration target is of low manufacturing difficulty and cost; with the new calibration target, the running efficiency and the image quality of the whole navigation system could be improved. The new calibration target is found to be of several advantages over the currently used ones; it is feasible and is under manufacturing already.

A method for extracting landmark geometric data from X-ray images / 生物医学工程学杂志

Extracting landmark geometric data in C-arm X-ray images is a key technique of camera calibration process in a fluoroscopic-image-based surgical navigation system. The geometric data is the basis of X-ray image distortion correction and C-arm camera calibration. In this paper we present an integrated clustering algorithm for landmark geometric data extraction. This proposed alorithm, with the advantages of an adaptive thresholding method as well as of a fast quasi connected-component labeling method, needs only one pass to extract the needed geometric data in a gray scale image. It is of high executive efficiency and good robustness as well.