Classification of cervical lesions based on multimodal features fusion.

Cervical cancer is a severe threat to women's health worldwide with a long cancerous cycle and a clear etiology, making early screening vital for the prevention and treatment. Based on the dataset provided by the Obstetrics and Gynecology Hospital of Fudan University, a four-category classification model for cervical lesions including Normal, low-grade squamous intraepithelial lesion (LSIL), high-grade squamous intraepithelial lesion (HSIL) and cancer (Ca) is developed. Considering the dataset characteristics, to fully utilize the research data and ensure the dataset size, the model inputs include original and acetic colposcopy images, lesion segmentation masks, human papillomavirus (HPV), thinprep cytologic test (TCT) and age, but exclude iodine images that have a significant overlap with lesions under acetic images. Firstly, the change information between original and acetic images is introduced by calculating the acetowhite opacity to mine the correlation between the acetowhite thickness and lesion grades. Secondly, the lesion segmentation masks are utilized to introduce prior knowledge of lesion location and shape into the classification model. Lastly, a cross-modal feature fusion module based on the self-attention mechanism is utilized to fuse image information with clinical text information, revealing the features correlation. Based on the dataset used in this study, the proposed model is comprehensively compared with five excellent models over the past three years, demonstrating that the proposed model has superior classification performance and a better balance between performance and complexity. The modules ablation experiments further prove that each proposed improved module can independently improve the model performance.

An improvement method for pancreas CT segmentation using superpixel-based active contour.

Objective.Pancreas is one of the most challenging organs for Computed Tomograph (CT) image automatic segmentation due to its complex shapes and fuzzy edges. It is simple and universal to use the traditional segmentation method as a post-processor of deep learning method for segmentation accuracy improvement. As the most suitable traditional segmentation method for pancreatic segmentation, the active contour model (ACM), still suffers from the problems of weak boundary leakage and slow contour evolution speed. Therefore, a convenient post-processor for any deep learning methods using superpixel-based active contour model (SbACM) is proposed to improve the segmentation accuracy.Approach.Firstly, the superpixels with strong adhesion to edges are used to guide the design of narrowband and energy function. A multi-scale evolution strategy is also proposed to reduce the weak boundary leakage and comprehensively improve the evolution speed. Secondly, using the original image and the coarse segmentation results obtained from deep learning methods as inputs, the proposed SbACM method is used as a post-processor for fine segmentation. Finally, the pancreatic segmentation public dataset TCIA from the National Institutes of Health(NIH, USA) is used for evaluation, and the Wilcoxon Test confirmed that the improvement of proposed method is statistically significant.Main results.(1) the superpixel-based narrowband shape and dynamic edge energy of the proposed SbACM work for boundary leakage reduction, as well as the multi-scale evolution strategy and dynamic narrowband width for the evolution speed improvement; (2) as a post-processor, SbACM can increase the Dice similarity coefficients (DSC) of five typical UNet-based models, including UNet, SS-UNet, PBR UNet, ResDSN, and nnUNet, 2.35% in average and 9.04% in maximum. (3) Based on the best backbone nnUNet, the proposed post-processor performs better than either adding edge awareness or adding edge loss in segmentation enhancement without increasing the complexity and training time of deep learning models.Significance.The proposed SbACM can improve segmentation accuracy with the lowest cost, especially in cases of squeezed fuzzy edges with similar neighborhood , and complex edges.

High-security automatic path planning of radiofrequency ablation for liver tumors.

BACKGROUND AND OBJECTIVE: Radiofrequency ablation (RFA) is an effective method for the treatment of liver tumors. Preoperative path planning, which plays a crucial role in RFA treatment, requires doctors to have significant experience and ability. Specifically, correct and highly active preoperative path planning should ensure the safety of the whole puncturing process, complete ablation of tumors and minimal damage to healthy tissues. METHODS: In this paper, a high-security automatic multiple puncture path planning method for liver tumors is proposed, in which the optimization of the ablation number, puncture number, target positions and puncture point positions subject to comprehensive clinical constraints are studied. In particular, both the safety of the puncture path and the distribution of ablation ellipsoids are taken into consideration. The influence of each constraint on the safety of the whole puncturing process is discussed in detail. On this basis, the efficiency of the planning method is obviously improved by simplifying the computational data and optimized variables. In addition, the performance and adaptability of the proposed method to large and small tumors are compared and summarized. RESULTS: The proposed method is evaluated on 10 liver tumors of various geometric characteristics from 7 cases. The test results show that the average path planning time and average ablation efficiency are 41.4 s and 60.19%, respectively. For tumors of different sizes, the planning results obtained from the proposed method have similar healthy tissue coverage. Through the clinical evaluation of doctors, the planning results meet the needs of RFA for liver tumors. CONCLUSIONS: The proposed method can provide reasonable puncture paths in RFA planning, which is beneficial to ensure the safety and efficiency of liver tumor ablation.

A fully automatic surgical registration method for percutaneous abdominal puncture surgical navigation.

Surgical registration that maps surgical space onto image space plays an important role in surgical navigation. Accurate surgical registration can help surgeons efficiently locate surgical instruments. The complicated marker-based surgical registration method is highly accurate, but it is time-consuming. Therefore, a marker-less surgical registration method with high-precision and high-efficiency is proposed without human intervention. Firstly, the surgical navigation system based on the multi-vision system is calibrated by using a specially-designed calibration board. When extracting the abdominal point cloud acquired by the structured light vision system, the constraint is constructed by using Computed Tomography (CT) image to filter out the points in irrelevant areas to improve the computational efficiency. The Coherent Point Drift (CPD) algorithm based on Gaussian Mixture Model (GMM) is applied in the registration of abdominal point cloud with lack of surface features. To enhance the efficiency of the CPD algorithm, firstly, the system calibration result is used in rough registration of the point cloud, and then the proper point cloud pretreatment method and its parameters are studied through experiments. Finally, the puncturing simulation experiments were carried out by using the abdominal phantom. The experimental results show that the proposed surgical registration method has high accuracy and efficiency, and has potential clinical application value.