Transrectal ultrasound (TRUS) imaging is a cost-effective and non-invasive modality widely used in the diagnosis of prostate cancer. The computer-aided diagnosis (CAD) relying on TRUS images has been extensively investigated recently. Compared to static images, TRUS video provides richer spatial-temporal information, which make it a promising alternative for improving the accuracy and robustness of CAD systems. However, TRUS video analysis also introduces new challenges. These include information redundancy, which increases computational costs; high intra- and inter-class similarity, which complicates feature extraction; and a low signal-to-noise ratio, which hinders the identification of clinically relevant information. To address these problems, we propose a heuristic frame selection (HFS) and a three-branch collaborative feature learning network (HFS-TriNet) for prostate cancer classification from TRUS videos. Specifically, selecting a clip of video frames at intervals for training can mitigate redundancy. The HFS strategy dynamically initializes the starting point of each training clip, which ensures that the sampled clips span the entire video sequence. For better feature extraction, besides a regular ResNet50 branch, we also utilize 1) a large model branch based a pre-trained medical segment anything model (SAM) to extract deep features of each frame and a normalization-based attention module to explore the temporal consistency; and 2) a wavelet transform convolutional residual (WTCR) branch that extracts lesion edge information in the high-frequency domain and performs denoising in the low-frequency domain.

Accurate prostate segmentation from transrectal ultrasound (TRUS) images is the key to the computer-aided diagnosis of prostate cancer. However, this task faces serious challenges, including various interferences, variational prostate shapes, and insufficient datasets. To address these challenges, a region-adaptive transformer convolution fusion net (TCF-Net) for accurate and robust segmentation of TRUS images is proposed. As a high-performance segmentation network, the TCF-Net contains a hierarchical encoder-decoder structure with two main modules: (1) a region-adaptive transformer-based encoder to identify and localize prostate regions, which learns the relationship between objects and pixels. It assists the model in overcoming various interferences and prostate shape variations. (2) A convolution-based decoder to improve the applicability to small datasets. Besides, a patch-based fusion module is also proposed to introduce an inductive bias for fine prostate segmentation. TCF-Net is trained and evaluated on a challenging clinical TRUS image dataset collected from the First Affiliated Hospital of Jinan University in China. The dataset contains 1000 TRUS images of 135 patients. Experimental results show that the mIoU of TCF-Net is 94.4%, which exceeds other state-of-the-art (SOTA) models by more than 1%.

Current convolutional neural network-based ultrasound automatic classification models for prostate cancer often rely on extensive manual labeling. Although Self-supervised Learning (SSL) have shown promise in addressing this problem, those data that from medical scenarios contains intra-class similarity conflicts, so using loss calculations directly that include positive and negative sample pairs can mislead training. SSL method tends to focus on global consistency at the image level and does not consider the internal informative relationships of the feature map. To improve the efficiency of prostate cancer diagnosis, using SSL method to learn key diagnostic information in ultrasound images, we proposed a self-supervised dual-head attentional bootstrap learning network (SDABL), including Online-Net and Target-Net. Self-Position Attention Module (SPAM) and adaptive maximum channel attention module (CAAM) are inserted in both paths simultaneously. They captures position and inter-channel attention and of the original feature map with a small number of parameters, solve the information optimization problem of feature maps in SSL. In loss calculations, we discard the construction of negative sample pairs, and instead guide the network to learn the consistency of the location space and channel space by drawing closer to the embedding representation of positive samples continuously. We conducted numerous experiments on the prostate Transrectal ultrasound (TRUS) dataset, experiments show that our SDABL pre-training method has significant advantages over both mainstream contrast learning methods and other attention-based methods. Specifically, the SDABL pre-trained backbone achieves 80.46% accuracy on our TRUS dataset after fine-tuning.

Objective:To explore the application value of transrectal ultrasound images classification network model of prostate cancer based on deep learning in the classification of benign and malignant prostate tissue in transrectal ultrasound images.Methods:A total of 1 462 two-dimensional images of transrectal prostate biopsy with clear pathologic results(including 658 images of malignant tumor, 804 images of benign tumor) from 203 patients with suspicious prostate cancer(including 89 cases of malignant tumor, 114 cases of benign tumor) were collected from May 2018 to May 2021 in the First Affiliated Hospital of Jinan University. They were divided into the training database, validation database, and test database. And the training and validation database were used to train and obtain the intelligence-assisted diagnosis network model, and then the test database was used to test the network model and two ultrasound doctors of different ages. With pathologic diagnosis as the gold standard, the diagnostic performance among them was evaluated.Results:①The sensitivity of network model was 66.7% the specificity was 91.9%, the accuracy was 80.5%, the precision(positive predictive value) was 87.1%. The area under the ROC curve was 0.922. ②The accuracy of the junior and senior ultrasound doctors was 57.5%, 62.0%; the specificity was 62.0%, 66.3%; the sensitivity was 51.5%, 56.8%; the precision was 53.1%, 58.1%, respectively. ③The accuracy, sensitivity, specificity, precision of classification: the network model > the ultrasound doctors, the differences were significant( P<0.05); the senior ultrasound doctor>the junior ultrasound doctor, the differences were not significant( P>0.05). Conclusions:The intelligence-assisted diagnosis network model based on deep learning can classify benign and malignant prostate tissue in transrectal ultrasound images, improve the accuracy of ultrasound doctors in diagnosing prostate cancer. It is of great significance to improve the efficiency of screening for patients with high clinical suspicion of prostate cancer.

The Gleason scoring system is a reliable method for quantifying the aggressiveness of prostate cancer, which provides an important reference value for clinical assessment on therapeutic strategies. However, to the best of our knowledge, no study has been done on the pathological grading of prostate cancer from single ultrasound images. In this work, a novel Automatic Region-based Gleason Grading (ARGG) network for prostate cancer based on deep learning is proposed. ARGG consists of two stages: (1) a region labeling object detection (RLOD) network is designed to label the prostate cancer lesion region; (2) a Gleason grading network (GNet) is proposed for pathological grading of prostate ultrasound images. In RLOD, a new feature fusion structure Skip-connected Feature Pyramid Network (CFPN) is proposed as an auxiliary branch for extracting features and enhancing the fusion of high-level features and low-level features, which helps to detect the small lesion and extract the image detail information. In GNet, we designed a synchronized pulse enhancement module (SPEM) based on pulse-coupled neural networks for enhancing the results of RLOD detection and used as training samples, and then fed the enhanced results and the original ones into the channel attention classification network (CACN), which introduces an attention mechanism to benefit the prediction of cancer grading. Experimental performance on the dataset of prostate ultrasound images collected from hospitals shows that the proposed Gleason grading model outperforms the manual diagnosis by physicians with a precision of 0.830. In addition, we have evaluated the lesions detection performance of RLOD, which achieves a mean Dice metric of 0.815.

The computer-aided diagnosis of prostate ultrasound images can aid in the detection and treatment of prostate cancer. However, the ultrasound images of the prostate sometimes come with serious speckle noise, low signal-to-noise ratio, and poor detection accuracy. To overcome this shortcoming, we proposed a deep learning model that integrates S-Mask R-CNN and Inception-v3 in the ultrasound image-aided diagnosis of prostate cancer in this paper. The improved S-Mask R-CNN was used to realize the accurate segmentation of prostate ultrasound images and generate candidate regions. The region of interest align algorithm was used to realize the pixel-level feature point positioning. The corresponding binary mask of prostate images was generated by the convolution network to segment the prostate region and the background. Then, the background information was shielded, and a data set of segmented ultrasound images of the prostate was constructed for the Inception-v3 network for lesion detection. A new network model was added to replace the original classification module, which is composed of forward and back propagation. Forward propagation mainly transfers the characteristics extracted from the convolution layer pooling layer below the pool_3 layer through the transfer learning strategy to the input layer and then calculates the loss value between the classified and label values to identify the ultrasound lesion of the prostate. The experimental results showed that the proposed method can accurately detect the ultrasound image of the prostate and segment prostate information at the pixel-level simultaneously. The proposed method has higher accuracy than that of the doctor’s manual diagnosis and other detection methods. Our simple and effective approach will serve as a solid baseline and help ease future research in the computer-aided diagnosis of prostate ultrasound images. Furthermore, this work will promote the development of prostate cancer ultrasound diagnostic technology.

以高分二号遥感影像中的高速公路收费站为研究对象,选取了北京、山西、河南、广东、福建5个省市2019年的高速公路收费站点位和0.8 m遥感影像,通过图像预处理、样本标注、裁切、数据增强、样本集划分的步骤制作训练样本集;引入"多尺度特征融合"的方法对SSD目标检测模型进行改进,通过增加"转置卷积"和"拼接"操作,将高层次特征图像的语义特征赋予低层次特征图像,以增强上采样质量与特征融合能力,从而提升了模型对小目标收费站的检测效果;将改进SSD模型用于2019年福建省高分二号影像中的收费站点位提取,沿福建省高速公路路网矢量对影像进行自动切片,将切片输入模型中进行目标检测;保留有收费站的切片,使用非极大值抑制去除多余的检测框,将剩余的检测框的坐标变换为中心点的坐标,可以直接输出得到高速公路收费站的中心点矢量,从而实现对于收费站点位的端到端自动化提取.研究结果表明:改进SSD模型的精度、召回率及二者的调和平均数分别为0.86、0.88和0.87,均优于传统的SSD,VGG,Faster R-CNN和特征金字塔网络模型.可见,对收费站点位的自动提取可以大大提高公路管理者的工作效率,有效满足公路管理者的实际工作需求.

人工智能(AI)技术现已用于辅助经直肠超声(TRUS)诊断前列腺癌(PCa),主要包括建立计算机辅助诊断系统和预测模型及引导前列腺穿刺活检,以提高诊断效率和诊断准确率,并降低检查成本.本文对基于AI的TRUS诊断PCa研究进展进行综述.

Improving health-related physical fitness plays an important role in enhancing the comprehensive physical fitness of college students. The traditional test of health-related physical fitness is characterized by complicated operations and low efficiency. A set of test methods for health-related physical fitness of college students based on an intelligent mobile terminal is designed in this work. The intelligent test method calculates the health-related physical fitness level of users through image acquisition and analysis by combining intelligent terminal software and test items of key elements. To verify the validity of the proposed method, a total of 116 college students (59 males and 57 females) aged between 18 and 22 are chosen as test respondents. The health-related physical fitness of respondents is tested and compared by using the traditional and proposed methods. The traditional test method reports the cardiorespiratory fitness scores of 69.6 ± 9.5 (M) and 77.1 ± 9.9 (F), and the proposed method reports 70.9 ± 9.7 (M) and 77.7 ± 9.8 (F). Twenty samples are chosen randomly to calculate the correlation coefficients r = 0.944 (M) and 0.965 (F) and significance p = 0.145 (M) and 0.489 (F). The flexibility fitness scores in the traditional method are 74.6 ± 11.9 (M) and 73.3 ± 11.5 (F), and the scores of the proposed method are 74.8 ± 11.5 (M) and 75.0 ± 11.2 (F). Twenty samples are chosen randomly to calculate r = 0.944 (M) and 0.938 (F) and p = 0.941 (M) and 0.098 (F). In the proposed method, muscle strength/muscular endurance fitness and a body composition module are tested by the traditional method, and test data are input into an artificial input system. The experimental results verify the accuracy of the proposed method in evaluating the health-related physical fitness of college students. Application of the proposed method can effectively reduce the cost of physical fitness testing, increase the convenience of testing, and direct substantial attention to the health-related physical fitness of college students.