Existing B-ultrasound images suffer from high noise and low contrast, which can easily affect lesion identification accuracy. This paper proposes a B-mode ultrasound video image enhancement and automatic detection system based on computer artificial intelligence. First, adaptive median filtering and histogram equalization are performed on the original video frames to achieve noise suppression and brightness compensation. Secondly, a deep feature extraction model based on a convolutional neural network (CNN) is constructed, incorporating a multi-scale attention mechanism to enhance the fidelity of edge details in target structures. Finally, a segmentation and detection module based on a U-Net architecture is designed to automatically locate and accurately segment tumors or critical tissues. Experiments were conducted on 500 clinical B-ultrasound video samples. The results showed that the proposed system improved the image signal-to-noise ratio (SNR) to 31.3 dB, the mean structural similarity index (SSIM) to 0.932, and the mean average precision (mAP) of object detection to 93.5%. These results validate the effectiveness and practicality of the system for B-ultrasound image visualization and lesion identification.

Tao Zeng, Tao Wu, Jing Zhou. Research on B-ultrasound Video Image Enhancement and Detection System Based on Computer Artificial Intelligence. International Journal of Multimedia Computing (2022), Vol. 3, Issue 2: 52-59. https://doi.org/10.38007/IJMC.2022.030207.

[1] Wang Yongfei, Wang Chun. Low-light medical color image enhancement algorithm based on improved Retinex algorithm [J]. Journal of Shandong University of Aeronautics and Astronautics, 2021, 42(2): 52-60

[2] Xu Li, Liu Liang, Zhao Fengjun. Multimodal rigid medical image layered enhancement simulation under grayscale transformation [J]. Computer Simulation, 2020, 41(4): 250-254

[3] Deng Erqiang, Qin Zhen, Zhu Guosong. EnGAN: Enhanced generative adversarial network in medical image segmentation [J]. Computer Application Research, 2019, 41(7): 2195-2202

[4] Shangguan Hongting, Liu Yuhong. A review of medical image enhancement methods [J]. Chinese Journal of Medical Physics, 2018, 40(4): 410-415

[5] Wang Jian, Cheng Chufan, Chen Fang. Realistic medical image enhancement method based on multi-loss hybrid adversarial function and heuristic projection algorithm [J]. Data Acquisition and Processing, 2020, 38(5): 1104-1111

[6] Dinh P H, Giang N L. A new medical image enhancement algorithm using adaptive parameters [J]. International Journal of Imaging Systems and Technology, 2019, 32(6): 2198-2218.

[7] Li H, Lin Z, Qiu Z, et al. Enhancing and adapting in the clinic: Source-free unsupervised domain adaptation for medical image enhancement [J]. IEEE Transactions on Medical Imaging, 2020, 43(4): 1323-1336.

[8] Chen Z, Pawar K, Ekanayake M, et al. Deep learning for image enhancement and correction in magnetic resonance imaging—state-of-the-art and challenges [J]. Journal of Digital Imaging, 2021, 36(1): 204-230.

[9] Wu H T, Cao X, Gao Y, et al. Fundus image enhancement via semi-supervised GAN and anatomical structure preservation [J]. IEEE Transactions on Emerging Topics in Computational Intelligence, 2020, 8(1): 313-326.

[10] Goceri E. Medical image data augmentation: techniques, comparisons and interpretations [J]. Artificial intelligence review, 2018, 56(11): 12561-12605.

[11] Hua H, Li Y, Wang T, et al. Edge computing with artificial intelligence: A machine learning perspective [J]. ACM Computing Surveys, 2021, 55(9): 1-35.

[12] Weber P, Carl K V, Hinz O. Applications of explainable artificial intelligence in finance—a systematic review of finance, information systems, and computer science literature [J]. Management Review Quarterly, 2020, 74(2): 867-907.

[13] Jaber T A. Artificial intelligence in computer networks [J]. Periodicals of Engineering and Natural Sciences, 2018, 10(1): 309-322.