Application of Artificial Intelligence in Cardiovascular Imaging

During the last two decades, as computer technology has matured and business scenarios have diversified, the scale of application of computer systems in various industries has continued to expand, resulting in a huge increase in industry data. As for the medical industry, huge unstructured data has been accumulated, so exploring how to use medical image data more effectively to efficiently complete diagnosis has an important practical impact. For a long time, China has been striving to promote the process of medical informatization, and the combination of big data and artificial intelligence and other advanced technologies in the medical field has become a hot industry and a new development trend. This paper focuses on cardiovascular diseases and uses relevant deep learning methods to realize automatic analysis and diagnosis of medical images and verify the feasibility of AI-assisted medical treatment. We have tried to achieve a complete diagnosis of cardiovascular medical imaging and localize the vulnerable lesion area. (1) We tested the classical object based on a convolutional neural network and experiment, explored the region segmentation algorithm, and showed its application scenarios in the field of medical imaging. (2) According to the data and task characteristics, we built a network model containing classification nodes and regression nodes. After the multitask joint drill, the effect of diagnosis and detection was also enhanced. In this paper, a weighted loss function mechanism is used to improve the imbalance of data between classes in medical image analysis, and the effect of the model is enhanced. (3) In the actual medical process, many medical images have the label information of high-level categories but lack the label information of low-level lesions. The proposed system exposes the possibility of lesion localization under weakly supervised conditions by taking cardiovascular imaging data to resolve these issues. Experimental results have verified that the proposed deep learning-enabled model has the capacity to resolve the aforementioned issues with minimum possible changes in the underlined infrastructure.

Effectiveness of virtual teaching of diagnostic and interventional imaging fundamentals to Egyptian medical students: an analytical cross-sectional study

Background There is a worldwide deficit in teaching and training in the field of radiology for undergraduate medical students. This educational gap is prominent in many medical schools as most radiology curricula are a part of other specialty trainings, usually provided by non-radiologists. After COVID-19 pandemic, there was an increased trend in online education. However, questions have been raised about the efficacy and acceptance of online education. We developed a course on the principles of radiology and medical imaging basics to target Egyptian medical students. We then assessed the impact of these educational videos through several online surveys. Our "The Principles of Radiology Online Course" was delivered to students at various Egyptian medical schools; it was a prerecorded series composed of nine sessions, and each session followed the sequence of a pre-test, video, and post-test. There was a final survey to assess the overall feedback. Finally, we analyzed the results to give insight onto how teaching radiology through online lectures can help build better physicians. Results Among various medical schools around Egypt, 1396 Egyptian medical students joined this cohort. Cohort population percentage was 56% female and 44% male. Ninety-eight percent of the students agreed that this program increased their understanding of radiology. Eighty-four percent of the students found the platform friendly and easy to use. Seventy-nine percent found these webinars were more convenient compared to in-person education. Statistical significance (p-value < 0.05) was achieved in all sessions after comparing students’ pre and post-test scores, and in students’ confidence and knowledge level before and after the course. Conclusions Radiology is an underrepresented subject for a lot of medical students. Online radiology webinars have proven to be a promising method of teaching medical students key medical imaging concepts. An online course of radiology basics and principles can help improve a medical student’s knowledge and enhance overall future patient care.

Benefits and limitations for the use of radiation dose management systems in medical imaging. practical experience in a university hospital

Objectives: Radiation dose management systems (DMS) are currently to help improve radiation protection in medical imaging and interventions. This study presents our experience using a homemade DMS called DOLQA (Dose On-Line for Quality Assurance). Methods: Our DMS is connected to 14 X-ray systems in a university hospital linked to the central data repository of a large network of 16 public hospitals in the Autonomous Community of Madrid, with 6.7 million inhabitants. The system allows us to manage individual patient dose data and groups of procedures with the same clinical indications, and compare them with diagnostic reference levels (DRLs). The system can also help to prioritize optimisation actions. Results: This study includes results of imaging examinations from 2020, with 3,7601 procedures and 28,6471 radiation events included in the radiation dose structured reports (RDSR), for computed tomography (CT), interventional procedures, positron emission tomography-CT (PET-CT) and mammography. Conclusions: The benefits of the system include: automatic registration and management of patient doses, creation of dose reports for patients, information on recurrent examinations, high dose alerts, and help to define optimisation actions. The system requires the support of medical physicists and implication of radiologists and radiographers. DMSs must undergo periodic quality controls and audit reports must be drawn up and submitted to the hospital’s quality committee. The drawbacks of DMSs include the need for continuous external support (medical physics experts, radiologists, radiographers, technical services of imaging equipment and hospital informatics services) and the need to include data on clinical indication for the imaging procedures. Advances in knowledge: DMS perform automatic management of radiation doses, produces patient dose reports, and registers high dose alerts to suggest optimisation actions. Benefits and limitations are derived from the practical experience in a large university hospital.

Fabrication of 2-D Capacitive Micromachined Ultrasonic Transducer (CMUT) Array through Silicon Wafer Bonding

Capacitive micromachined ultrasound transducers (CMUTs) have broad application prospects in medical imaging, flow monitoring, and nondestructive testing. CMUT arrays are limited by their fabrication process, which seriously restricts their further development and application. In this paper, a vacuum-sealed device for medical applications is introduced, which has the advantages of simple manufacturing process, no static friction, repeatability, and high reliability. The CMUT array suitable for medical imaging frequency band was fabricated by a silicon wafer bonding technology, and the adjacent array devices were isolated by an isolation slot, which was cut through the silicon film. The CMUT device fabricated following this process is a 4 × 16 array with a single element size of 1 mm × 1 mm. Device performance tests were conducted, where the center frequency of the transducer was 3.8 MHz, and the 6 dB fractional bandwidth was 110%. The static capacitance (29.4 pF) and center frequency (3.78 MHz) of each element of the array were tested, and the results revealed that the array has good consistency. Moreover, the transmitting and receiving performance of the transducer was evaluated by acoustic tests, and the receiving sensitivity was −211 dB @ 3 MHz, −213 dB @ 4 MHz. Finally, reflection imaging was performed using the array, which provides certain technical support for the research of two-dimensional CMUT arrays in the field of 3D ultrasound imaging.

Three-Dimensional Liver Image Segmentation Using Generative Adversarial Networks Based on Feature Restoration

Medical imaging provides a powerful tool for medical diagnosis. In the process of computer-aided diagnosis and treatment of liver cancer based on medical imaging, accurate segmentation of liver region from abdominal CT images is an important step. However, due to defects of liver tissue and limitations of CT imaging procession, the gray level of liver region in CT image is heterogeneous, and the boundary between the liver and those of adjacent tissues and organs is blurred, which makes the liver segmentation an extremely difficult task. In this study, aiming at solving the problem of low segmentation accuracy of the original 3D U-Net network, an improved network based on the three-dimensional (3D) U-Net, is proposed. Moreover, in order to solve the problem of insufficient training data caused by the difficulty of acquiring labeled 3D data, an improved 3D U-Net network is embedded into the framework of generative adversarial networks (GAN), which establishes a semi-supervised 3D liver segmentation optimization algorithm. Finally, considering the problem of poor quality of 3D abdominal fake images generated by utilizing random noise as input, deep convolutional neural networks (DCNN) based on feature restoration method is designed to generate more realistic fake images. By testing the proposed algorithm on the LiTS-2017 and KiTS19 dataset, experimental results show that the proposed semi-supervised 3D liver segmentation method can greatly improve the segmentation performance of liver, with a Dice score of 0.9424 outperforming other methods.