中国医疗器械杂志最新文献

Objective: To reduce the variability in detection results and enhance the stability of product quality in fully automated flow fluorescence immunoassay analyzers by introducing robust parameter design.

Methods: Applying the principles of robust parameter design, magnetic bead transfer rate was selected as the target characteristic. Magnetic rod retraction distance (A), adsorption cycles (B), and cleaning solution volume (C) were chosen as controllable factors, while environmental factors (temperature and humidity) served as noise factors and the tests were carried out by using orthogonal experimental design combined with the composite noise factor method. Results After optimizing the product design using robust design methods, the coefficient of variation (CV) in detection results decreased from approximately 3.08% $-$ 5.72% to 1.63% $-$ 2.93%. Conclusion Robust parameter design substantially improved the stability of detection results, enhanced the overall quality of the product at the design stage, and was proved to be a systematic, effective, and easily mastered design approach.

As an important part of clinical diagnosis and treatment, the safety and effectiveness of medical devices are directly related to patients' lives and health. With the continuous development of science and technology in China, a growing number of investigator-initiated trials (IIT) involve unlisted medical devices, which poses significant risks and challenges. In accordance with China's existing laws and regulations, unlisted medical devices shall not be used in investigator-initiated trials in principle. This paper proposes that by means of policy guidance and technical support provided by regulatory authorities, risk assessment and hierarchical management conducted by clinical research management departments, capacity improvement of medical and health institutions, ethics committees and investigators, and the joint establishment and improvement of a rights and interests protection system for research participants by all parties, the development of investigator-initiated trials using unlisted medical devices in China can be promoted towards a more scientific and standardized direction.

This study reviewed the pre-market regulatory requirements for the division of registration units, registration and filing status, as well as the actual sales and usage status of in vitro diagnostic reagent registration units from the perspective of full lifecycle supervision. A comparative analysis was conducted on the regulatory requirements of China, the United States, and the European Union for in vitro diagnostic reagent registration and sales units, with a focus on immune platforms, mass spectrometry platforms, and genetic testing products. This study analyzed the unit division and its risks in product classification and supervision, and provided regulatory recommendations to offer ideas and technical references for developing regulatory documents and scientific supervision of in vitro diagnostic reagent registration and sales units.

This paper elaborates on the research progress of super-resolution ultrasound imaging technique, detailing each step involved in the implementation process of this technology (ultrasound transmission and scanning, beamforming, clutter filtering, microbubble localization and tracking, etc.), and summarizes existing problems and potential solutions based on existing literature. Targeting relevant clinical applications and demand scenarios, the paper focuses on four future directions of super-resolution imaging technology, including 3D super-resolution imaging, realization of productization, contrast-free super-resolution, and artificial intelligence (AI) technology. In addition, the paper introduces the calculation methods of multiple quantitative parameters based on the function of super-resolution imaging, and demonstrates the calculation results and correlations of different parameters (fractal dimension, directional entropy, and circular variance) through simulations.

Objective: To construct a cross-organ AI model based on multi-task deep learning (DL) to achieve unified prediction of malignancy risks of thyroid and breast nodules.

Methods: Clinical data of patients with thyroid nodules ( n=2386) and breast nodules ( n=2753) from three hospitals were collected. A multi-task deep learning model was constructed based on the Transformer architecture, and a design that combines feature sharing layers with organ-specific layers was adopted. Performance was evaluated through five-fold cross-validation, and tested on an independent external validation set ( n=835), and SHAP analysis was used to explain model decisions.

Results: The constructed AI model for pan-endocrine nodules achieved an AUC of 0.932 (95%CI: 0.914-0.951) in the prediction of thyroid nodules, with a sensitivity of 86.5% and a specificity of 89.2%. The predictive AUC of breast nodules was 0.917 (95%CI: 0.896-0.938), with a sensitivity of 84.3% and a specificity of 88.7%. Compared with the single-organ model, the pan-model performed better on small sample datasets ( P<0.01), and remained stable in external validation (AUC>0.90). SHAP analysis revealed that edge irregularity, calcification type, and internal echo were common important features, while blood flow signals and TI-RADS/BI-RADS classification were organ-specific features.

Conclusion: A high-performance cross-organ malignancy risk prediction model for thyroid-breast nodules was successfully constructed, confirming that pan-endocrine nodules can be precisely stratified for risk through a unified deep learning architecture, providing a new paradigm for AI-assisted diagnosis of endocrine tumors.

The overall treatment process of adolescent idiopathic scoliosis (AIS) encompasses prevention, detection, diagnosis, and treatment. With advancements in technology and in-depth clinical research, this paper provides a comprehensive review of cutting-edge developments in the field of AIS. These include preventive and diagnostic measures such as genetic testing, optimized observational risk assessment, and the construction of risk models, as well as the evolution of precise non-surgical and minimally invasive, intelligent surgical treatment plans. The study aims to assist medical professionals in scientific research and clinical decision-making.

Injectable fillers for plastic surgery are widely used in the field of aesthetic plastic surgery, and their safety and effectiveness are of vital importance. To ensure the safety and effectiveness of the products, it is indispensable to conduct scientific, rigorous, comprehensive and standardized preclinical physicochemical evaluations. According to the raw materials, injectable fillers for plastic surgery can be classified into sodium hyaluronate type, collagen type and polyester type, etc. The evaluation items and indicators for different materials vary. Preclinical physicochemical evaluations not only include general physicochemical project evaluation, such as the content of active ingredients, pH value, osmotic pressure, total heavy metal content, trace elements, etc., but also need to carry out corresponding evaluation items according to the particularity of different materials, such as the crosslinking degree of sodium hyaluronate gel, the collagen content of collagen products and the copolymerization rate of polyester products. This article not only briefly introduces the test methods of some general evaluation items, but also provides detailed introductions to some evaluation items that have no mature methods or have room for improvement in existing methods, aiming to provide references for the research and development and quality control of injectable fillers for plastic surgery, and help improve the safety and effectiveness of injectable fillers for plastic surgery.

Objective From the perspectives of algorithmic transparency and data security to establish a practical and feasible ethical review mechanism for medical artificial intelligence (AI) research, effectively safeguarding the life, health, personal dignity, and legitimate rights and interests of research participants. Methods To sort out the basis for ethical review of medical AI research at home and abroad, analyze the problems and difficulties in relevant ethical review practices, and explore the procedures and key points of ethical review of medical AI research based on practical work. Results It has been clarified that the ethical review of medical AI research needs to strengthen interdisciplinary cooperation, invite experts from multiple fields such as software engineering and computer science to participate in the review work, optimize the process of ethical application and acceptance, standardize the way of ethical review, and develop corresponding review points based on different types of medical AI research, focusing on the evaluation of research risk-benefit ratio and the standardization of informed consent procedures. Conclusion This study proposes a standardized and effective implementation path for ethical review of medical AI research, while adhering to the basic requirements of traditional ethical review and taking into account the particularity of AI technology. The path is highly practicable.

This paper introduces and analyzes the principles and structure inside the high-voltage generator, and for the first time proposes the viewpoint that using an inverter to increase the inverter frequency can enhance the imaging performance of the digital radiography (DR) system. The theoretical basis for using an inverter to increase the inverter frequency to eliminate ripple is analyzed. Considering the characteristics of ripple, a pulse width modulation (PWM) inverter circuit based on insulated gate bipolar transistor (IGBT) is designed to suppress the ripple generated inside the high-voltage generator. For the first time, detective quantum efficiency (DQE) is introduced to quantitatively analyze the DR imaging performance before and after the installation of the inverter, in order to verify whether the addition of the inverter frequency module has improved the imaging performance.

In order to support the demonstration application of domestic innovative medical device magnetocardiography (MCG) in China and promote the systematic display and promotion of clinical diagnosis solutions, this paper designs and constructs a networked, intelligent and standardized information service system of MCG diagnosis technology for coronary microvascular dysfunction. In response to the functional and performance requirements of the system, an efficient and scalable architecture is designed, which is based on the Node.js, supported by the React (frontend) and NestJS (backend) frameworks, and JSON Web Token (JWT) authentication. This architecture enables efficient and convenient data sharing and information exchange among multiple medical institutions. Currently, the system has provided demonstration, promotion, and resource integration of MCG clinical diagnostic solutions for 10 hospitals, with a cumulative collection of 845 case report forms (CRF). It provides technical support for the clinical diagnostic solution of homegrown innovative MCG in coronary microvascular dysfunction from the perspectives of information platform and service.