Medical imaging can intuitively show people the internal structure, morphological information, and organ functions of the organism, which is one of the most important inspection methods in clinical medical diagnosis. Currently used medical imaging methods can only be applied to some diagnostic occasions after qualitative lesions have been generated, and the general imaging technology is usually accompanied by radiation and other conditions. However, electrical impedance tomography has the advantages of being noninvasive and non-radiative. EIT (Electrical Impedance Tomography) is also widely used in the early diagnosis and treatment of some diseases because of these advantages. At present, EIT is relatively mature and more and more image reconstruction algorithms are used to improve imaging resolution. Hardware technology is also developing rapidly, and the accuracy of data collection and processing is continuously improving. In terms of clinical application, EIT has also been used for pathological treatment of lungs, the brain, and the bladder. In the future, EIT has a good application prospect in the medical field, which can meet the needs of real-time, long-term monitoring and early diagnosis. Aiming at the application of EIT in the treatment of lung pathology, this article reviews the research progress of EIT, image reconstruction algorithms, hardware system design, and clinical applications used in the treatment of lung diseases. Through the research and introduction of several core components of EIT technology, it clarifies the characteristics of EIT system complexity and its solutions, provides research ideas for subsequent research, and once again verifies the broad development prospects of EIT technology in the future.

Accurate defect detection is an important element in ensuring product quality and safe equipment operation. However, due to the lack of deep cross-modal interactions during vision feature extraction, existing methods often suffer from attention bias, which ultimately limits detection accuracy. To address this issue, this paper proposes a Vision-Language Cyclic Interaction Model (VLCIM), which progressively optimizes vision feature extraction by integrating domain prior knowledge and generic large model, effectively bridging the dual-domain barrier between “generic-specific” and “vision-language”. Specifically, progressive cyclic interaction learning is proposed for the first time, which integrates a recursive guidance module (RGM) and cross-modal interaction (CMI) strategy to realize bidirectional dynamic fusion and collaborative optimization of vision and language features. Furthermore, the proposed dual-view synergistic detection mechanism enhances discriminative decision responses, significantly improving the model’s boundary perception ability and decision-making accuracy in complex scenarios. VLCIM achieves high-precision defect detection by establishing a cyclic interaction mechanism between domain-specific language features and vision representations. Experimental results on three industrial datasets demonstrate that VLCIM achieves improvements of 5.9%, 5.6%, and 4.1% in mIoU over the state-of-the-art (SOTA) methods, indicating its validity and generalization in different scenarios.

This article focuses on the event-triggered adaptive fuzzy tracking control for a pneumatic servo system (PSS) with input voltage saturation and position output constraint. First, a state observer is introduced to estimate the unmeasurable state through the displacement sensor signal. Second, the barrier Lyapunov function is constructed to guarantee that the output constraint is not violated. Meanwhile, the event-triggered adaptive controller is constructed to ensure the control performance of the PSS as well as to reduce communication burden. It is proved that the proposed control strategy can guarantee that the convergence of the tracking error can be ensured, all the signals in the closed-loop PSS are bounded, and position output does not transgress the constrained set. Finally, the experimental results show that the control precision and response speed can be improved by using the proposed algorithm.

Non aqueous phase liquid (NAPL) pollution exists in porous media such as soil. SVE technology can be used to remove this pollution in soil. However, few domestic and international studies have paid attention to the changes of soil temperature in the field, which we believe can be useful information to optimize the layout of heating wells. In this research we established partial differential equations of soil heat transfer using the COMSOL multi-field coupling tool to simulate the field distribution of the change in soil internal temperature in the process of SVE to obtain the change of effective heating area with time under certain initial heating conditions. At the same time, we used liquid ethylbenzene to represent NAPL pollutants, and designed a simulation of soil temperature field distribution under the movement of liquid ethylbenzene under external pressure. It was found that under the action of Darcy’s velocity field, the utilization efficiency of the SVE system for the heat source was significantly improved, that is, the temperature distribution of the soil was more uniform. However, the temperature of the heated area increased slowly because the extraction well took away the heat energy. The heat source power should be increased or prolonged to improve the effect of Darcy’s field. Through a coupled simulation, we obtained a variation relationship of the soil temperature field in 1800 min under the action of one extraction well and four heating wells. These data will provide the basis for our next step in designing an algorithm to optimize the distribution of heating wells.