RESUMEN
Recently, the use of photon counting detectors in computed tomography (PCCT) has attracted extensive attention. It is highly desired to improve the quality of material basis image and the quantitative accuracy of elemental composition, particularly when PCCT data is acquired at lower radiation dose levels. In this work, we develop a physics-constrained and noise-controlled diffusion model, PRECISION in short, to address the degraded quality of material basis images and inaccurate quantification of elemental composition mainly caused by imperfect noise model and/or hand-crafted regularization of material basis images, such as local smoothness and/or sparsity, leveraged in the existing direct material basis image reconstruction approaches. In stark contrast, PRECISION learns distribution-level regularization to describe the feature of ideal material basis images via training a noise-controlled spatial-spectral diffusion model. The optimal material basis images of each individual subject are sampled from this learned distribution under the constraint of the physical model of a given PCCT and the measured data obtained from the subject. PRECISION exhibits the potential to improve the quality of material basis images and the quantitative accuracy of elemental composition for PCCT.
RESUMEN
The employment of flexible piezoresistive sensors has sparked growing interest within the realm of wearable electronic devices, specifically in the fields of health detection and e-skin. Nevertheless, the advancement of piezoresistive sensors has been impeded by their limited sensitivity and restricted operating ranges. Consequently, it is imperative to fabricate sensors with heightened sensitivity and expanded operating ranges through the utilization of the appropriate methodologies. In this paper, piezoresistive sensors were fabricated utilizing electrospun polyvinylidene fluoride/polyacrylonitrile/polyethylene-polypropylene glycol multilayer fibrous membranes anchored with polypyrrole granules as the sensing layer, while electrospun thermoplastic polyurethane (TPU) fibers were employed as the flexible substrate. The sensitivity of the sensor is investigated by varying the fiber diameter of the sensing layer. The experimental findings reveal that a concentration of 14 wt % in the spinning solution exhibits high sensitivity (996.7 kPa-1) within a wide working range (0-10 kPa). This is attributed to the favorable diameter of the fibers prepared at this concentration, which facilitates the uniform in situ growth of pyrrole. The highly deformable TPU flexible fibers and multilayer sensing layer structure enable different linear responses across a broad pressure range (0-1 MPa). Furthermore, the sensor demonstrates good cyclic stability and can detect human movements under different pressures. These results suggest that the piezoresistive sensor with a wide operating range and high sensitivity has significant potential for future health monitoring and artificial intelligence applications.
RESUMEN
Background: This study aimed to assess the feasibility, safety, and accuracy of a low-dose CT fluoroscopy-guided remote-controlled robotic real-time puncture procedure. Methods: The study involved two control groups with Taguchi method: Group A, which underwent low-dose traditional CT-guided manual puncture (blank control), and Group B, which underwent conditional control puncture. Additionally, an experimental group, Group C, underwent CT fluoroscopy-guided remote-controlled robotic real-time puncture. In a phantom experiment, various simulated targets were punctured, while in an animal experiment, attempts were made to puncture targets in different organs of four pigs. The number of needle adjustments, puncture time, total puncture operation time, and radiation dose were analyzed to evaluate the robot system. Results: Successful punctures were achieved for each target, and no complications were observed. Dates were calculated for all parameters using Taguchi method. Conclusion: The low-dose CT fluoroscopy-guided puncture robot system is a safe, feasible, and equally accurate alternative to traditional manual puncture procedures.
RESUMEN
BACKGROUND: Microvascular invasion (MVI) is an independent prognostic factor that is associated with early recurrence and poor survival after resection of hepatocellular carcinoma (HCC). However, the traditional pathology approach is relatively subjective, time-consuming, and heterogeneous in the diagnosis of MVI. The aim of this study was to develop a deep-learning model that could significantly improve the efficiency and accuracy of MVI diagnosis. MATERIALS AND METHODS: We collected H&E-stained slides from 753 patients with HCC at the First Affiliated Hospital of Zhejiang University. An external validation set with 358 patients was selected from The Cancer Genome Atlas database. The deep-learning model was trained by simulating the method used by pathologists to diagnose MVI. Model performance was evaluated with accuracy, precision, recall, F1 score, and the area under the receiver operating characteristic curve. RESULTS: We successfully developed a MVI artificial intelligence diagnostic model (MVI-AIDM) which achieved an accuracy of 94.25% in the independent external validation set. The MVI positive detection rate of MVI-AIDM was significantly higher than the results of pathologists. Visualization results demonstrated the recognition of micro MVIs that were difficult to differentiate by the traditional pathology. Additionally, the model provided automatic quantification of the number of cancer cells and spatial information regarding MVI. CONCLUSIONS: We developed a deep learning diagnostic model, which performed well and improved the efficiency and accuracy of MVI diagnosis. The model provided spatial information of MVI that was essential to accurately predict HCC recurrence after surgery.
Asunto(s)
Carcinoma Hepatocelular , Aprendizaje Profundo , Neoplasias Hepáticas , Humanos , Carcinoma Hepatocelular/patología , Neoplasias Hepáticas/patología , Inteligencia Artificial , Estudios Retrospectivos , Invasividad NeoplásicaRESUMEN
A novel method for background signal suppression is introduced to improve the selectivity of dynamic nuclear polarization (DNP) NMR spectroscopy in the study of target molecules within complex mixtures. The method uses subtraction between positively and negatively enhanced DNP spectra, leading to an improved contrast factor, which is the ratio between the target and background signal intensities. The proposed approach was experimentally validated using a reverse-micelle system that confines the target molecules together with the polarizing agent, OX063 trityl. A substantial increase in the contrast factor was observed, and the contrast factor was optimized through careful selection of the DNP build-up time. A simulation study based on the experimental results provides insights into a strategy for choosing the appropriate DNP build-up time and the corresponding selectivity of the method. Further analysis revealed a broad applicability of the technique, encompassing studies from large biomolecules to surface-modified polymers, depending on the nuclear spin diffusion rate with a range of gyromagnetic ratios.