Найдено 5
Ничего не найдено
,
Laser-Treated Surfaces for VADs: From Inert Titanium to Potential Biofunctional Materials
Bock E., Pfleging W., Tada D., Macedo E., Premazzi N., Sá R., Solheid J., Besser H., Andrade A.
Q1
Objective. Laser-treated surfaces for ventricular assist devices. Impact Statement. This work has scientific impact since it proposes a biofunctional surface created with laser processing in bioinert titanium. Introduction. Cardiovascular diseases are the world’s leading cause of death. An especially debilitating heart disease is congestive heart failure. Among the possible therapies, heart transplantation and mechanical circulatory assistance are the main treatments for its severe form at a more advanced stage. The development of biomaterials for ventricular assist devices is still being carried out. Although polished titanium is currently employed in several devices, its performance could be improved by enhancing the bioactivity of its surface. Methods. Aiming to improve the titanium without using coatings that can be detached, this work presents the formation of laser-induced periodic surface structures with a topology suitable for cell adhesion and neointimal tissue formation. The surface was modified by femtosecond laser ablation and cell adhesion was evaluated in vitro by using fibroblast cells. Results. The results indicate the formation of the desired topology, since the cells showed the appropriate adhesion compared to the control group. Scanning electron microscopy showed several positive characteristics in the cells shape and their surface distribution. The in vitro results obtained with different topologies point that the proposed LIPSS would provide enhanced cell adhesion and proliferation. Conclusion. The laser processes studied can create new interactions in biomaterials already known and improve the performance of biomaterials for use in ventricular assist devices.
Multiphoton Microscopes Go Big: Large-Scale In Vivo Imaging of Neural Dynamics
Pakan J.M., Tang Y.
Q1
Preoperative Prediction of Lymph Node Metastasis in Colorectal Cancer with Deep Learning
Liu H., Zhao Y., Yang F., Lou X., Wu F., Li H., Xing X., Peng T., Menze B., Huang J., Zhang S., Han A., Yao J., Fan X.
Q1
Objective. To develop an artificial intelligence method predicting lymph node metastasis (LNM) for patients with colorectal cancer (CRC). Impact Statement. A novel interpretable multimodal AI-based method to predict LNM for CRC patients by integrating information of pathological images and serum tumor-specific biomarkers. Introduction. Preoperative diagnosis of LNM is essential in treatment planning for CRC patients. Existing radiology imaging and genomic tests approaches are either unreliable or too costly. Methods. A total of 1338 patients were recruited, where 1128 patients from one centre were included as the discovery cohort and 210 patients from other two centres were involved as the external validation cohort. We developed a Multimodal Multiple Instance Learning (MMIL) model to learn latent features from pathological images and then jointly integrated the clinical biomarker features for predicting LNM status. The heatmaps of the obtained MMIL model were generated for model interpretation. Results. The MMIL model outperformed preoperative radiology-imaging diagnosis and yielded high area under the curve (AUCs) of 0.926, 0.878, 0.809, and 0.857 for patients with stage T1, T2, T3, and T4 CRC, on the discovery cohort. On the external cohort, it obtained AUCs of 0.855, 0.832, 0.691, and 0.792, respectively (T1-T4), which indicates its prediction accuracy and potential adaptability among multiple centres. Conclusion. The MMIL model showed the potential in the early diagnosis of LNM by referring to pathological images and tumor-specific biomarkers, which is easily accessed in different institutes. We revealed the histomorphologic features determining the LNM prediction indicating the model ability to learn informative latent features.
Real-Time High-Resolution MRI Endoscopy at up to 10 Frames per Second
Liu X., Karmarkar P., Voit D., Frahm J., Weiss C.R., Kraitchman D.L., Bottomley P.A.
Q1
Objective. Atherosclerosis is a leading cause of mortality and morbidity. Optical endoscopy, ultrasound, and X-ray offer minimally invasive imaging assessments but have limited sensitivity for characterizing disease and therapeutic response. Magnetic resonance imaging (MRI) endoscopy is a newer idea employing tiny catheter-mounted detectors connected to the MRI scanner. It can see through vessel walls and provide soft-tissue sensitivity, but its slow imaging speed limits practical applications. Our goal is high-resolution MRI endoscopy with real-time imaging speeds comparable to existing modalities. Methods. Intravascular (3 mm) transmit-receive MRI endoscopes were fabricated for highly undersampled radial-projection MRI in a clinical 3-tesla MRI scanner. Iterative nonlinear reconstruction was accelerated using graphics processor units connected via a single ethernet cable to achieve true real-time endoscopy visualization at the scanner. MRI endoscopy was performed at 6-10 frames/sec and 200-300 μm resolution in human arterial specimens and porcine vessels ex vivo and in vivo and compared with fully sampled 0.3 frames/sec and three-dimensional reference scans using mutual information (MI) and structural similarity (3-SSIM) indices. Results. High-speed MRI endoscopy at 6-10 frames/sec was consistent with fully sampled MRI endoscopy and histology, with feasibility demonstrated in vivo in a large animal model. A 20-30-fold speed-up vs. 0.3 frames/sec reference scans came at a cost of ~7% in MI and ~45% in 3-SSIM, with reduced motion sensitivity. Conclusion. High-resolution MRI endoscopy can now be performed at frame rates comparable to those of X-ray and optical endoscopy and could provide an alternative to existing modalities, with MRI’s advantages of soft-tissue sensitivity and lack of ionizing radiation.
