The Red Kangaroo pericardium as a material source for the manufacture of percutaneous heart valves.

BACKGROUND: The kangaroo pericardium might be considered to be a good candidate material for use in the manufacture of the leaflets of percutaneous heart valves based upon the unique lifestyle. The diet consists of herbs, forbs and strubs. The kangaroo pericardium holds an undulated structure of collagen. MATERIAL AND METHOD: A Red Kangaroo was obtained after a traffic fatality and the pericardium was dissected. Four compasses were cut from four different sites: auricular (AUR), atrial (ATR), sternoperitoneal (SPL) and phrenopericardial (PPL). They were investigated by means of scanning electron microscopy, light microscopy and transmission electron microscopy. RESULTS: All the samples showed dense and wavy collagen bundles without vascularisation from both the epicardium and the parietal pericardium. The AUR and the ATR were 150±25µm thick whereas the SPL and the PPL were thinner at 120±20µm. The surface of the epicardium was smooth and glistening. The filaments of collagen were well individualized without any aggregation, but the banding was poorly defined and somewhat blurry. CONCLUSION: This detailed morphological analysis of the kangaroo pericardium illustrated a surface resistant to thrombosis and physical characteristics resistant to fatigue. The morphological characteristics of the kangaroo pericardium indicate that it represents an outstanding alternative to the current sources e.g., bovine and porcine. However, procurement of tissues from the wild raises supply and sanitary issues. Health concerns based upon sanitary uncertainty and reliability of supply of wild animals remain real problems.

Microstructural alterations owing to handling of bovine pericardium to manufacture bioprosthetic heart valves: A potential risk for cusp dehiscence.

INTRODUCTION: Cross-linking and anti-calcification of prosthetic heart valves have been continuously improved to prevent degeneration and calcification. However, non-calcific structural deteriorations such as cuspal dehiscences along the stent still require further analysis. MATERIAL AND METHOD: Based upon the previous analysis of an explanted valve after 7 years, a fresh commercial aortic valve was embedded in poly(methyl methacrylate) (PMMA) and cut into slices to ensure the detailed observation of the assembly and material structures. A pericardial patch embossed to provide the adequate shape of the cusps was investigated after paraffin embedding and appropriate staining. The microstructural damages that occurred during manufacturing process were identified and evaluated by light microscopy, polarized microscopy, scanning electron microscopy (SEM) and transmission electron microscopy (TEM). RESULTS: The wavy collagen bundles, the key structure of the pericardium patch, were damaged to a great extent at suture sites along the stent and in the compressed areas around the stent post. The fixation of the embossed pericardium patch along the plots of the stent aggravated the microstructural modifications. The damages mainly appeared as the elimination of collagen bundle waviness and delamination between the bundles. CONCLUSION: Considering the modes of failure of the explant, the damages to the collagen bundles may identify the vulnerable sites that play an important role in the cusp dehiscence of heart valve implants. Such information is important to the manufacturers. Recommendations to prevent in vivo cusp dehiscence can therefore be formulated.

Transcatheter heart valve crimping and the protecting effects of a polyester cuff.

INTRODUCTION: Prior to deployment, the percutaneous heart valves must be crimped and loaded into sheaths of diameters that can be as low as 6mm for a 23mm diameter valve. However, as the valve leaflets are fragile, any damage caused during this crimping process may contribute to reducing its long-term durability in vivo. MATERIAL AND METHOD: Bovine pericardium percutaneous valves were manufactured as follows. The leaflets were sutured on a nitinol frame. A polyester cuff fabric served as a buffer between the pericardium and the stent. Two valves were crimped and one valve was used as control. The valves were examined in gross observation and micro-CT scan and then the leaflets were processed for histology and analyzed in scanning electron microscopy, light microscopy and transmission electron microscopy. RESULT: Crimping of the valves resulted in the increase thickness of the leaflets and there was no evidence of additional delamination. The heavy prints of the stents were irregularly distributed on the outflow surface in the crimped devices and were shallow and did not penetrate throughout the thickness of the leaflets. However, the wavy microscopy of collagen fiber bundles was well preserved. They were found to remain individualized without any agglutination as shown by the regular banding appearance. CONCLUSION: Crimping of self-deployable valves per se caused only minor damages to the leaflets. However, the procedure could be refined in order to minimize areas of high pressure and swelling of the tissue that can be accompanied with flow surface disruption and increase of the hydraulic conductance. The incorporation of a polyester buffer serves to prevent the deleterious effects that may be caused if the pericardium tissue were in direct contact with the nitinol stent.

Deep learning approach towards accurate state of charge estimation for lithium-ion batteries using self-supervised transformer model.

Accurate state of charge (SOC) estimation of lithium-ion (Li-ion) batteries is crucial in prolonging cell lifespan and ensuring its safe operation for electric vehicle applications. In this article, we propose the deep learning-based transformer model trained with self-supervised learning (SSL) for end-to-end SOC estimation without the requirements of feature engineering or adaptive filtering. We demonstrate that with the SSL framework, the proposed deep learning transformer model achieves the lowest root-mean-square-error (RMSE) of 0.90% and a mean-absolute-error (MAE) of 0.44% at constant ambient temperature, and RMSE of 1.19% and a MAE of 0.7% at varying ambient temperature. With SSL, the proposed model can be trained with as few as 5 epochs using only 20% of the total training data and still achieves less than 1.9% RMSE on the test data. Finally, we also demonstrate that the learning weights during the SSL training can be transferred to a new Li-ion cell with different chemistry and still achieve on-par performance compared to the models trained from scratch on the new cell.