Lung ultrasound comets are "comet-tail" artifacts appearing in lung ultrasound images. They are particularly useful in detecting several lung pathologies and may indicate the amount of extravascular lung water. However, the comets are not always well defined and large variations in the counting results exist between observers. This study uses a convolutional neural network to quantify these lung ultrasound comets on a 4864-image clinical lung ultrasound dataset labeled by the authors. The neural network counted the number of comets correctly on 43.4% of the images and has an intraclass correlation (ICC) of 0.791 with respect to human counting on the test set. The ICC level indicates a higher correlation level than previously reported ICC between human observers. The neural network was then deployed and applied to a clinical 6272-image dataset. The correlation between the automated comet counts and the clinical parameters was examined. The comet counts correlate positively with the diastolic blood pressure (p = 0.047, r = 0.448), negatively with ejection fraction (p = 0.061, r = -0.513), and negatively with BMI (p = 0.009, r = -0.566). The neural network can be alternatively formulated as a diagnostic test for comet-positive images with 80.8% accuracy. The results could potentially be improved with a larger dataset and a refined approach to the neural networks used.

Background/aims

Recently, ultrasound signals termed 'lung water comets' associated with pulmonary edema have been correlated with adverse clinical events in dialysis patients. These comets fluctuate substantially during the ultrasound exam highlighting the need for objective quantitative measurement methods.

Methods

We developed an image-processing algorithm for the detection and quantification of lung comets. Quantification measures included comet number (comet count) and the fraction of the ultrasound beams with comet findings (comet fraction). We used this algorithm in a pilot study in 20 stable dialysis outpatients to identify associations between ultrasound comets and clinical parameters including blood pressure (BP), percent blood volume reduction on dialysis (%BV), ejection fraction (EF), and ultrafiltration on dialysis (UF).

Results

Positive findings included associations with lung comet measurements with pre-dialysis Diastolic BP (r = 0.534, p = 0.015), subject age (r = -0.446, p = 0.049), and a combination of EF and end dialysis %BV reduction (r = -0.585, p = 0.028). Comet fraction and comet count were closely correlated due to the inherent relationship between these two metrics (r = 0.973, p < 0.001). Negative findings included ultrasound comets that did not change from beginning to end of dialysis (p = 0.756), and were not significantly correlated with single dialysis treatment UF (p = 0.522), subject body weight (p = 0.208), or BMI (p = 0.358).

Conclusions

Ultrasound signal processing methods may help quantify lung ultrasound comets. Additional findings include algorithmic lung comet measurement that did not change significantly during single dialysis sessions in these stable outpatients, but were associated with cardiovascular and fluid status parameters.

Semi-crystalline thermoplastics are an important class of biomaterials with applications in creating extracorporeal and implantable medical devices. In situ release of nitric oxide (NO) from medical devices can enhance their performance via NO's potent anti-thrombotic, bactericidal, anti-inflammatory, and angiogenic activity. However, NO-releasing semi-crystalline thermoplastic systems are limited and the relationship between polymer crystallinity and NO release profile is unknown. In this paper, the functionalization of poly(ether-block-amide) (PEBA), Nylon 12, and polyurethane tubes, as examples of semi-crystalline polymers, with the NO donor S-nitroso-N-acetylpenicillamine (SNAP) within, is demonstrated via a polymer swelling method. The degree of crystallinity of the polymer plays a crucial role in both SNAP impregnation and NO release. Nylon 12, which has a relatively high degree of crystallinity, exhibits an unprecedented NO release duration of over 5 months at a low NO level, while PEBA tubing exhibits NO release over days to weeks. As a new biomedical application of NO, the NO-releasing PEBA tubing is examined as a cannula for continuous subcutaneous insulin infusion. The released NO is shown to enhance insulin absorption into the bloodstream probably by suppressing the tissue inflammatory response, and thereby could benefit insulin pump therapy for diabetes management.

Semi-crystalline thermoplastics are an important class of biomaterials with applications in creating extracorporeal and implantable medical devices. In situ release of nitric oxide (NO) from medical devices can enhance their performance via NO's potent anti-thrombotic, bactericidal, anti-inflammatory, and angiogenic activity. However, NO-releasing semi-crystalline thermoplastic systems are limited and the relationship between polymer crystallinity and NO release profile is unknown. In this paper, the functionalization of poly(ether-block-amide) (PEBA), Nylon 12, and polyurethane tubes, as examples of semi-crystalline polymers, with the NO donor S-nitroso-N-acetylpenicillamine (SNAP) within, is demonstrated via a polymer swelling method. The degree of crystallinity of the polymer plays a crucial role in both SNAP impregnation and NO release. Nylon 12, which has a relatively high degree of crystallinity, exhibits an unprecedented NO release duration of over 5 months at a low NO level, while PEBA tubing exhibits NO release over days to weeks. As a new biomedical application of NO, the NO-releasing PEBA tubing is examined as a cannula for continuous subcutaneous insulin infusion. The released NO is shown to enhance insulin absorption into the bloodstream probably by suppressing the tissue inflammatory response, and thereby could benefit insulin pump therapy for diabetes management.

Nitric oxide (NO) has many important physiological functions, including its ability to inhibit platelet activation and serve as potent antimicrobial agent. The multiple roles of NO in vivo have led to great interest in the development of biomaterials that can deliver NO for specific biomedical applications. Herein, we report a simple solvent impregnation technique to incorporate a nontoxic NO donor, S-nitroso-N-acetylpenicillamine (SNAP), into a more biocompatible biomedical grade polymer, CarboSil 20 80A. The resulting polymer-crystal composite material yields a very stable, long-term NO release biomaterial. The SNAP impregnation process is carefully characterized and optimized, and it is shown that SNAP crystal formation occurs in the bulk of the polymer after solvent evaporation. LC-MS results demonstrate that more than 70% of NO release from this new composite material originates from the SNAP embedded CarboSil phase, and not from the SNAP species leaching out into the soaking solution. Catheters prepared with CarboSil and then impregnated with 15 wt % SNAP provide a controlled NO release over a 14 d period at physiologically relevant fluxes and are shown to significantly reduce long-term (14 day) bacterial biofilm formation against Staphylococcus epidermidis and Pseudonomas aeruginosa in a CDC bioreactor model. After 7 h of catheter implantation in the jugular veins of rabbit, the SNAP CarboSil catheters exhibit a 96% reduction in thrombus area (0.03 ± 0.01 cm²/catheter) compared to the controls (0.84 ± 0.19 cm²/catheter) (n = 3). These results suggest that SNAP impregnated CarboSil can become an attractive new biomaterial for use in preparing intravascular catheters and other implanted medical devices.