Applications of Machine Learning in Cardiac Electrophysiology.

Artificial intelligence through machine learning (ML) methods is becoming prevalent throughout the world, with increasing adoption in healthcare. Improvements in technology have allowed early applications of machine learning to assist physician efficiency and diagnostic accuracy. In electrophysiology, ML has applications for use in every stage of patient care. However, its use is still in infancy. This article will introduce the potential of ML, before discussing the concept of big data and its pitfalls. The authors review some common ML methods including supervised and unsupervised learning, then examine applications in cardiac electrophysiology. This will focus on surface electrocardiography, intracardiac mapping and cardiac implantable electronic devices. Finally, the article concludes with an overview of how ML may impact on electrophysiology in the future.

Early and Late Posttraumatic Epilepsy in the Setting of Traumatic Brain Injury: A Meta-analysis and Review of Antiepileptic Management.

BACKGROUND: Moderate to severe traumatic brain injury confers increased risk of posttraumatic seizures (PTSs). Early PTSs are diagnosed when seizures develop within 7 days after injury, whereas seizures diagnosed as late PTSs occur later. Patients have been treated with phenytoin (PHT) to prevent early PTSs and more recently with levetiracetam (LEV). Various regimens have been tried in patients to prevent late PTSs with variable success. We assessed and compared effectiveness of these drugs on early and late PTS prevention. METHODS: A literature search revealed 120 articles. Data were included if the same factors were compared across studies with identical treatment arms. Random effects models were used for meta-analysis to combine data into an overriding odds ratio (OR) comparing PTS incidence. For early PTSs, PHT was compared with placebo and LEV with PHT. For late PTSs, each drug was compared with placebo. RESULTS: Sixteen studies were included. PHT was associated with decreased odds of early seizures relative to placebo (OR = 0.34, 95% confidence interval [CI] 0.19-0.62). There was no difference in early seizure incidence between LEV and PHT (OR = 0.83, 95% CI 0.33-2.1). Neither LEV (OR = 0.69, 95% CI 0.24-1.96) nor PHT (OR = 0.4, 95% CI 0.1-1.6) was associated with fewer late PTSs than placebo. CONCLUSIONS: New literature is consistent with current guidelines supporting antiepileptic drug administration for prevention of early, but not late, PTSs. With regard to early PTS prevention, LEV and PHT are similarly efficacious, which is consistent with current guidelines. Side-effect profiles favor LEV administration over PHT.

Caveolin-1 polarization in transmigrating endothelial cells requires binding to intermediate filaments.

Caveolin-1 influences cell migration through multiple signaling pathways. In a previous report, we have shown that caveolin-1 is polarized in three-dimensional migrating endothelial cells (EC), and that caveolin-1 accumulation at the front of transmigrating cells requires the phosphorylatable Tyr(14) residue of caveolin-1. Immuno-electron microscopy further indicated that caveolin-1 was distributed along cytoskeletal structures in the anterior of transmigrating EC [Parat MO, Anand-Apte B, Fox PL (Mol Biol Cell 14:3156-3168, 2003)]. In the present study, we investigate whether caveolin-1 interacts with intermediate filaments (IF) and whether this interaction is required for caveolin-1 polarization in transmigrating cells. The distribution of vimentin is polarized in cells traversing a filter pore and overlaps with the distribution of caveolin-1, which accumulates in the cell front. In vivo sprouting EC also exhibit an anterior polarization of these two proteins. Furthermore, caveolin-1 co-purifies with intermediate filaments, suggesting an interaction between caveolin-1 and IF. Vimentin-deficient SW13 cells exhibit a dramatically altered polarization of caveolin-1-GFP, which no longer accumulates in the protruding cell extension. In addition, the Tyr(14) residue of caveolin-1 is required for co-purification of the protein with IF. Taken together, our results show that caveolin-1 Tyr(14) is necessary for binding to intermediate filaments, which in turn is required for anterior polarization of caveolin-1 in transmigrating cells.

The tumor inhibitor and antiangiogenic agent withaferin A targets the intermediate filament protein vimentin.

The natural product withaferin A (WFA) exhibits antitumor and antiangiogenesis activity in vivo, which results from this drug's potent growth inhibitory activities. Here, we show that WFA binds to the intermediate filament (IF) protein, vimentin, by covalently modifying its cysteine residue, which is present in the highly conserved alpha-helical coiled coil 2B domain. WFA induces vimentin filaments to aggregate in vitro, an activity manifested in vivo as punctate cytoplasmic aggregates that colocalize vimentin and F-actin. WFA's potent dominant-negative effect on F-actin requires vimentin expression and induces apoptosis. Finally, we show that WFA-induced inhibition of capillary growth in a mouse model of corneal neovascularization is compromised in vimentin-deficient mice. These findings identify WFA as a chemical genetic probe of IF functions, and illuminate a potential molecular target for withanolide-based therapeutics for treating angioproliferative and malignant diseases.

Acute effects of desmin mutations on cytoskeletal and cellular integrity in cardiac myocytes.

Mutations in desmin have been associated with a subset of human myopathies. Symptoms typically appear in the second to third decades of life, but in the most severe cases can manifest themselves earlier. How desmin mutations lead to aberrant muscle function, however, remains poorly defined. We created a series of four mutations in rat desmin and tested their in vitro filament assembly properties. RDM-G, a chimera between desmin and green fluorescent protein, formed protofilament-like structures in vitro. RDM-1 and RDM-2 blocked in vitro assembly at the unit-length filament stage, while RDM-3 had more subtle effects on assembly. When expressed in cultured rat neonatal cardiac myocytes via adenovirus infection, these mutant proteins disrupted the endogenous desmin filament to an extent that correlated with their defects in in vitro assembly properties. Disruption of the desmin network by RDM-1 was also associated with disruption of plectin, myosin, and alpha-actinin organization in a significant percentage of infected cells. In contrast, expression of RDM-2, which is similar to previously characterized human mutant desmins, took longer to disrupt desmin and plectin organization and had no significant effect on myosin or alpha-actinin organization over the 5-day time course of our studies. RDM-3 had the mildest effect on in vitro assembly and no discernable effect on either desmin, plectin, myosin, or alpha-actinin organization in vivo. These results indicate that mutations in desmin have both direct and indirect effects on the cytoarchitecture of cardiac myocytes.