Cattle-FRETS71, a novel fluorogenic substrate with broad applicability for characterizing ADAMTS13 properties and function.

BACKGROUND: Current ADAMTS13 activity assays are important for diagnosing thrombotic thrombocytopenic purpura (TTP) but are unreliable to assay ADAMTS13 activity in animal models. The Cattle-FRETS71 assay is capable of detecting ADAMTS13 activity in plasma from multiple animal species, making it a potentially useful reagent at all stages of clinical research. The performance of Cattle-FRETS71 in TTP diagnosis is not yet known. OBJECTIVES: We evaluated the performance of the Cattle-FRETS71 substrate against the human FRETS-rVWF71 and the FRETS-VWF73 commercial substrates in human plasma and serum samples to validate its utility in diagnosing TTP in patients. METHODS: Internal validation was performed using heparinized plasma samples (n = 81). External validation was a blinded study using serum samples from the Oklahoma TTP Registry (n = 118, collected 2004-2014) that had been initially assayed by FRETS-VWF73 within 1 year of collection. Additional validation was performed with citrated plasma samples with variable ADAMTS13 activities (n = 32) that were analyzed by FRETS-VWF73. RESULTS: There was an excellent correlation (r = 0.94) between Cattle-FRETS71 and FRETS-rVWF71 for assayed heparinized plasma samples (n = 81). Assay results between Cattle-FRETS71 and FRETS-VWF73 of Oklahoma TTP Registry serum samples (n = 118) and citrated plasma samples (n = 32) were comparably good (r = 0.81 and r = 0.85, respectively). CONCLUSION: The Cattle-FRETS71 assay is comparable with other assays in quantifying ADAMTS13 activity in human plasma collected from patients with documented or suspected TTP. The versatility of Cattle-FRETS71, combined with its specificity and sensitivity, makes it a useful tool for the standardization of ADAMTS13 activity across basic and clinical research paradigms.

Reinforcement Learning for Central Pattern Generation in Dynamical Recurrent Neural Networks.

Lifetime learning, or the change (or acquisition) of behaviors during a lifetime, based on experience, is a hallmark of living organisms. Multiple mechanisms may be involved, but biological neural circuits have repeatedly demonstrated a vital role in the learning process. These neural circuits are recurrent, dynamic, and non-linear and models of neural circuits employed in neuroscience and neuroethology tend to involve, accordingly, continuous-time, non-linear, and recurrently interconnected components. Currently, the main approach for finding configurations of dynamical recurrent neural networks that demonstrate behaviors of interest is using stochastic search techniques, such as evolutionary algorithms. In an evolutionary algorithm, these dynamic recurrent neural networks are evolved to perform the behavior over multiple generations, through selection, inheritance, and mutation, across a population of solutions. Although, these systems can be evolved to exhibit lifetime learning behavior, there are no explicit rules built into these dynamic recurrent neural networks that facilitate learning during their lifetime (e.g., reward signals). In this work, we examine a biologically plausible lifetime learning mechanism for dynamical recurrent neural networks. We focus on a recently proposed reinforcement learning mechanism inspired by neuromodulatory reward signals and ongoing fluctuations in synaptic strengths. Specifically, we extend one of the best-studied and most-commonly used dynamic recurrent neural networks to incorporate the reinforcement learning mechanism. First, we demonstrate that this extended dynamical system (model and learning mechanism) can autonomously learn to perform a central pattern generation task. Second, we compare the robustness and efficiency of the reinforcement learning rules in relation to two baseline models, a random walk and a hill-climbing walk through parameter space. Third, we systematically study the effect of the different meta-parameters of the learning mechanism on the behavioral learning performance. Finally, we report on preliminary results exploring the generality and scalability of this learning mechanism for dynamical neural networks as well as directions for future work.

Robotic Inguinal Hernia Repair Outcomes: Operative Time and Cost Analysis.

BACKGROUND: Robotic inguinal hernia repair is the latest iteration of minimally invasive herniorrhaphy. Previous studies have shown expedited learning curves compared to traditional laparoscopy, which may be offset by higher cost and longer operative time. We sought to compare operative time and direct cost across the evolving surgical practice of 10 surgeons in our healthcare system. METHODS: This is a retrospective review of all transabdominal preperitoneal robotic inguinal hernia repairs performed by 10 general surgeons from July 2015 to September 2018. Patients requiring conversion to an open procedure or undergoing simultaneous procedures were excluded. The data was divided to compare each surgeon's initial 20 cases to their subsequent cases. Direct operative cost was calculated based on the sum of supplies used intra-operatively. Multivariate analysis, using a generalized estimating equation, was adjusted for laterality and resident involvement to evaluate outcomes. RESULTS: Robotic inguinal hernia repairs were divided into two groups: early experience (n = 167) and late experience (n = 262). The late experience had a shorter mean operative time by 17.6 min (confidence interval: 4.06 - 31.13, p = 0.011), a lower mean direct operative cost by $538.17 (confidence interval: 307.14 - 769.20, p < 0.0001), and fewer postoperative complications (p = 0.030) on multivariate analysis. Thirty-day readmission rates were similar between both groups. CONCLUSION: Increasing surgeon experience with robotic inguinal hernia repair is associated with a predictable reduction in operative time, complication rates, and direct operative cost per case. Thirty-day readmission rates are not affected by the learning curve.