Data analytics in a clinical setting: Applications to understanding breathing patterns and their relevance to neonatal disease.

In this review, we focus on the use of contemporary linear and non-linear data analytics as well as machine learning/artificial intelligence algorithms to inform treatment of pediatric patients. We specifically focus on methods used to quantify changes in breathing that can lead to increased risk for apnea of prematurity, retinopathy of prematurity (ROP), necrotizing enterocolitis (NEC) and provide a list of potentially useful algorithms that comprise a suite of software tools to enhance prediction of outcome. Next, we provide a brief overview of machine learning/artificial intelligence methods and applications within the sphere of perinatal care. Finally, we provide an overview of the infrastructure needed to use these tools in a clinical setting for real-time data acquisition, data synchrony, data storage and access, and bedside data visualization to assist in clinical decision making and support the medical informatics mission. Our goal is to provide an overview and inspire other investigators to adopt these tools for their own research and optimization of perinatal patient care.

An Inexpensive Open-Source Chamber for Controlled Hypoxia/Hyperoxia Exposure.

Understanding hypoxia/hyperoxia exposure requires either a high-altitude research facility or a chamber in which gas concentrations are precisely and reproducibly controlled. Hypoxia-induced conditions such as hypoxic-ischemic encephalopathy (HIE), obstructive or central apneas, and ischemic stroke present unique challenges for the development of models with acute or chronic hypoxia exposure. Many murine models exist to study these conditions; however, there are a variety of different hypoxia exposure protocols used across laboratories. Experimental equipment for hypoxia exposure typically includes flow regulators, nitrogen concentrators, and premix oxygen/nitrogen tanks. Commercial hypoxia/hyperoxia chambers with environmental monitoring are incredibly expensive and require proprietary software with subscription fees or highly expensive software licenses. Limitations exist in these systems as most are single animal systems and not designed for extended or intermittent hypoxia exposure. We have developed a simple hypoxia chamber with off-the-shelf components, and controlled by open-source software for continuous data acquisition of oxygen levels and other environmental factors (temperature, humidity, pressure, light, sound, etc.). Our chamber can accommodate up to two mouse cages and one rat cage at any oxygen level needed, when using a nitrogen concentrator or premixed oxygen/nitrogen tank with a flow regulator, but is also scalable. Our system uses a Python-based script to save data in a text file using modules from the sensor vendor. We utilized Python or R scripts for data analysis, and we have provided examples of data analysis scripts and acquired data for extended exposure periods (≤7 days). By using FLOS (Free-Libre and open-source) software and hardware, we have developed a low-cost and customizable system that can be used for a variety of exposure protocols. This hypoxia/hyperoxia exposure chamber allows for reproducible and transparent data acquisition and increased consistency with a high degree of customization for each experimenter's needs.

Developing a standardized system of exposure and intervention endpoints for isoflurane in preclinical stroke models.

Isoflurane is a regularly used anesthetic in translational research. Isoflurane facilitates invasive surgery and a rapid recovery. Specifically, in the pathology of stroke, controversy has surrounded isoflurane's intrinsic neuroprotective abilities, affecting apoptosis, excitotoxicity, and blood brain barrier disruption. Due to the intrinsic neuroprotective nature and lack of standardized guidelines for the use of isoflurane, research has shifted away from this gas in most animal models. Antagonistically, studies have also reported that no neuroprotective effects are observed when a surgery is accompanied with isoflurane exposure under 20 minutes. Isoflurane affects the pathophysiology in stroke patients by altering critical pathways in endothelial, neuronal, and microglial cells. Current studies have elucidated isoflurane neuroprotection to be time dependent and may be minimized in experimental designs if the exposure time is limited to a specific window. Therefore, with detailed and extensive literature on anesthetics, we can hypothesize that isoflurane exposure under the 20-minute benchmark, behavior and molecular pathways can be evaluated at any time-point following ischemic insult without confounding artifacts from isoflurane; however, If the exposure to isoflurane exceeds 20 minutes, the acute neuroprotective effects are evident for 2 weeks in the model, which should be accounted for in molecular and behavioral assessments, with either isoflurane inhibitors or a control group at 2 weeks post middle cerebral artery occlusion. The purpose of this review is to suggest a detailed and standardized outline for interventions and behavioral assessments after the use of isoflurane in experimental designs.

Role of lipid metabolism in smoothened derepression in hedgehog signaling.

The binding of Hedgehog (Hh) to its receptor Patched causes derepression of Smoothened (Smo), resulting in the activation of the Hh pathway. Here, we show that Smo activation is dependent on the levels of the phospholipid phosphatidylinositol-4 phosphate (PI4P). Loss of STT4 kinase, which is required for the generation of PI4P, exhibits hh loss-of-function phenotypes, whereas loss of Sac1 phosphatase, which is required for the degradation of PI4P, results in hh gain-of-function phenotypes in multiple settings during Drosophila development. Furthermore, loss of Ptc function, which results in the activation of Hh pathway, also causes an increase in PI4P levels. Sac1 functions downstream of STT4 and Ptc in the regulation of Smo membrane localization and Hh pathway activation. Taken together, our results suggest a model in which Ptc directly or indirectly functions to suppress the accumulation of PI4P. Binding of Hh to Ptc derepresses the levels of PI4P, which, in turn, promotes Smo activation.