Assessing the symptoms of proximal weakness caused by neurological deficits requires the knowledge and experience of neurologists. Recent advances in machine learning and the Internet of Things have resulted in the development of automated systems that emulate physicians’ assessments. The application of those systems requires not only accuracy in the classification but also reliability regardless of users’ proficiency in the real environment for the clinical point-of-care and the personalized health management. This study provides an agreement and reliability analysis of using a machine learning-based scaling of Medical Research Council (MRC) proximal scores to evaluate proximal weakness by experts and non-experts. The system trains an ensemble learning model using the signals from sensors attached to the limbs of patients in a neurological intensive care unit. For the agreement analysis, we investigated the percent agreement of MRC proximal scores and Bland-Altman plots of kinematic features between the expert- and non-expert scaling. We also analyzed the intra-class correlation coefficients (ICCs) of kinematic features and Krippendorff’s alpha of the observers’ scaling for the reliability analysis. The mean percent agreement between the expert- and the non-expert scaling was 0.542 for manual scaling and 0.708 for autonomous scaling. The ICCs of kinematic features measured using sensors ranged from 0.742 to 0.850, whereas the Krippendorff’s alpha of manual scaling for the three observers was 0.275. The autonomous assessment system can be utilized by the caregivers, paramedics, or other observers during an emergency to evaluate acute stroke patients.
Critically ill patients often develop hyperglycemia because of the metabolic response to trauma and stress. In response to any form of damage to the organism, it reacts by increasing its own glucose production which subsequently causes hyperglycemia. This adaptive reaction of the organism is directed to aid in the rapid restoration after the damage. Therefore, glucose is an indispensable substrate in the critically ill which aids the reparation process. Severe and persistent hyperglycemia is associated with unfavorable outcomes and is considered to be an independent predictor of in-hospital mortality. The discussion remains on whether hyperglycemia is just a marker of increased stress which makes it a surrogate indicator of disease severity or if it is the reason for the unfavorable outcome. A few years ago, several published articles suggested that a tight glycemic control within the normal range improves treatment outcome. Over time, researchers have changed their point of view and currently there is a discussion on this matter in the scientific literatures. At the same time, the question of what glycemic level should be maintained for patients in the Neurological Intensive Care Unit is a matter of discussion. In this review, the authors analyzed the latest guidelines on treatment of critical patients with neurosurgical and neurological pathologies, specifically the glycemic control in this category of patients.
The last decade has been marked by the advent of continuous EEG (cEEG) monitoring, which is now recommended as the standard of care in numerous medical conditions seen in the neurological intensive care unit (ICU).1, 2 In clinical practice, its main indications are seizure detection, treatment monitoring, and prognostication in various conditions such as seizures/status epilepticus, ischemic stroke, subarachnoid hemorrhage, intracerebral hemorrhage, traumatic brain injury, brain tumor, encephalitis/sepsis-associated encephalopathy, extracorporeal membrane oxygenation, targeted temperature management, and cardiac arrest.2, 3
AbstractDespite advances in research and improved neurological intensive care in recent years, the clinical outcome of severely head injured patients is still poor. Primary insult is followed by a complex cascade of molecular and biochemical events that lead to neuroinflammation, brain edema, and delayed neuronal death. No specific pharmacological therapy is currently available which prevents the development of secondary brain injuries, and most therapeutic strategies have failed in translation from bench to bedside. There are limitations of clinical and radiological methods in delineating the exact severity and prognosis of traumatic brain injury (TBI). A myriad complex biochemical markers are under investigation to delineate the extent of brain tissue damage and to independently predict the outcome, but a search for simple biomarker still eludes the research. Progesterone, a gonadal hormone and a neurosteroid, although controversial as a neuroprotective agent, may hold promise as a simple biochemical marker of the outcome in severe TBI.
Agreement and Reliability Analysis of Machine Learning Scaling and Wireless Monitoring in the Assessment of Acute Proximal Weakness by Experts and Non-Experts: A Feasibility Study
