NAPS Fusion: A framework to overcome experimental data limitations to predict human performance and cognitive task outcomes.

In the area of human performance and cognitive research, machine learning (ML) problems become increasingly complex due to limitations in the experimental design, resulting in the development of poor predictive models. More specifically, experimental study designs produce very few data instances, have large class imbalances and conflicting ground truth labels, and generate wide data sets due to the diverse amount of sensors. From an ML perspective these problems are further exacerbated in anomaly detection cases where class imbalances occur and there are almost always more features than samples. Typically, dimensionality reduction methods (e.g., PCA, autoencoders) are utilized to handle these issues from wide data sets. However, these dimensionality reduction methods do not always map to a lower dimensional space appropriately, and they capture noise or irrelevant information. In addition, when new sensor modalities are incorporated, the entire ML paradigm has to be remodeled because of new dependencies introduced by the new information. Remodeling these ML paradigms is time-consuming and costly due to lack of modularity in the paradigm design, which is not ideal. Furthermore, human performance research experiments, at times, creates ambiguous class labels because the ground truth data cannot be agreed upon by subject-matter experts annotations, making ML paradigm nearly impossible to model. This work pulls insights from Dempster-Shafer theory (DST), stacking of ML models, and bagging to address uncertainty and ignorance for multi-classification ML problems caused by ambiguous ground truth, low samples, subject-to-subject variability, class imbalances, and wide data sets. Based on these insights, we propose a probabilistic model fusion approach, Naive Adaptive Probabilistic Sensor (NAPS), which combines ML paradigms built around bagging algorithms to overcome these experimental data concerns while maintaining a modular design for future sensor (new feature integration) and conflicting ground truth data. We demonstrate significant overall performance improvements using NAPS (an accuracy of 95.29%) in detecting human task errors (a four class problem) caused by impaired cognitive states and a negligible drop in performance with the case of ambiguous ground truth labels (an accuracy of 93.93%), when compared to other methodologies (an accuracy of 64.91%). This work potentially sets the foundation for other human-centric modeling systems that rely on human state prediction modeling.

Exploring inspiratory occlusion metrics to assess respiratory drive in patients under acute intermittent hypoxia.

Patients living with Amyotrophic Lateral Sclerosis (ALS) experience respiratory weakness and, eventually, failure due to inspiratory motor neuron degeneration. Routine pulmonary function tests (e.g., maximum inspiratory pressure (MIP)) are used to assess disease progression and ventilatory compromise. However, these tests are poor discriminators between respiratory drive and voluntary respiratory function at rest. To better understand ALS disease progression, we can look into compensatory strategies and how patients consciously react to the occlusion and the effort produced to meet the ventilatory challenge of the occlusion. This ventilatory challenge, especially beyond the P0.1 (200 ms and 300 ms), provides information regarding the patient's ability to recruit additional respiratory muscles as a compensatory strategy. Utilizing a standard P0.1 protocol to assess respiratory drive, we extend the occlusion time analysis to 200 ms and 300 ms (Detected Occlusion Response (DOR)) in order to capture compensatory respiratory mechanics. Furthermore, we followed an Acute Intermittent Hypoxia (AIH) protocol known to increase phrenic nerve discharge to evaluate the compensatory strategies. Inspiratory pressure, the rate of change in pressure, and pressure generation normalized to MIP were measured at 100 ms, 200 ms, and 300 ms after an occlusion. Airway occlusions were performed three times during the experiment (i.e., baseline, 30 and 60 minutes post-AIH). Results indicated that while AIH did not elicit change in the P0.1 or MIP, the DOR increased for ALS patients. These results support the expected therapeutic role of AIH and indicate the potential of the DOR as a metric to detect compensatory changes.

Postcardiac Arrest Neuroprognostication Practices: A Survey of Brazilian Physicians.

End-of-life care and decisions on withdrawal of life-sustaining therapies vary across countries, which may affect the feasibility of future multicenter cardiac arrest trials. In Brazil, withdrawal of life-sustaining therapy is reportedly uncommon, allowing the natural history of postcardiac arrest hypoxic-ischemic brain injury to present itself. We aimed to characterize approaches to neuroprognostication of cardiac arrest survivors among physicians in Brazil. DESIGN: Cross-sectional study. SETTING: Between August 2, 2019, and July 31, 2020, we distributed a web-based survey to physicians practicing in Brazil. SUBJECTS: Physicians practicing in Brazil and members of the Brazilian Association of Neurointensive Care, who care for patients resuscitated following cardiac arrest. INTERVENTIONS: Not applicable. MEASUREMENTS AND MAIN RESULTS: Responses from 185 physicians were obtained. Pupillary reflexes, corneal reflexes, and motor responses were considered critical to prognostication, whereas neuroimaging and electroencephalography were also regarded as important. For patients without targeted temperature management, absent pupillary and corneal reflexes at 24 hours postarrest were considered strongly predictive of poor neurologic outcome by 31.8% and 33.0%, respectively. For targeted temperature management-treated patients, absent pupillary and corneal reflexes at 24-hour postrewarming were considered prognostic by 22.9% and 20.0%, respectively. Physicians felt comfortable making definitive prognostic recommendations at day 6 postarrest or later (34.2%) for nontargeted temperature management-treated patients, and at day 6 postrewarming (20.4%) for targeted temperature management-treated patients. Over 90% believed that improving neuroprognostic accuracy would affect end-of-life decision-making. CONCLUSIONS: There is significant variability in neuroprognostic approaches to postcardiac arrest patients and timing of prognostic studies among Brazilian physicians, with practices frequently deviating from current guidelines, underscoring a need for greater neuroprognostic accuracy. Nearly all physicians believed that improving neuroprognostication will impact end-of-life decision-making. Given the tendency to delay prognostic recommendations while using similar neuroprognostic tools, Brazil offers a unique cohort in which to examine the natural history of hypoxic-ischemic brain injury in future studies.

Characterizing and Modeling Breathing Dynamics: Flow Rate, Rhythm, Period, and Frequency.

The characterization of breathing dynamics provides researchers and clinicians the ability to differentiate respiratory compensation, impairment, disease progression, ventilator assistance, and the onset of respiratory failure. However, within many sub-fields of respiratory physiology, we still have challenges identifying changes within the breathing dynamics and critical respiratory states. We discuss one fundamental modeling of breathing and how modeling imprecise assumptions decades ago regarding breathing are still propagating into our quantitative analysis today, limiting our characterization and modeling of breathing. The assumption that breathing is a continuous sinusoidal wave that can consist of a single frequency which is composed of a stationary time-invariant process has limited our expanded discussion of breathing dynamics, modeling, functional testings, and metrics. Therefore, we address major misnomers regarding breathing dynamics, specifically rate, rhythm, frequency, and period. We demonstrate how these misnomers impact the characterization and modeling through the force equations that are linked to the Work of Breathing (WoB) and our interpretation of breathing dynamics through the fundamental models and create possible erroneous evaluations of work of breathing. This discussion and simplified non-periodic WoB models ultimately sets the foundation for improved quantitative approaches needed to further our understanding of breathing dynamics, compensation, and adaptation.

Activation Complexity: A Cognitive Impairment Tool for Characterizing Neuro-isolation.

Electroencephalography (EEG) is a method for recording electrical activity, indicative of cortical brain activity from the scalp. EEG has been used to diagnose neurological diseases and to characterize impaired cognitive states. When the electrical activity of neurons are temporally synchronized, the likelihood to reach their threshold potential for the signal to propagate to the next neuron, increases. This phenomenon is typically analyzed as the spectral intensity increasing from the summation of these neurons firing. Non-linear analysis methods (e.g., entropy) have been explored to characterize neuronal firings, but only analyze temporal information and not the frequency spectrum. By examining temporal and spectral entropic relationships simultaneously, we can better characterize how neurons are isolated, (the signal's inability to propagate to adjacent neurons), an indicator of impairment. A novel time-frequency entropic analysis method, referred to as Activation Complexity (AC), was designed to quantify these dynamics from key EEG frequency bands. The data was collected during a cognitive impairment study at NASA Langley Research Center, involving hypoxia induction in 49 human test subjects. AC demonstrated significant changes in EEG firing patterns characterize within explanatory (p < 0.05) and predictive models (10% increase in accuracy). The proposed work sets the methodological foundation for quantifying neuronal isolation and introduces new potential technique to understand human cognitive impairment for a range of neurological diseases and insults.

Relative Mortality Analysis Of The "Golden Hour": A Comprehensive Acuity Stratification Approach To Address Disagreement In Current Literature.

OBJECTIVE: This study sought to address the disagreement in literature regarding the "golden hour" in trauma by using the Relative Mortality Analysis to overcome previous studies' limitations in accounting for acuity when evaluating the impact of prehospital time on mortality. METHODS: The previous studies that failed to support the "golden hour" suffered from limitations in their efforts to account for the confounding effects of patient acuity on the relationship between prehospital time and mortality in their trauma populations. The Relative Mortality Analysis was designed to directly address these limitations using a novel acuity stratification approach, based on patients' probability of survival (PoS), a comprehensive triage metric calculated using Trauma and Injury Severity Score methodology. For this analysis, the population selection and analysis methods of these previous studies were compared to the Relative Mortality Analysis on how they capture the relationship between prehospital time and mortality in the University of Virginia (UVA) Trauma Center population. RESULTS: The methods of the previous studies that failed to support the "golden hour" also failed to do so when applied to the UVA Trauma Center population. However, when applied to the same population, the Relative Mortality Analysis identified a subgroup, 9.9% (with a PoS 23%-91%), of the 5,063 patient population with significantly lower mortality when transported to the hospital within 1 hour, supporting the "golden hour." CONCLUSION: These results suggest that previous studies failed to support the "golden hour" not due to a lack of patients significantly impacted by prehospital time within their trauma populations, but instead due to limitations in their efforts to account for patient acuity. As a result, these studies inappropriately rejected the "golden hour," leading to the current disagreement in literature regarding the relationship between prehospital time and trauma patient mortality. The Relative Mortality Analysis was shown to overcome the limitations of these studies and demonstrated that the "golden hour" was significant for patients who were not low acuity (PoS >91%) or severely high acuity (PoS <23%).

Uncertainty in heart rate complexity metrics caused by R-peak perturbations.

Heart rate complexity (HRC) is a proven metric for gaining insight into human stress and physiological deterioration. To calculate HRC, the detection of the exact instance of when the heart beats, the R-peak, is necessary. Electrocardiogram (ECG) signals can often be corrupted by environmental noise (e.g., from electromagnetic interference, movement artifacts), which can potentially alter the HRC measurement, producing erroneous inputs which feed into decision support models. Current literature has only investigated how HRC is affected by noise when R-peak detection errors occur (false positives and false negatives). However, the numerical methods used to calculate HRC are also sensitive to the specific location of the fiducial point of the R-peak. This raises many questions regarding how this fiducial point is altered by noise, the resulting impact on the measured HRC, and how we can account for noisy HRC measures as inputs into our decision models. This work uses Monte Carlo simulations to systematically add white and pink noise at different permutations of signal-to-noise ratios (SNRs), time segments, sampling rates, and HRC measurements to characterize the influence of noise on the HRC measure by altering the fiducial point of the R-peak. Using the generated information from these simulations provides improved decision processes for system design which address key concerns such as permutation entropy being a more precise, reliable, less biased, and more sensitive measurement for HRC than sample and approximate entropy.

Outcomes after Falls Continue to Worsen Despite Trauma and Geriatric Care Advancements.

The most common mechanism of traumatic injury is ground-level fall. The objective of this study was to understand how patients sustaining falls and their outcomes have evolved. An institutional trauma database was used to identify adult patients who suffered a fall and were admitted to a Level I trauma center during two distinct time periods: 1998 to 2003 (past) and 2008 to 2013 (current). Data on anticoagulant use and comorbidities was gathered by retrospective chart review of patients treated during 2003 and 2013. Univariable analyses and multivariable regression were used to evaluate demographics and outcomes. A total of 6116 patients were identified, with a 24 per cent increase in number of falls between groups. Current fall patients are older (70 vs 66 years, P < 0.001), more often admitted to intensive care (28 vs 12%, P < 0.001), have longer lengths of stay (5 vs 4 days, P < 0.001), are frequently discharged to skilled nursing facilities (24 vs 8%, P < 0.001), and have higher mortality (5 vs 3%, P = 0.002). The adjusted odds of mortality for patients treated during 2003 and 2013 was associated with age, gender, injury severity score, and Glasgow Coma Scale score. Current fall patients use more health care resources and have worse outcomes, despite advances in trauma and geriatric care.

Trauma Survival Margin Analysis: A Dissection of Trauma Center Performance through Initial Lactate.

Measurement of trauma center performance presently relies on W-score calculation and comparison to national data sets. A limitation to this practice is a skewing of the W score, as it determines overall performance of a trauma population that is often heavily weighted by patients of low acuity. The University of Virginia relative mortality metric (RMM) was formulated to provide higher resolution in identifying areas of performance improvement within subpopulations of a trauma center using traditional Trauma Injury Severity Score methodology. Lactic acidosis has been established as a risk factor for mortality in the setting of trauma. This study aims to compare survival margin, defined as the area between actual and predicted mortality curves, in patients with either normal or elevated initial lactate. W score and RMM were calculated and compared in these cohorts. Whereas the W score suggested increased survival within the high initial lactate group, the RMM demonstrated the expected finding of increased survival margin in the normal lactate cohort. The RMM is a potentially valuable tool for trauma centers to monitor and improve performance. In addition, these findings validate the use of lactate as a triage and risk adjustment tool in the trauma setting.

An EMG comparative analysis of quadriceps during isoinertial strength training using nonlinear scaled wavelets.

High-speed resistance training is used to increase power; however, momentum can reduce the effectiveness of high-speed (HS) training when using weight-stack (WS) machines. This study used a non-linear scaled wavelet analysis to assess differences between pneumatic (P) and WS during seven HS or controlled speed (CS) repetitions. Vastus medialis (VM) and lateralis (VL), and rectus femoris (RF) EMG data were collected during leg extension exercises performed by five regular weight-trainers (mean age ± SD, 23.2 ± 2.9 years). Data were analyzed using continuous wavelet analysis to assess temporal Intensity distribution across eight frequency bands. Significant differences occurred due to speed for all muscles (p<.0001). P produced higher Intensity than WS for all muscles during HS (p<.0001), and VM and RF during CS (p<.001). The CON phase produced higher Intensity than ECC for the vasti muscles during CS (p<.0003), and VM and RF during HS (p<.0001). Intensity increased across repetitions plateauing earlier for the vasti than RF during CS. Regardless of the machine, Intensity levels peaked between the 25-53 Hz and 46-82 Hz (2nd and 3rd wavelets) bands. The results indicate that when the objective is increasing power through isoinertial training, P machines at HS appear to be the most effective alternative.