scholarly journals An interpretable machine learning model based on a quick pre-screening system enables accurate deterioration risk prediction for COVID-19

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Lijing Jia ◽  
Zijian Wei ◽  
Heng Zhang ◽  
Jiaming Wang ◽  
Ruiqi Jia ◽  
...  
Keyword(s):  
Heart Rate ◽  
Clinical Symptoms ◽  
Predictive Performance ◽  
International Normalized Ratio ◽  
Urine Specific Gravity ◽  
Prothrombin Activity ◽  
High Performing ◽  
High Risk Factors ◽  
Machine Learning Model ◽  
Interpretable Model

AbstractA high-performing interpretable model is proposed to predict the risk of deterioration in coronavirus disease 2019 (COVID-19) patients. The model was developed using a cohort of 3028 patients diagnosed with COVID-19 and exhibiting common clinical symptoms that were internally verified (AUC 0.8517, 95% CI 0.8433, 0.8601). A total of 15 high risk factors for deterioration and their approximate warning ranges were identified. This included prothrombin time (PT), prothrombin activity, lactate dehydrogenase, international normalized ratio, heart rate, body-mass index (BMI), D-dimer, creatine kinase, hematocrit, urine specific gravity, magnesium, globulin, activated partial thromboplastin time, lymphocyte count (L%), and platelet count. Four of these indicators (PT, heart rate, BMI, HCT) and comorbidities were selected for a streamlined combination of indicators to produce faster results. The resulting model showed good predictive performance (AUC 0.7941 95% CI 0.7926, 0.8151). A website for quick pre-screening online was also developed as part of the study.

scholarly journals The impact of data-complexity and team characteristics on performance in the classification model

2022 ◽  
Vol 9 (1) ◽  
pp. 0-0
Keyword(s):  
Machine Learning ◽  
Binary Classification ◽  
Class Imbalance ◽  
Predictive Ability ◽  
Predictive Performance ◽  
Classification Model ◽  
Data Complexity ◽  
High Performing ◽  
Machine Learning Model ◽  
The Impact

This article investigates the impact of data-complexity and team-specific characteristics on machine learning competition scores. Data from five real-world binary classification competitions hosted on Kaggle.com were analyzed. The data-complexity characteristics were measured in four aspects including standard measures, sparsity measures, class imbalance measures, and feature-based measures. The results showed that the higher the level of the data-complexity characteristics was, the lower the predictive ability of the machine learning model was as well. Our empirical evidence revealed that the imbalance ratio of the target variable was the most important factor and exhibited a nonlinear relationship with the model’s predictive abilities. The imbalance ratio adversely affected the predictive performance when it reached a certain level. However, mixed results were found for the impact of team-specific characteristics measured by team size, team expertise, and the number of submissions on team performance. For high-performing teams, these factors had no impact on team score.

scholarly journals Mortality prediction using a novel combination of biomarkers in the first day of sepsis in intensive care units

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Junkun Liu ◽  
Chengwen Bai ◽  
Binbin Li ◽  
Aijun Shan ◽  
Fei Shi ◽  
...  
Keyword(s):  
Septic Shock ◽  
Scoring Systems ◽  
Predictive Performance ◽  
Serum Levels ◽  
International Normalized Ratio ◽  
Individual Performance ◽  
Mortality Prediction ◽  
New Combination ◽  
Thrombin Time ◽  
Survivor Group

AbstractEarly identification of infection severity and organ dysfunction is crucial in improving outcomes of patients with sepsis. We aimed to develop a new combination of blood-based biomarkers that can early predict 28-day mortality in patients with sepsis or septic shock. We enrolled 66 patients with sepsis or septic shock and compared 14 blood-based biomarkers in the first 24 h after ICU admission. The serum levels of interleukin-6 (IL-6) (median 217.6 vs. 4809.0 pg/ml, P = 0.001), lactate (median 2.4 vs. 6.3 mmol/L, P = 0.014), N-terminal prohormone of brain natriuretic peptide (NT-proBNP) (median 1596.5 vs. 32,905.3 ng/ml, P < 0.001), prothrombin time (PT) (median 15.6 vs. 20.1 s, P = 0.030), activated partial thrombin time (APTT) (median 45.1 vs. 59.0 s, P = 0.026), and international normalized ratio (INR) (median 1.3 vs. 1.8, P < 0.001) were significantly lower in the survivor group. IL-6, NT-proBNP, and INR provided the best individual performance in predicting 28-day mortality of patients with sepsis or septic shock. Furthermore, the combination of these three biomarkers achieved better predictive performance (AUC 0.890, P < 0.001) than conventional scoring systems. In summary, the combination of IL-6, NT-proBNP, and INR may serve as a potential predictor of 28-day mortality in critically ill patients with sepsis or septic shock.

scholarly journals Predictive Value of Subacute Heart Rate Variability for Determining Outcome Following Adolescent Concussion

2021 ◽  
Vol 10 (1) ◽  
pp. 161
Author(s):  
Colt A. Coffman ◽  
Jacob J. M. Kay ◽  
Kat M. Saba ◽  
Adam T. Harrison ◽  
Jeffrey P. Holloway ◽  
...  
Keyword(s):  
Heart Rate ◽  
Heart Rate Variability ◽  
Cognitive Performance ◽  
Predictive Value ◽  
System Integration ◽  
Sports Medicine ◽  
Clinical Symptoms ◽  
Physiological Stress ◽  
Neurological Dysfunction ◽  
Neurobehavioral Function

Objective assessments of concussion recovery are crucial for facilitating effective clinical management. However, predictive tools for determining adolescent concussion outcomes are currently limited. Research suggests that heart rate variability (HRV) represents an indirect and objective marker of central and peripheral nervous system integration. Therefore, it may effectively identify underlying deficits and reliably predict the symptomology following concussion. Thus, the present study sought to evaluate the relationship between HRV and adolescent concussion outcomes. Furthermore, we sought to examine its predictive value for assessing outcomes. Fifty-five concussed adolescents (12–17 years old) recruited from a local sports medicine clinic were assessed during the initial subacute evaluation (within 15 days postinjury) and instructed to follow up for a post-acute evaluation. Self-reported clinical and depressive symptoms, neurobehavioral function, and cognitive performance were collected at each timepoint. Short-term HRV metrics via photoplethysmography were obtained under resting conditions and physiological stress. Regression analyses demonstrated significant associations between HRV metrics, clinical symptoms, neurobehavioral function, and cognitive performance at the subacute evaluation. Importantly, the analyses illustrated that subacute HRV metrics significantly predicted diminished post-acute neurobehavioral function and cognitive performance. These findings indicate that subacute HRV metrics may serve as a viable predictive biomarker for identifying underlying neurological dysfunction following concussion and predict late cognitive outcomes.

Abstract 222: Nerve Conduction Study in Postural Orthostatic Tachycardia Syndrome

2016 ◽  
Vol 9 (suppl_2) ◽  
Author(s):  
Chandralekha Ashangari ◽  
Samreen F Asghar ◽  
Sadaf Syed ◽  
Amna A Butt ◽  
Amer Suleman
Keyword(s):  
Heart Rate ◽  
Nerve Conduction Study ◽  
Nerve Conduction ◽  
Clinical Symptoms ◽  
Orthostatic Intolerance ◽  
Tibial Nerve ◽  
Sensory Nerves ◽  
Exercise Intolerance ◽  
Postural Orthostatic Tachycardia Syndrome ◽  
Medical Diagnostic

Background: Postural orthostatic tachycardia syndrome (POTS) is an autonomic disturbance characterized by the clinical symptoms of orthostatic intolerance, mainly light headedness, fatigue, sweating, tremor, anxiety, palpitation, exercise intolerance and near syncope on upright posture. These are relieved on lying down. Patients also have a heart rate >120 beats/min (bpm) on standing or increase their heart rate by 30 bpm from a resting heart rate after standing for 10 min. A nerve conduction study (NCS) is a medical diagnostic test commonly used to evaluate the function, especially the ability of electrical conduction, of the motor and sensory nerves of the human body. The aim of this study is to demonstrate median, ulnar, peroneal, tibial nerve conduction results POTS patients. Methods: 177 patients were selected randomly from our clinic with POTS. Nerve conduction results of median, ulnar, peroneal, tibial nerves were reviewed from electronic medical records. Results: Out of 177 patients, 151 patients are females (85%, n=151, age 32.07±11.10), 26 patients are males (15%, n=26, age 29.08±17.40).Median nerve conduction results are 57.83 m/sec ±7.58 m/sec, Ulnar nerve conduction results are 56.62 m/sec ±6.85 m/sec, Peroneal nerve conduction results are 49.96 m/sec ±6.85 m/sec, Tibial nerve conduction results are 50.70 m/sec ±6.86 m/sec. Conclusion: The nerve conduction velocities tend to be within normal range in Postural Orthostatic Tachycardia Syndrome (POTS) patients.

scholarly journals Morning Sympathetic Activity after Evening Binge Alcohol Consumption

2020 ◽  
Author(s):  
Ian Mark Greenlund ◽  
Hannah A. Cunningham ◽  
Anne L Tikkanen ◽  
Jeremy A Bigalke ◽  
Carl A Smoot ◽  
...  
Keyword(s):  
Heart Rate ◽  
Alcohol Consumption ◽  
Binge Drinking ◽  
Rem Sleep ◽  
Muscle Sympathetic Nerve Activity ◽  
Urine Specific Gravity ◽  
Total Urine ◽  
The Impact ◽  
Insight Into

Binge alcohol consumption elicits acute and robust increases of muscle sympathetic nerve activity (MSNA), yet the impact of evening binge drinking on morning-after MSNA is unknown. The present study examined the effects of evening binge alcohol consumption on polysomnographic sleep and morning-after MSNA. We hypothesized that evening binge drinking (i.e. 4-5 drink equivalent in <2hrs) would reduce sleep quality and increase morning-after blood pressure (BP) and MSNA. Following a familiarization night within the sleep laboratory, twenty-two participants (12 men, 10 women; 25±1 years) were examined after simulated binge drinking or fluid control (randomized, crossover design). Morning MSNA was successfully recorded across both conditions in 16 participants (8 men, 8 women) during a 10-minute baseline and three Valsalva's maneuvers (VM). Binge drinking reduced rapid eye movement (REM) sleep (15±1 vs. 20±1%; p=0.003), increased stage II sleep (54±1 vs. 51±1%; p=0.002), increased total urine output (2.9±0.2 vs. 2.1±0.1 liters; p<0.001), but did not alter morning-after urine specific gravity. Binge drinking increased morning-after heart rate (65 (54-72) vs. 58 (51-67) beats/min; p=0.013), but not resting BP or MSNA. Binge drinking elicited greater sympathoexcitation during VM (38±3 vs. 43±3 bursts/min, p=0.036). Binge drinking augmented heart rate (p=0.002), systolic BP (p=0.022) and diastolic (p=0.037) BP reactivity to VM phase IV, and blunted cardiovagal baroreflex sensitivity during VM phases II (p=0.028) and IV (p=0.043). In conclusion, evening binge alcohol consumption disrupted REM sleep and morning-after autonomic function. These findings provide new mechanistic insight into the potential role of binge drinking on cardiovascular risk.

Antihypertensive Effect of Guanfacine: Long-Term ‘Once-a-Day’ Treatment and Sudden Withdrawal

1981 ◽  
Vol 61 (s7) ◽  
pp. 469s-471s
Author(s):  
I. Szám ◽  
J. Holló
Keyword(s):  
Blood Pressure ◽  
Heart Rate ◽  
Antihypertensive Effect ◽  
Clinical Symptoms ◽  
Day Treatment ◽  
Pressure Rise ◽  
Common Side Effect ◽  
Blood Pressure Rise ◽  
Abrupt Discontinuation

1. Twenty patients with essential hypertension were treated with guanfacine given in single daily doses of 1–5 mg over a period of 24 weeks. Compared with the initial values at the end of the first wash-out period, there was a significant decrease of blood pressure and heart rate. The most common side effect, dryness of the mouth, usually disappeared after 8–10 weeks of treatment. No changes in laboratory values were seen. In the post-treatment placebo period there were significant increases in blood pressure and heart rate compared with the last readings during the treatment period. However, these never exceeded the pretreatment values. 2. In a second trial guanfacine (1–5 mg daily) was abruptly discontinued in 11 patients after 6–20 weeks' treatment. Blood pressure was measured twice a day, in lying and standing positions, during the 4 days before abrupt withdrawal of guanfacine and for 7 days after discontinuation. Clopamide was given concurrently to two patients, and this was continued after withdrawal of guanfacine. Only in two patients did the blood pressure rise to values above the initial levels (30 mmHg systolic and 10 mmHg diastolic), but no clinical symptoms were observed during the withdrawal. A transitory increase of heart rate of between 10 and 30 beats/min was observed in five patients after abrupt discontinuation of the drug.

scholarly journals Serum Level of D-Lactate in Patients with Cystic Fibrosis: Preliminary Data

2018 ◽  
Vol 2018 ◽  
pp. 1-5 ◽  
Author(s):  
Sabina Więcek ◽  
Jerzy Chudek ◽  
Halina Woś ◽  
Maria Bożentowicz-Wikarek ◽  
Bożena Kordys-Darmolinska ◽  
...  
Keyword(s):  
Cystic Fibrosis ◽  
Diet Composition ◽  
Intestinal Inflammation ◽  
Clinical Symptoms ◽  
Pancreatic Insufficiency ◽  
Lactate Concentration ◽  
Intestinal Flora ◽  
International Normalized Ratio ◽  
Exocrine Pancreatic Insufficiency ◽  
Faecal Calprotectin

D-Lactate is produced by the intestinal biota and later absorbed into circulation. Some patients with cystic fibrosis (CF) develop exocrine pancreatic insufficiency that may disturb the gut microbiome and enhance the production of D-lactate. However, this concept has not been studied yet. The aim of the study was to assess D-lactate concentration in relation to the occurrence of clinical features, activity of CF, and diet composition in paediatric patients. Patients and Method. Serum concentrations of D-lactate were measured in 38 CF patients (19 girls and 19 boys) from 6 months to 18 years of age. The analysis included age, sex, clinical symptoms, diet (the variety and calorie needs), the laboratory tests for pancreatic efficiency (serum levels of albumin and glucose, faecal elastase activity, and faecal fat index) and faecal calprotectin (the marker of intestinal inflammation), and parameters of liver damage and of cholestasis (the activity of aminotransferases, γ-glutamyltransferase, level of bilirubin, and international normalized ratio). Results. The median level of D-lactate was 0.86 μg/ml (1Q–3Q: 0.48–2.03) and correlated with the CF severity in the Schwachman-Kulczycki score, parameters of pancreatic insufficiency, and the presence of intestinal inflammation. An increased level of D-lactate was observed in the subgroup with pancreas insufficiency (1.05 versus 0.73; p<0.05), parallel with an elevated level of calprotectin (0.948 versus 0.755; p=0.08). There was no relationship between energy consumption and diet composition and serum D-lactates. Conclusion. Serum D-lactate concentration in CF patients is a promising new marker of exocrine pancreatic insufficiency probably related to intestinal flora dysbiosis/overgrowth.

scholarly journals Preoperative prediction of complicated appendicitis using machine learning method

2020 ◽  
Author(s):  
Chunbo Kang ◽  
Xubin Li ◽  
Xiaoqian Chi ◽  
Yabin Yang ◽  
Haifeng Shan ◽  
...  
Keyword(s):  
Machine Learning ◽  
Clinical Symptoms ◽  
Laboratory Data ◽  
Learning Model ◽  
Complicated Appendicitis ◽  
Support Vector ◽  
Training Set ◽  
Preoperative Prediction ◽  
Machine Learning Model ◽  
Testing Set

Abstract BACKGROUND Accurate preoperative prediction of complicated appendicitis (CA) could help selecting optimal treatment and reducing risks of postoperative complications. The study aimed to develop a machine learning model based on clinical symptoms and laboratory data for preoperatively predicting CA.METHODS 136 patients with clinicopathological diagnosis of acute appendicitis were retrospectively included in the study. The dataset was randomly divided (94: 42) into training and testing set. Predictive models using individual and combined selected clinical and laboratory data features were built separately. Three combined models were constructed using logistic regression (LR), support vector machine (SVM) and random forest (RF) algorithms. The CA prediction performance was evaluated with Receiver Operating Characteristic (ROC) analysis, using the area under the curve (AUC), sensitivity, specificity and accuracy factors.RESULTS The features of the abdominal pain time, nausea and vomiting, the highest temperature, high sensitivity-CRP (hs-CRP) and procalcitonin (PCT) had significant differences in the CA prediction (P<0.001). The ability to predict CA by individual feature was low (AUC<0.8). The prediction by combined features was significantly improved. The AUC of the three models (LR, SVM and RF) in the training set and the testing set were 0.805, 0.888, 0.908 and 0.794, 0.895, 0.761, respectively. The SVM-based model showed a better performance for CA prediction. RF had a higher AUC in the training set, but its poor efficiency in the testing set indicated a poor generalization ability.CONCLUSIONS The SVM machine learning model applying clinical and laboratory data can well predict CA preoperatively which could assist diagnosis in resource limited settings.

scholarly journals Comparative study of neuroimaging features and clinical symptoms in patients with eclampsia

2021 ◽  
Vol 10 (11) ◽  
pp. 4171
Author(s):  
Neha Meena ◽  
Suman Meena ◽  
Khushbu Meena ◽  
Savitri Verma
Keyword(s):  
Ct Scan ◽  
Maternal Morbidity ◽  
Clinical Symptoms ◽  
Occipital Lobe ◽  
Study Data ◽  
Morbidity And Mortality ◽  
Computed Tomographic ◽  
High Risk Factors ◽  
Maternal Morbidity And Mortality ◽  
Life Threatening

Background: Eclampsia is a life-threatening emergency that remains a major cause for feto-maternal morbidity and mortality. The purpose of our study was to access various computed tomographic scan (CT) findings in eclampsia patients and compare neurological symptoms with radiological findings. Methods: A prospective analytical study was undertaken in department of obstetrics and gynecology, J. K. Lon hospital, Kota during the period of 2019-2020. Women who presented as eclampsia and admitted in indoor wards were included in the study. Data analyzed included various maternal and fetal parameters, CT scan findings and outcome of pregnancy. Results: The incidence of eclampsia was 1.1% of total deliveries. High risk factors associated with eclampsia were primigravida (70%), maternal age (70% in 21-25 years age group), illiteracy (64%), inadequate antenatal care (96%), early gestation (68%). On CT scan findings 52% patients had abnormal CT scan findings of which most common was cerebral edema (57.7%). Parieto-occipital lobe was most common region to be affected. Altered sensorium was found to be significantly associated with abnormal CT scan findings.Conclusions: Eclampsia is a major cause of fetal and maternal morbidity and mortality.  CT scan in eclampsia have significant role in early diagnosis of patients with cerebral pathologies and these CT scan findings were associated with the level of consciousness and number of convulsive episodes.  Thus, CT scan helps in further management of these patients by multidisciplinary approach.

scholarly journals A Strategy for Automatic Performance Tuning of Stencil Computations on GPUs

2018 ◽  
Vol 2018 ◽  
pp. 1-24 ◽  
Author(s):  
Joseph D. Garvey ◽  
Tarek S. Abdelrahman
Keyword(s):  
Machine Learning ◽  
Random Sampling ◽  
Graphics Processing Units ◽  
Performance Tuning ◽  
Stencil Computations ◽  
High Performing ◽  
Expert Search ◽  
Machine Learning Model ◽  
Novel Strategy ◽  
Graphics Processing

We propose and evaluate a novel strategy for tuning the performance of a class of stencil computations on Graphics Processing Units. The strategy uses a machine learning model to predict the optimal way to load data from memory followed by a heuristic that divides other optimizations into groups and exhaustively explores one group at a time. We use a set of 104 synthetic OpenCL stencil benchmarks that are representative of many real stencil computations. We first demonstrate the need for auto-tuning by showing that the optimization space is sufficiently complex that simple approaches to determining a high-performing configuration fail. We then demonstrate the effectiveness of our approach on NVIDIA and AMD GPUs. Relative to a random sampling of the space, we find configurations that are 12%/32% faster on the NVIDIA/AMD platform in 71% and 4% less time, respectively. Relative to an expert search, we achieve 5% and 9% better performance on the two platforms in 89% and 76% less time. We also evaluate our strategy for different stencil computational intensities, varying array sizes and shapes, and in combination with expert search.

Sign in / Sign up

Export Citation Format

Share Document