Prediction of Hypotension Events with Physiologic Vital Sign Signatures in The Intensive Care Unit

2020 ◽  
Author(s):  
Joo Heung Yoon ◽  
Vincent Jeanselme ◽  
Artur Dubrawski ◽  
Marilyn Hravnak ◽  
Michael R. Pinsky ◽  
...  
Keyword(s):  
Intensive Care Unit ◽  
Intensive Care ◽  
Risk Score ◽  
Medical Information ◽  
Characteristic Curve ◽  
Real Life ◽  
Vital Signs ◽  
Control Group ◽  
Alert System ◽  
And Performance

Abstract Background. Even brief hypotension is associated with increased morbidity and mortality. We developed a machine learning model to predict the initial hypotension event among intensive care unit (ICU) patients, and designed an alert system for bedside implementation. Materials and Methods. From the Medical Information Mart for Intensive Care III (MIMIC-3) dataset minute-by-minute vital signs were extracted. A hypotension event was defined as at least 5 measurements within a 10-minute period of systolic blood pressure ≤ 90 mmHg and mean arterial pressure ≤ 60 mmHg. A random forest (RF) classifier was used to predict hypotension, and performance was measured with area under the receiver operating characteristic curve (AUROC) and area under the precision recall curve (AUPRC). Hypotension alerts were generated using risk score thresholds, then a stacked RF model and a lock-out time were applied for real-life implementation. Results. We identified 1307 subjects (1580 ICU stays) as the case (hypotension) group and 1619 subjects (2279 ICU stays) as the control group. The RF model showed AUROC of 0.93 and 0.88 at 15 and 60 minutes respectively before hypotension, and AUPRC of 0.77 at 60 minutes before. Risk score trajectories revealed 80% and > 60% of cases predicted at 15 and 60 minutes before the hypotension, respectively. The stacked model with 15-minute lock-out produced on average 0.79 alerts/subject/hour (sensitivity 92.4%). Conclusion. Clinically significant hypotension events in the ICU can be predicted at least 1 hour before the initial hypotension episode. Developing a high-sensitive and reliable practical alert system is feasible, with low rate of alerts.

scholarly journals Prediction of hypotension events with physiologic vital sign signatures in the intensive care unit

Critical Care ◽  
2020 ◽  
Vol 24 (1) ◽  
Author(s):  
Joo Heung Yoon ◽  
Vincent Jeanselme ◽  
Artur Dubrawski ◽  
Marilyn Hravnak ◽  
Michael R. Pinsky ◽  
...  
Keyword(s):  
Intensive Care Unit ◽  
Time Series ◽  
Intensive Care ◽  
Risk Score ◽  
Medical Information ◽  
Time Series Data ◽  
Characteristic Curve ◽  
Real Life ◽  
Series Data ◽  
Alert System

Abstract Background Even brief hypotension is associated with increased morbidity and mortality. We developed a machine learning model to predict the initial hypotension event among intensive care unit (ICU) patients and designed an alert system for bedside implementation. Materials and methods From the Medical Information Mart for Intensive Care III (MIMIC-3) dataset, minute-by-minute vital signs were extracted. A hypotension event was defined as at least five measurements within a 10-min period of systolic blood pressure ≤ 90 mmHg and mean arterial pressure ≤ 60 mmHg. Using time series data from 30-min overlapping time windows, a random forest (RF) classifier was used to predict risk of hypotension every minute. Chronologically, the first half of extracted data was used to train the model, and the second half was used to validate the trained model. The model’s performance was measured with area under the receiver operating characteristic curve (AUROC) and area under the precision recall curve (AUPRC). Hypotension alerts were generated using risk score time series, a stacked RF model. A lockout time were applied for real-life implementation. Results We identified 1307 subjects (1580 ICU stays) as the hypotension group and 1619 subjects (2279 ICU stays) as the non-hypotension group. The RF model showed AUROC of 0.93 and 0.88 at 15 and 60 min, respectively, before hypotension, and AUPRC of 0.77 at 60 min before. Risk score trajectories revealed 80% and > 60% of hypotension predicted at 15 and 60 min before the hypotension, respectively. The stacked model with 15-min lockout produced on average 0.79 alerts/subject/hour (sensitivity 92.4%). Conclusion Clinically significant hypotension events in the ICU can be predicted at least 1 h before the initial hypotension episode. With a highly sensitive and reliable practical alert system, a vast majority of future hypotension could be captured, suggesting potential real-life utility.

scholarly journals Application of an Automatic Medical Information System to Implement Bundle Care for the Prevention of Ventilator-Associated Pneumonia

2021 ◽  
Vol 18 (21) ◽  
pp. 11128
Author(s):  
Hung-Hui Lee ◽  
Li-Ying Lin ◽  
Hsiu-Fen Yang ◽  
Yu-Yi Tang ◽  
Pei-Hern Wang
Keyword(s):  
Intensive Care Unit ◽  
Information System ◽  
Intensive Care ◽  
Oral Hygiene ◽  
Medical Information ◽  
Medical Information System ◽  
Care Bundle ◽  
Control Group ◽  
Ventilator Associated Pneumonia ◽  
Care Bundles

Ventilator-associated pneumonia is a common hospital-acquired infection. It causes patients to stay longer in the hospital and increases medical costs. This study explores the effect of applying an automatic medical information system to implement five-item prevention care bundles on the prevention of ventilator-related pneumonia. This study was a retrospective cohort study. This study was conducted from October 2017 to February 2018 and collected data from the intensive care unit of a medical center in southern Taiwan from January 2013 to May 2016. The control group (enrolled from January 2013 to June 2014) received oral hygiene. The experimental group (enrolled from July 2014 to December 2015) received five-item ventilator-associated pneumonia prevention care bundles, which consisted of (1) elevation of the head of the bed to 30–45°; (2) daily oral care with 0.12−0.2% chlorhexidine twice daily; (3) daily assessment of readiness to extubate; (4) daily sedative interruption; and (5) emptying water from the respirator tube. Results showed the incidence of ventilator-associated pneumonia in the bundle group was significantly less than the oral hygiene group (p = 0.029). The factors that significantly affected the incidence of ventilator-associated pneumonia were ventilator-associated pneumonia care bundle, ventilator-days, and intensive care unit length of stay. A significant reduction in ventilator-associated pneumonia rate in the bundle group compared to the oral hygiene group (OR = 0.366, 95% CI = 0.159–0.840) was observed, with 63.4% effectiveness. Application of an automatic medical information system to implement bundle care can significantly reduce the incidence of ventilator-associated pneumonia.

scholarly journals A Long Short-Term Memory Ensemble Approach for Improving the Outcome Prediction in Intensive Care Unit

2019 ◽  
Vol 2019 ◽  
pp. 1-10 ◽  
Author(s):  
Jing Xia ◽  
Su Pan ◽  
Min Zhu ◽  
Guolong Cai ◽  
Molei Yan ◽  
...  
Keyword(s):  
Intensive Care Unit ◽  
Intensive Care ◽  
Medical Information ◽  
Short Term Memory ◽  
Characteristic Curve ◽  
Scoring Systems ◽  
Superior Performance ◽  
Short Term ◽  
Term Memory ◽  
Long Short Term Memory

In intensive care unit (ICU), it is essential to predict the mortality of patients and mathematical models aid in improving the prognosis accuracy. Recently, recurrent neural network (RNN), especially long short-term memory (LSTM) network, showed advantages in sequential modeling and was promising for clinical prediction. However, ICU data are highly complex due to the diverse patterns of diseases; therefore, instead of single LSTM model, an ensemble algorithm of LSTM (eLSTM) is proposed, utilizing the superiority of the ensemble framework to handle the diversity of clinical data. The eLSTM algorithm was evaluated by the acknowledged database of ICU admissions Medical Information Mart for Intensive Care III (MIMIC-III). The investigation in total of 18415 cases shows that compared with clinical scoring systems SAPS II, SOFA, and APACHE II, random forests classification algorithm, and the single LSTM classifier, the eLSTM model achieved the superior performance with the largest value of area under the receiver operating characteristic curve (AUROC) of 0.8451 and the largest area under the precision-recall curve (AUPRC) of 0.4862. Furthermore, it offered an early prognosis of ICU patients. The results demonstrate that the eLSTM is capable of dynamically predicting the mortality of patients in complex clinical situations.

scholarly journals The Effect of Music Therapy on Perceived Pain, Mental Health, Vital Signs, and Medication Usage of Burn Patients Hospitalized in the Intensive Care Unit: A Randomized Controlled Feasibility Study Protocol

2021 ◽  
Vol 12 ◽  
Author(s):  
Mark Ettenberger ◽  
Rafael Maya ◽  
Andrés Salgado-Vasco ◽  
Sofia Monsalve-Duarte ◽  
William Betancourt-Zapata ◽  
...  
Keyword(s):  
Intensive Care Unit ◽  
Clinical Trials ◽  
Intensive Care ◽  
Music Therapy ◽  
Study Protocol ◽  
Intervention Group ◽  
Vital Signs ◽  
Control Group ◽  
Burn Patients ◽  
Randomized Controlled

Background: Burn patients experience major physiological and psychological stressors during treatment and rehabilitation, including elevated levels of pain, anxiety, stress, or depression. Music interventions inclusive of music therapy (MT) have been shown to improve such symptoms, but rigorous clinical trials investigating specific music therapy methods in adult burn patients are scarce.Methods: This is a single center Randomized Controlled Trial (RCT) protocol with two parallel arms. Participants are 81 adult burn patients admitted to the Intensive Care Unit (ICU) of the University Hospital Fundación Santa Fe de Bogotá in Colombia. The intervention consists of a Music Assisted Relaxation (MAR) protocol, a music therapy technique composed of entrained live music combined with a guided relaxation and/or the use of imagery. The effects of the MAR will be compared to a control group (treatment as usual) over a period of maximum 2 weeks or six interventions. The primary outcome measure is perceived background pain, as measured with a Visual Analog Scale (VAS) before and after each intervention. Secondary outcomes are anxiety and depression levels; vital signs; and the use of pain medication. Additionally, some patients in the intervention group will be invited to participate in electroencephalography, electromyography, and electrocardiography recordings during the MAR.Discussion: This study protocol follows the SPIRIT guidelines for defining items of clinical trials and is the first study in Colombia to evaluate the effects of music therapy for adult burn patients. With this RCT it is hoped to gather new knowledge about the potential of music therapy to help critical care patients cope and recover from their injuries during the hospitalization in the ICU.Trial registration:www.clinicaltrials.gov, Identifier: NCT04571255.Protocol version: V1.0, May 24th 2021

scholarly journals Development and Validation of an Interpretable 3-day Intensive Care Unit Readmission Prediction Model Using Explainable Boosting Machines

2021 ◽  
Author(s):  
Stefan Hegselmann ◽  
Christian Ertmer ◽  
Thomas Volkert ◽  
Antje Gottschalk ◽  
Martin Dugas ◽  
...  
Keyword(s):  
Machine Learning ◽  
Intensive Care Unit ◽  
Intensive Care ◽  
Medical Information ◽  
Characteristic Curve ◽  
External Validation ◽  
Simplified Acute Physiology Score ◽  
Gradient Boosting ◽  
Patients At Risk ◽  
High Level

Intensive care unit readmissions are associated with mortality and bad outcomes. Machine learning could help to identify patients at risk to improve discharge decisions. However, many models are black boxes, so that dangerous properties might remain unnoticed. In this study, an inherently interpretable model for 3-day ICU readmission prediction was developed. We used a retrospective cohort of 15,589 ICU stays and 169 variables collected between 2006 and 2019. A team of doctors inspected the model, checked the plausibility of each component, and removed problematic parts. Qualitative feedback revealed several challenges for interpretable machine learning in healthcare. The resulting model used 67 features and showed an area under the precision-recall curve of 0.119+/-0.020 and an area under the receiver operating characteristic curve of 0.680+/-0.025. This is on par with state-of-the-art gradient boosting machines and outperforms the Simplified Acute Physiology Score II. External validation with the Medical Information Mart for Intensive Care database version IV confirmed our findings. Hence, a machine learning model for readmission prediction with a high level of human control is feasible without sacrificing performance.

scholarly journals Effects of COVID-19 pandemic on breastfeeding rates in a neonatal intensive care unit

2021 ◽  
Vol 0 (0) ◽  
Author(s):  
Asli Okbay Gunes ◽  
Emre Dincer ◽  
Nilgun Karadag ◽  
Sevilay Topcuoglu ◽  
Guner Karatekin
Keyword(s):  
Intensive Care Unit ◽  
Intensive Care ◽  
Breast Milk ◽  
Neonatal Intensive Care Unit ◽  
Neonatal Intensive Care ◽  
Control Group ◽  
Study Group ◽  
Intake Rate ◽  
Informed Consent Form ◽  
Group 2

Abstract Objectives To find out if the expressed breast milk delivery rate to neonatal intensive care unit (NICU) for babies who were hospitalized for any reason other than COVID-19, and exclusive breastfeeding (EB) rates between discharge date and 30th day of life of those babies were affected by COVID-19 pandemic. Methods Babies who were hospitalized before the date first coronavirus case was detected in our country were included as control group (CG). The study group was divided into two groups; study group 1 (SG1): the mothers whose babies were hospitalized in the period when mother were asked not to bring breast milk to NICU, study group 2 (SG2): the mothers whose babies were hospitalized after the date we started to use the informed consent form for feeding options. The breast milk delivery rates to NICU during hospitalization and EB rates between discharge and 30th day of life were compared between groups. Results Among 154 mother-baby dyads (CG, n=50; SG1, n=46; SG2, n=58), the percentage of breast milk delivery to NICU was 100%, 79% for CG, SG2, respectively (p<0.001). The EB rate between discharge and 30th day of life did not change between groups (CG:90%, SG1:89%, SG2:75.9; p=0.075). Conclusions If the mothers are informed about the importance of breast milk, the EB rates are not affected by the COVID-19 pandemic in short term, even if the mothers are obligatorily separated from their babies. The breast milk intake rate of the babies was lowest while our NICU protocol was uncertain, and after we prepared a protocol this rate increased.

Topical Lidocaine Patch for Postthoracotomy and Poststernotomy Pain in Cardiothoracic Intensive Care Unit Adult Patients

2019 ◽  
Vol 39 (5) ◽  
pp. 51-57 ◽  
Author(s):  
Michael Liu ◽  
Mabel Wai ◽  
James Nunez
Keyword(s):  
Intensive Care Unit ◽  
Cardiac Surgery ◽  
Intensive Care ◽  
Standard Care ◽  
Academic Medical Center ◽  
Adult Patients ◽  
Control Group ◽  
Lidocaine Group ◽  
Lengths Of Stay ◽  
Topical Lidocaine

Background Transdermal lidocaine patches have few systemic toxicities and may be useful analgesics in cardiac surgery patients. However, few studies have evaluated their efficacy in the perioperative setting. Objective To compare the efficacy of topical lidocaine 5% patch plus standard care (opioid and nonopioid analgesics) with standard care alone for postthoracotomy or poststernotomy pain in adult patients in a cardiothoracic intensive care unit. Methods A single-center, retrospective cohort evaluation was conducted from January 2015 through December 2015 in the adult cardiothoracic intensive care unit at a tertiary academic medical center. Cardiac surgery patients with new sternotomies or thoracotomies were included. Patients in the lidocaine group received 1 to 3 topical lidocaine 5% patches near sternotomy and/or thoracotomy sites daily. Patches remained in place for 12 hours daily. Patients in the control group received standard care alone. Results The primary outcome was numeric pain rating for sternotomy/thoracotomy sites. Secondary outcomes were cardiothoracic intensive care unit and hospital lengths of stay and total doses of analgesics received. Forty-seven patients were included in the lidocaine group; 44 were included in the control group. Mean visual analogue scores for pain did not differ between groups (lidocaine, 2; control, 1.9; P = .58). Lengths of stay were similar for both groups (cardiothoracic intensive care unit: lidocaine, 3.06 days; control, 3.11 days; P = .86; hospital: lidocaine, 8.26 days; control, 7.61 days; P = .47). Conclusions Adjunctive lidocaine 5% patches did not reduce acute pain in postthoracotomy and post-sternotomy patients in the cardiothoracic intensive care unit.

scholarly journals The use of score for neonatal acute physiology perinatal extention II (SNAPPE II) in predicting neonatal outcome in neonatal intensive care unit

2016 ◽  
Vol 45 (6) ◽  
pp. 241
Author(s):  
Mia R A ◽  
Risa Etika ◽  
Agus Harianto ◽  
Fatimah Indarso ◽  
Sylviati M Damanik
Keyword(s):  
Intensive Care Unit ◽  
Intensive Care ◽  
Neonatal Intensive Care Unit ◽  
Neonatal Intensive Care ◽  
Characteristic Curve ◽  
Newborn Care ◽  
Scoring Systems ◽  
Laboratory Observation ◽  
The Mean ◽  
Overall Mortality

Background Scoring systems which quantify initial risks have animportant role in aiding execution of optimum health services by pre-dicting morbidity and mortality. One of these is the score for neonatalacute physiology perinatal extention (SNAPPE), developed byRichardson in 1993 and simplified in 2001. It is derived of 6 variablesfrom the physical and laboratory observation within the first 12 hoursof admission, and 3 variables of perinatal risks of mortality.Objectives To assess the validity of SNAPPE II in predicting mor-tality at neonatal intensive care unit (NICU), Soetomo Hospital,Surabaya. The study was also undertaken to evolve the best cut-offscore for predicting mortality.Methods Eighty newborns were admitted during a four-month periodand were evaluated with the investigations as required for the specifi-cations of SNAPPE II. Neonates admitted >48 hours of age or afterhaving been discharged, who were moved to lower newborn care <24hours and those who were discharged on request were excluded. Re-ceiver operating characteristic curve (ROC) were constructed to derivethe best cut-off score with Kappa and McNemar Test.Results Twenty eight (35%) neonates died during the study, 22(82%) of them died within the first six days. The mean SNAPPE IIscore was 26.3+19.84 (range 0-81). SNAPPE II score of thenonsurvivors was significantly higher than the survivors(42.75+18.59 vs 17.4+14.05; P=0.0001). SNAPPE II had a goodperformance in predicting overall mortality and the first-6-daysmortality, with area under the ROC 0.863 and 0.889. The best cut-off score for predicting mortality was 30 with sensitivity 81.8%,specificity 76.9%, positive predictive value 60.0% and negativepredictive value 90.0%.Conclusions SNAPPE II is a measurement of illness severity whichcorrelates well with neonatal mortality at NICU, Soetomo Hospital.The score of more than 30 is associated with higher mortality

scholarly journals INVESTIGATING EFFECT OF NURSING INTERVENTIONS, BASED ON WELLS SCORE RESULTS, ON THE INCIDENCE OF DEEP VEIN THROMBOSIS IN PATIENTS ADMITTED TO THE INTENSIVE CARE UNIT

2018 ◽  
Vol 11 (5) ◽  
pp. 377
Author(s):  
Morteza Habibi Moghadam ◽  
Marzieh Asadizaker ◽  
Simin Jahani ◽  
Elham Maraghi ◽  
Hakimeh Saadatifar ◽  
...  
Keyword(s):  
Intensive Care Unit ◽  
Intensive Care ◽  
Deep Vein Thrombosis ◽  
Intervention Group ◽  
Control Group ◽  
Nursing Interventions ◽  
Vein Thrombosis ◽  
Wells Score ◽  
Deep Vein ◽  
Pitting Edema

 Objective: Venous thromboembolism, including deep vein thrombosis (DVT) and pulmonary embolism (PE), is a common complaint in critically ill patients. Therefore, the present study was conducted to determine the effect of nursing interventions, based on the Wells results, on the incidence of DVT in intensive care unit (ICU) patients.Methods: The present clinical trial was conducted on 72 ICU patients without DVT and PE who met the inclusion criteria according to Wells score in Dr. Ganjavian Hospital, Dezful in 2012. The participants were investigated and randomly divided into intervention (n=36) and control groups (n=36). The intervention group received preventive nursing measures based on the risk level determined by the Wells score, and routine therapeutic interventions were performed for the control group. Then, patients were evaluated using Wells score, D-dimer testing, and Doppler sonography on the 1st, 5th, and 10th days. Data were finally coded and entered into SPSS version 23. Data analysis was performed using Chi-square, Fisher’s exact, and Mann–Whitney U tests.Results: The incidence of DVT in both groups showed that 2 patients of the control group who were identified to be at risk using the Wells score were diagnosed with DVT while none of the patients of the intervention group experienced DVT. The present study showed that 22.2% of the patients of the control group suffered from non-pitting edema, which was significantly different from the intervention group (p=0.005).Conclusion: The results of the present study showed that using the Wells score for early identification of the at-risk patients and nursing interventions based on this score’s results is helpful in the prevention of DVT. Appropriate nursing interventions were also effective in reducing the incidence of non-pitting edema in the lower extremities.

scholarly journals Perception of Stress and Styles of Coping with It in Parents Giving Kangaroo Mother Care to Their Children during Hospitalization in NICU

2021 ◽  
Vol 18 (23) ◽  
pp. 12694
Author(s):  
Barbara Zych ◽  
Witold Błaż ◽  
Ewa Dmoch-Gajzlerska ◽  
Katarzyna Kanadys ◽  
Anna Lewandowska ◽  
...  
Keyword(s):  
Intensive Care Unit ◽  
Intensive Care ◽  
Neonatal Intensive Care Unit ◽  
Parental Stress ◽  
Neonatal Intensive Care ◽  
Premature Baby ◽  
Hospital Environment ◽  
Control Group ◽  
Skin Contact ◽  
Task Oriented

The experience of hospitalization of a newborn in the Neonatal Intensive Care Unit (NICU) may become distressing both for the baby and parent. The study aimed to assess the degree of parental stress and coping strategies in parents giving KMC to their babies hospitalized in NICU compared to the control group parents not giving KMC. The prospective observational study enrolled a cohort of 337 parents of premature babies hospitalized in NICU in 2016 in Eastern Poland. The Parental Stressor Scale: Neonatal Intensive Care Unit, Coping Inventory for Stressful Situations were used. The level of stress in parents giving KMC was defined as low or moderate. Analysis confirmed its greater presence in the group of parents initiating KMC late (2–3 weeks) compared to those starting this initiative in week 1 of a child’s life. An additional predictor of a higher level of stress in parents initiating KMC “late” was the hospital environment of a premature baby. Task oriented coping was the most common coping strategy in the study group. KMC and direct skin-to-skin contact of the parent with the baby was associated with a higher level of parental stress only initially and decreased with time and KMC frequency.

Sign in / Sign up

Export Citation Format

Share Document