scholarly journals Application of Deep Learning Technology in Predicting the Risk of Inpatient Death in Intensive Care Unit

2021 ◽  
Vol 2021 ◽  
pp. 1-9
Author(s):  
Ming Li ◽  
HuiLin Chen ◽  
ShuYing Yan ◽  
Xiao Xu ◽  
HuaJuan Xu
Keyword(s):  
Neural Network ◽  
Intensive Care Unit ◽  
Intensive Care ◽  
Deep Learning ◽  
Clinical Application ◽  
Experimental Results ◽  
Fuzzy Artmap ◽  
Icu Patients ◽  
Life And Death ◽  
Clinical Scoring

The Intensive Care Unit (ICU) is an important unit for the rescue of critically ill patients in hospitals, and patient mortality is an important indicator to measure the level of ICU treatment. Currently, a variety of clinical scoring systems are used to evaluate the patient's condition and predict survival, but these systems require a lot of resources. However, due to the rapid development of artificial intelligence and deep learning, machine learning based methods have been used to study the survival prediction of ICU patients. Additionally, these methods have made significant progress, but there is still a distance from clinical application, and equally metric interpretability of the deep learning method is not very mature. Therefore, in this paper, we have proposed a predicting model for the life and death of ICU patients, which is primarily based on the Fuzzy ARTMAP model. With a thorough analysis of the existing ICU patient condition assessment and life and death prediction methods, we have observed that patient’s ICU monitoring information performs integrated analysis and extracts features according to the clinical characteristics of physiological indicators. Finally, fuzzy ARTMAP neural network is used to predict the life and death of patients. Likewise, prediction results are combined with the clinical scoring system and logistic regression, artificial neural network, support vector machine, and AdaBoost. Experimental results of these algorithms were compared, which verifies that the proposed method has outperformed the existing model. The main purpose of the proposed mode is to design a life and death prediction method for ICU patients, which has high predictive performance and is an acceptable method for clinical medical staff, where ICU monitoring data is used. Experimental results show that the method proposed has achieved better prediction performance and accuracy ratio, which provide theoretical reference for clinical application.

scholarly journals Deep learning to predict long-term mortality in patients requiring 7 days of mechanical ventilation

PLoS ONE ◽  
2021 ◽  
Vol 16 (6) ◽  
pp. e0253443
Author(s):  
Naomi George ◽  
Edward Moseley ◽  
Rene Eber ◽  
Jennifer Siu ◽  
Mathew Samuel ◽  
...  
Keyword(s):  
Neural Network ◽  
Intensive Care Unit ◽  
Mechanical Ventilation ◽  
Intensive Care ◽  
Deep Learning ◽  
Medical Information ◽  
Prediction Models ◽  
Prolonged Mechanical Ventilation ◽  
External Validation ◽  
Saps Ii Score

Background Among patients with acute respiratory failure requiring prolonged mechanical ventilation, tracheostomies are typically placed after approximately 7 to 10 days. Yet half of patients admitted to the intensive care unit receiving tracheostomy will die within a year, often within three months. Existing mortality prediction models for prolonged mechanical ventilation, such as the ProVent Score, have poor sensitivity and are not applied until after 14 days of mechanical ventilation. We developed a model to predict 3-month mortality in patients requiring more than 7 days of mechanical ventilation using deep learning techniques and compared this to existing mortality models. Methods Retrospective cohort study. Setting: The Medical Information Mart for Intensive Care III Database. Patients: All adults requiring ≥ 7 days of mechanical ventilation. Measurements: A neural network model for 3-month mortality was created using process-of-care variables, including demographic, physiologic and clinical data. The area under the receiver operator curve (AUROC) was compared to the ProVent model at predicting 3 and 12-month mortality. Shapley values were used to identify the variables with the greatest contributions to the model. Results There were 4,334 encounters divided into a development cohort (n = 3467) and a testing cohort (n = 867). The final deep learning model included 250 variables and had an AUROC of 0.74 for predicting 3-month mortality at day 7 of mechanical ventilation versus 0.59 for the ProVent model. Older age and elevated Simplified Acute Physiology Score II (SAPS II) Score on intensive care unit admission had the largest contribution to predicting mortality. Discussion We developed a deep learning prediction model for 3-month mortality among patients requiring ≥ 7 days of mechanical ventilation using a neural network approach utilizing readily available clinical variables. The model outperforms the ProVent model for predicting mortality among patients requiring ≥ 7 days of mechanical ventilation. This model requires external validation.

scholarly journals Lung ultrasound predicts clinical course but not outcome in COVID-19 ICU patients: a retrospective single-center analysis

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Stephanie-Susanne Stecher ◽  
Sofia Anton ◽  
Alessia Fraccaroli ◽  
Jeremias Götschke ◽  
Hans Joachim Stemmler ◽  
...  
Keyword(s):  
Intensive Care Unit ◽  
Intensive Care ◽  
Lung Ultrasound ◽  
Clinical Course ◽  
Point Of Care ◽  
Clinical Deterioration ◽  
Normal Lung ◽  
Icu Patients ◽  
Standardized Protocol ◽  
Ultrasound Score

Abstract Background Point-of-care lung ultrasound (LU) is an established tool in the first assessment of patients with coronavirus disease (COVID-19). Purpose of this study was to evaluate the value of lung ultrasound in COVID-19 intensive care unit (ICU) patients in predicting clinical course and outcome. Methods We analyzed lung ultrasound score (LUS) of all COVID-19 patients admitted from March 2020 to December 2020 to the Internal Intensive Care Unit, Ludwig-Maximilians-University (LMU) of Munich. LU was performed according to a standardized protocol at ICU admission and in case of clinical deterioration with the need for intubation. A normal lung scores 0 points, the worst LUS has 24 points. Patients were stratified in a low (0–12 points) and a high (13–24 points) lung ultrasound score group. Results The study included 42 patients, 69% of them male. The most common comorbidities were hypertension (81%) and obesity (57%). The values of pH (7.42 ± 0.09 vs 7.35 ± 0.1; p = 0.047) and paO2 (107 [80–130] vs 80 [66–93] mmHg; p = 0.034) were significantly reduced in patients of the high LUS group. Furthermore, the duration of ventilation (12.5 [8.3–25] vs 36.5 [9.8–70] days; p = 0.029) was significantly prolonged in this group. Patchy subpleural thickening (n = 38; 90.5%) and subpleural consolidations (n = 23; 54.8%) were present in most patients. Pleural effusion was rare (n = 4; 9.5%). The median total LUS was 11.9 ± 3.9 points. In case of clinical deterioration with the need for intubation, LUS worsened significantly compared to baseline LU. Twelve patients died during the ICU stay (29%). There was no difference in survival in both LUS groups (75% vs 66.7%, p = 0.559). Conclusions LU can be a useful monitoring tool to predict clinical course but not outcome of COVID-19 ICU patients and can early recognize possible deteriorations.

scholarly journals Should the CIWA-Ar be the standard monitoring strategy for alcohol withdrawal syndrome in the intensive care unit?

2021 ◽  
Vol 16 (1) ◽  
Author(s):  
Tessa L. Steel ◽  
Shewit P. Giovanni ◽  
Sarah C. Katsandres ◽  
Shawn M. Cohen ◽  
Kevin B. Stephenson ◽  
...  
Keyword(s):  
Intensive Care Unit ◽  
Intensive Care ◽  
Alcohol Withdrawal ◽  
Withdrawal Syndrome ◽  
Alcohol Withdrawal Syndrome ◽  
Patient Characteristics ◽  
Response To Treatment ◽  
Alternative Measure ◽  
Icu Patients ◽  
Verbal Responses

Abstract Background The Clinical Institute Withdrawal Assessment for Alcohol-Revised (CIWA-Ar) is commonly used in hospitals to titrate medications for alcohol withdrawal syndrome (AWS), but may be difficult to apply to intensive care unit (ICU) patients who are too sick or otherwise unable to communicate. Objectives To evaluate the frequency of CIWA-Ar monitoring among ICU patients with AWS and variation in CIWA-Ar monitoring across patient demographic and clinical characteristics. Methods The study included all adults admitted to an ICU in 2017 after treatment for AWS in the Emergency Department of an academic hospital that standardly uses the CIWA-Ar to assess AWS severity and response to treatment. Demographic and clinical data, including Richmond Agitation-Sedation Scale (RASS) assessments (an alternative measure of agitation/sedation), were obtained via chart review. Associations between patient characteristics and CIWA-Ar monitoring were tested using logistic regression. Results After treatment for AWS, only 56% (n = 54/97) of ICU patients were evaluated using the CIWA-Ar; 94% of patients had a documented RASS assessment (n = 91/97). Patients were significantly less likely to receive CIWA-Ar monitoring if they were intubated or identified as Black. Conclusions CIWA-Ar monitoring was used inconsistently in ICU patients with AWS and completed less often in those who were intubated or identified as Black. These hypothesis-generating findings raise questions about the utility of the CIWA-Ar in ICU settings. Future studies should assess alternative measures for titrating AWS medications in the ICU that do not require verbal responses from patients and further explore the association of race with AWS monitoring.

scholarly journals Beta-Lactams Toxicity in the Intensive Care Unit: An Underestimated Collateral Damage?

2021 ◽  
Vol 9 (7) ◽  
pp. 1505
Author(s):  
Claire Roger ◽  
Benjamin Louart
Keyword(s):  
Risk Factors ◽  
Intensive Care Unit ◽  
Intensive Care ◽  
Clinical Manifestations ◽  
Acute Interstitial Nephritis ◽  
Convulsive Status Epilepticus ◽  
Beta Lactam ◽  
Beta Lactams ◽  
Drug Induced ◽  
Icu Patients

Beta-lactams are the most commonly prescribed antimicrobials in intensive care unit (ICU) settings and remain one of the safest antimicrobials prescribed. However, the misdiagnosis of beta-lactam-related adverse events may alter ICU patient management and impact clinical outcomes. To describe the clinical manifestations, risk factors and beta-lactam-induced neurological and renal adverse effects in the ICU setting, we performed a comprehensive literature review via an electronic search on PubMed up to April 2021 to provide updated clinical data. Beta-lactam neurotoxicity occurs in 10–15% of ICU patients and may be responsible for a large panel of clinical manifestations, ranging from confusion, encephalopathy and hallucinations to myoclonus, convulsions and non-convulsive status epilepticus. Renal impairment, underlying brain abnormalities and advanced age have been recognized as the main risk factors for neurotoxicity. In ICU patients, trough concentrations above 22 mg/L for cefepime, 64 mg/L for meropenem, 125 mg/L for flucloxacillin and 360 mg/L for piperacillin (used without tazobactam) are associated with neurotoxicity in 50% of patients. Even though renal complications (especially severe complications, such as acute interstitial nephritis, renal damage associated with drug induced hemolytic anemia and renal obstruction by crystallization) remain rare, there is compelling evidence of increased nephrotoxicity using well-known nephrotoxic drugs such as vancomycin combined with beta-lactams. Treatment mainly relies on the discontinuation of the offending drug but in the near future, antimicrobial optimal dosing regimens should be defined, not only based on pharmacokinetics/pharmacodynamic (PK/PD) targets associated with clinical and microbiological efficacy, but also on PK/toxicodynamic targets. The use of dosing software may help to achieve these goals.

scholarly journals Contamination of Propofol Infusions in the Intensive Care Unit: Incidence and Clinical Significance

1998 ◽  
Vol 26 (2) ◽  
pp. 162-164 ◽  
Author(s):  
S. A. R. Webb ◽  
B. Roberts ◽  
F. X. Breheny ◽  
C. L. Golledge ◽  
P. D. Cameron ◽  
...  
Keyword(s):  
Intensive Care Unit ◽  
Intensive Care ◽  
Observational Study ◽  
Nosocomial Infection ◽  
Wound Infection ◽  
Clinical Significance ◽  
Low Grade ◽  
Clinical Infection ◽  
Icu Patients

Epidemics of bacteraemia and wound infection have been associated with the infusion of bacterially contaminated propofol administered during anaesthesia. We conducted an observational study to determine the incidence and clinical significance of administration of potentially contaminated propofol to patients in an ICU setting. One hundred patients received a total of 302 infusions of propofol. Eighteen episodes of possible contamination of propofol syringes were identified, but in all cases contamination was by a low-grade virulence pathogen. There were no episodes of clinical infection or colonization which could be attributed to the administration of contaminated propofol. During the routine use of propofol to provide sedation in ICU patients the risk of nosocomial infection secondary to contamination of propofol is extremely low.

Comparing Vitamin A and Moist Chamber in Preventing Ocular Surface Disorders

2017 ◽  
Vol 27 (6) ◽  
pp. 714-729 ◽  
Author(s):  
Hassan Babamohamadi ◽  
Monir Nobahar ◽  
Jalaladin Razi ◽  
Raheb Ghorbani
Keyword(s):  
Clinical Trial ◽  
Intensive Care Unit ◽  
Intensive Care ◽  
Vitamin A ◽  
Ocular Surface ◽  
Slight Reduction ◽  
Moist Chamber ◽  
Icu Patients ◽  
Schirmer’S Test

The present study was conducted to determine the effectiveness of vitamin A eye ointment (VAEO) and moist chamber (MC) in preventing ocular surface disorders (OSD) in intensive care unit (ICU) patients. A total of 38 eligible patients were selected for participation in the present clinical trial. All the patients were randomly administered VAEO in one eye every 6 hr for 5 days and had a polyethylene cover (PC) placed on their other eye to create an MC that was replaced every 12 hr as well. The results of Schirmer’s test also increased by 2.06 mm in the VAEO group ( p < .001), while they showed a slight reduction by 0.15 mm in the MC group ( p = .669). VAEO was more effective in preventing OSD in ICU patients than MC and is, therefore, recommended to be used as a method of preventing OSD.

scholarly journals Comparison of Non–Intensive Care Unit (ICU) versus ICU Rates of Catheter-Associated Urinary Tract Infection in Community Hospitals

2013 ◽  
Vol 34 (7) ◽  
pp. 744-747 ◽  
Author(s):  
Sarah S. Lewis ◽  
Lauren P. Knelson ◽  
Rebekah W. Moehring ◽  
Luke F. Chen ◽  
Daniel J. Sexton ◽  
...  
Keyword(s):  
Intensive Care Unit ◽  
Urinary Tract Infection ◽  
Intensive Care ◽  
Urinary Tract ◽  
Tract Infection ◽  
Community Hospitals ◽  
Icu Patients ◽  
Important Target

We describe and compare the epidemiology of catheter-associated urinary tract infection (CAUTI) occurring in non-intensive care unit (ICU) versus ICU wards in a network of community hospitals over a 2-year period. Overall, 72% of cases of CAUTI occurred in non-ICU patients, which indicates that this population is an important target for dedicated surveillance and prevention efforts.

scholarly journals Presymptomatic Prediction of Sepsis in Intensive Care Unit Patients

2008 ◽  
Vol 15 (7) ◽  
pp. 1089-1094 ◽  
Author(s):  
R. A. Lukaszewski ◽  
A. M. Yates ◽  
M. C. Jackson ◽  
K. Swingler ◽  
J. M. Scherer ◽  
...  
Keyword(s):  
Neural Network ◽  
Intensive Care ◽  
Clinical Diagnosis ◽  
Predictive Accuracy ◽  
High Sensitivity ◽  
Icu Patients ◽  
The Neural Network ◽  
Sensitivity And Selectivity ◽  
Ccl2 Mrna ◽  
Factor Α

ABSTRACT Postoperative or posttraumatic sepsis remains one of the leading causes of morbidity and mortality in hospital populations, especially in populations in intensive care units (ICUs). Central to the successful control of sepsis-associated infections is the ability to rapidly diagnose and treat disease. The ability to identify sepsis patients before they show any symptoms would have major benefits for the health care of ICU patients. For this study, 92 ICU patients who had undergone procedures that increased the risk of developing sepsis were recruited upon admission. Blood samples were taken daily until either a clinical diagnosis of sepsis was made or until the patient was discharged from the ICU. In addition to standard clinical and laboratory parameter testing, the levels of expression of interleukin-1β (IL-1β), IL-6, IL-8, and IL-10, tumor necrosis factor-α, FasL, and CCL2 mRNA were also measured by real-time reverse transcriptase PCR. The results of the analysis of the data using a nonlinear technique (neural network analysis) demonstrated discernible differences prior to the onset of overt sepsis. Neural networks using cytokine and chemokine data were able to correctly predict patient outcomes in an average of 83.09% of patient cases between 4 and 1 days before clinical diagnosis with high sensitivity and selectivity (91.43% and 80.20%, respectively). The neural network also had a predictive accuracy of 94.55% when data from 22 healthy volunteers was analyzed in conjunction with the ICU patient data. Our observations from this pilot study indicate that it may be possible to predict the onset of sepsis in a mixed patient population by using a panel of just seven biomarkers.

scholarly journals Patterns of Antimicrobial Resistance Among Intensive Care Unit Patients of A Private Medical College Hospital in Dhaka

2015 ◽  
Vol 25 (2) ◽  
pp. 47-51 ◽  
Author(s):  
Quazi Tarikul Islam ◽  
Md Mahmudur Rahman Siddiqui ◽  
Farhana Raz ◽  
Mohammad Asrafuzzaman ◽  
Md Robed Amin
Keyword(s):  
Intensive Care Unit ◽  
Intensive Care ◽  
Tracheal Aspirate ◽  
College Hospital ◽  
Icu Patients ◽  
Gram Negative ◽  
Pseudomonas Spp ◽  
Medical College ◽  
Acinetobacter Spp ◽  
Klebsiella Spp

Because of importance of Hospital acquired infections (HAIs), it is critical to conduct surveillance studies to obtain the required data about the regional microorganisms and their susceptibility to antibiotics. This study to investigate antimicrobial resistance pattern among Intensive Care Unit (ICU) patients in a private medical college hospital setup. In a cross sectional study, 100 specimens from patients admitted in the ICU who had signs or symptoms of nosocomial infection were collected from 2012 - 2013. For each patient, samples of blood, urine, tracheal aspirate, sputum, wound swab, pus, and endotracheal tubes were obtained, cultured and analyzed with antibiogram. The most common primary diagnosis were aspiration pneumonia (49%) and UTI (20%) respectively. The most common locations for infection were tracheal aspirate (54%). The most frequent gram negative microorganisms derived from samples were Acinetobacter spp (29%), Klebsiella spp (26%) and Pseudomonas spp (18%). Klebsiella spp, Acinetobacter spp and Pseudomonas spp were most common resistant organisms among all. Klebsiella spp were resistant against Ceftriaxone (84.6%), Ceftazidime (82.6%), Amikacin (46.1%), Gentamicin (66.6%) and Quinolones (65-66.6%) respectively. Acinetobacter spp were resistant against Ceftriaxone (85%), Ceftazidime (88.8%), Cefotaxime (85.7%), Meropenem (79.3%),Amikacin (86.2%), Gentamicin (84.5%) and Quinolons (86.2-89.2%) respectively. Pseudomonas spp were resistant against Ceftriaxone (70.5%), Ceftazidime (66.6%), Amikacin (68.7%), Gentamicin (58.8%), Meropenem (52.9%) and Quinolones (81.2-86.6%) respectively. Meropenem was the most sensitive antibiotic against Klebsiella spp (84.6%) but Cotrimoxazole in case of Acinetobacter spp (60%) respectively. Escherichia coli were mostly isolated from urine, which was sensitive to Amikacin (73.3%) and Meropenem (86.6%) respectively. Gram-negative pathogens obtained from ICU patients in our settings show high resistance to antibiotics. Regular monitoring of the pattern of resistance of common pathogens in the ICUs is essential to up-to-date the use of rational antibiotics regiments.Bangladesh J Medicine Jul 2014; 25 (2) : 47-51

Sign in / Sign up

Export Citation Format

Share Document