Machine learning algorithms to predict weight gain at discharge in neonatal intensive care unit: state of the art

2021 ◽  
Vol 46 ◽  
pp. S724-S725
Author(s):  
N. Yalcin ◽  
H.T. Çelik ◽  
K. Demirkan ◽  
S. Yiğit
Keyword(s):  
Machine Learning ◽  
Intensive Care Unit ◽  
Intensive Care ◽  
Weight Gain ◽  
Neonatal Intensive Care Unit ◽  
Neonatal Intensive Care ◽  
State Of The Art ◽  
Learning Algorithms ◽  
Machine Learning Algorithms

scholarly journals Prediction of Mortality in Surgical Intensive Care Unit Patients Using Machine Learning Algorithms

2021 ◽  
Vol 8 ◽  
Author(s):  
Kyongsik Yun ◽  
Jihoon Oh ◽  
Tae Ho Hong ◽  
Eun Young Kim
Keyword(s):  
Machine Learning ◽  
Intensive Care Unit ◽  
Intensive Care ◽  
Decision Tree ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
Receiver Operating Curve ◽  
Surgical Intensive Care Unit ◽  
Hospital Death ◽  
Surgical Intensive Care

Objective: Predicting prognosis of in-hospital patients is critical. However, it is challenging to accurately predict the life and death of certain patients at certain period. To determine whether machine learning algorithms could predict in-hospital death of critically ill patients with considerable accuracy and identify factors contributing to the prediction power.Materials and Methods: Using medical data of 1,384 patients admitted to the Surgical Intensive Care Unit (SICU) of our institution, we investigated whether machine learning algorithms could predict in-hospital death using demographic, laboratory, and other disease-related variables, and compared predictions using three different algorithmic methods. The outcome measurement was the incidence of unexpected postoperative mortality which was defined as mortality without pre-existing not-for-resuscitation order that occurred within 30 days of the surgery or within the same hospital stay as the surgery.Results: Machine learning algorithms trained with 43 variables successfully classified dead and live patients with very high accuracy. Most notably, the decision tree showed the higher classification results (Area Under the Receiver Operating Curve, AUC = 0.96) than the neural network classifier (AUC = 0.80). Further analysis provided the insight that serum albumin concentration, total prenatal nutritional intake, and peak dose of dopamine drug played an important role in predicting the mortality of SICU patients.Conclusion: Our results suggest that machine learning algorithms, especially the decision tree method, can provide information on structured and explainable decision flow and accurately predict hospital mortality in SICU hospitalized patients.

Machine Learning to Support Hemodynamic Intervention in the Neonatal Intensive Care Unit

2020 ◽  
Vol 47 (3) ◽  
pp. 435-448
Author(s):  
David Van Laere ◽  
Marisse Meeus ◽  
Charlie Beirnaert ◽  
Victor Sonck ◽  
Kris Laukens ◽  
...  
Keyword(s):  
Machine Learning ◽  
Intensive Care Unit ◽  
Intensive Care ◽  
Neonatal Intensive Care Unit ◽  
Neonatal Intensive Care

Early diagnosis of bloodstream infections in the intensive care unit using machine-learning algorithms

2020 ◽  
Vol 46 (3) ◽  
pp. 454-462 ◽  
Author(s):  
Michael Roimi ◽  
Ami Neuberger ◽  
Anat Shrot ◽  
Mical Paul ◽  
Yuval Geffen ◽  
...  
Keyword(s):  
Machine Learning ◽  
Intensive Care Unit ◽  
Intensive Care ◽  
Early Diagnosis ◽  
Learning Algorithms ◽  
Bloodstream Infections ◽  
Machine Learning Algorithms

scholarly journals Longitudinal Microbiome Composition and Stability Correlate with Increased Weight and Length of Very-Low-Birth-Weight Infants

mSystems ◽  
2019 ◽  
Vol 4 (1) ◽  
Author(s):  
Alyson L. Yee ◽  
Elizabeth Miller ◽  
Larry J. Dishaw ◽  
Jessica M. Gordon ◽  
Ming Ji ◽  
...  
Keyword(s):  
Intensive Care Unit ◽  
Intensive Care ◽  
Weight Gain ◽  
Birth Weight ◽  
Neonatal Intensive Care Unit ◽  
Beta Diversity ◽  
Neonatal Intensive Care ◽  
Very Low Birth Weight ◽  
Alpha Diversity ◽  
Vlbw Infants

ABSTRACT The microbiomes of 83 preterm very-low-birth-weight (VLBW) infants and clinical covariates were analyzed weekly over the course of their initial neonatal intensive care unit (NICU) stay, with infant growth as the primary clinical outcome. Birth weight significantly correlated with increased rate of weight gain in the first 6 weeks of life, while no significant relationship was observed between rate of weight gain and feeding type. Microbial diversity increased with age and was significantly correlated with weight gain and percentage of the mother’s own milk. As expected, infants who received antibiotics during their NICU stay had significantly lower alpha diversity than those who did not. Of those in the cohort, 25 were followed into childhood. Alpha diversity significantly increased between NICU discharge and age 2 years and between age 2 years and age 4 years, but the microbial alpha diversity of 4-year-old children was not significantly different from that of mothers. Infants who showed improved length over the course of their NICU stay had significantly more volatile microbial beta diversity results than and a significantly decreased microbial maturity index compared with infants who did not; interestingly, all infants who showed improved length during the NICU stay were delivered by Caesarean section. Microbial beta diversity results were significantly different between the time of the NICU stay and all other time points (for children who were 2 or 4 years old and mothers when their children were 2 or 4 years old). IMPORTANCE Preterm infants are at greater risk of microbial insult than full-term infants, including reduced exposure to maternal vaginal and enteric microbes, higher rates of formula feeding, invasive procedures, and administration of antibiotics and medications that alter gastrointestinal pH. This investigation of the VLBW infant microbiome over the course of the neonatal intensive care unit (NICU) stay, and at ages 2 and 4 years, showed that the only clinical variables associated with significant differences in taxon abundance were weight gain during NICU stay (Klebsiella and Staphylococcus) and antibiotic administration (Streptococcus and Bifidobacterium). At 2 and 4 years of age, the microbiota of these VLBW infants became similar to the mothers’ microbiota. The number of microbial taxa shared between the infant or toddler and the mother varied, with least the overlap between infants and mothers. Overall, there was a significant association between the diversity and structure of the microbial community and infant weight and length gain in an at-risk childhood population.

scholarly journals Machine Learning Approaches to Predict In-Hospital Mortality among Neonates with Clinically Suspected Sepsis in the Neonatal Intensive Care Unit

2021 ◽  
Vol 11 (8) ◽  
pp. 695
Author(s):  
Jen-Fu Hsu ◽  
Ying-Feng Chang ◽  
Hui-Jun Cheng ◽  
Chi Yang ◽  
Chun-Yuan Lin ◽  
...  
Keyword(s):  
Machine Learning ◽  
Intensive Care Unit ◽  
Intensive Care ◽  
Hospital Mortality ◽  
Neonatal Intensive Care Unit ◽  
Clinical Features ◽  
Neonatal Intensive Care ◽  
Learning Approaches ◽  
Suspected Sepsis ◽  
Test Sets

Background: preterm and critically ill neonates often experience clinically suspected sepsis during their prolonged hospitalization in the neonatal intensive care unit (NICU), which can be the initial sign of final adverse outcomes. Therefore, we aimed to utilize machine learning approaches to predict neonatal in-hospital mortality through data-driven learning. Methods: a total of 1095 neonates who experienced clinically suspected sepsis in a tertiary-level NICU in Taiwan between August 2017 and July 2020 were enrolled. Clinically suspected sepsis was defined based on clinical features and laboratory criteria and the administration of empiric antibiotics by clinicians. The variables used for analysis included patient demographics, clinical features, laboratory data, and medications. The machine learning methods used included deep neural network (DNN), k-nearest neighbors, support vector machine, random forest, and extreme gradient boost. The performance of these models was evaluated using the area under the receiver operating characteristic curve (AUC). Results: the final in-hospital mortality of this cohort was 8.2% (90 neonates died). A total of 765 (69.8%) and 330 (30.2%) patients were randomly assigned to the training and test sets, respectively. Regarding the efficacy of the single model that most accurately predicted the outcome, DNN exhibited the greatest AUC (0.923, 95% confidence interval [CI] 0.953–0.893) and the best accuracy (95.64%, 95% CI 96.76–94.52%), Cohen’s kappa coefficient value (0.74, 95% CI 0.79–0.69) and Matthews correlation coefficient value (0.75, 95% CI 0.80–0.70). The top three most influential variables in the DNN importance matrix plot were the requirement of ventilator support at the onset of suspected sepsis, the feeding conditions, and intravascular volume expansion. The model performance was indistinguishable between the training and test sets. Conclusions: the DNN model was successfully established to predict in-hospital mortality in neonates with clinically suspected sepsis, and the machine learning algorithm is applicable for clinicians to gain insights and have better communication with families in advance.

scholarly journals GROWTH ANALYSIS OF PRETERM NEWBORNS WITH GASTROSCHISIS DURING HOSPITALIZATION IN A NEONATAL INTENSIVE CARE UNIT

2021 ◽  
Vol 58 (4) ◽  
pp. 504-508
Author(s):  
Juliana Zoboli Del BIGIO ◽  
Mário Cícero FALCÃO ◽  
Ana Cristina Aoun TANNURI
Keyword(s):  
Intensive Care Unit ◽  
Intensive Care ◽  
Weight Gain ◽  
Birth Weight ◽  
Parenteral Nutrition ◽  
Neonatal Intensive Care Unit ◽  
Gestational Age ◽  
Neonatal Intensive Care ◽  
Growth Analysis ◽  
Preterm Newborns

ABSTRACT BACKGROUND: Gastroschisis, especially complex type, prematurity and low birth weight are associated with a worse clinical outcome with higher mortality, higher incidence of sepsis and catheter-related infection, cholestasis, short bowel syndrome, greater number of days to achieve full diet, longer time of parenteral nutrition and longer hospitalization time. OBJECTIVE: To evaluate the growth of preterm newborns with gastroschisis during their hospitalization in the neonatal intensive care unit. METHODS: Descriptive study, based on a retrospective cohort (January 2012 to December 2018), including preterm newborns (gestational age less than 37 weeks) with simple and complex gastroschisis admitted in a tertiary neonatal intensive care unit. The following parameters were analyzed: maternal age, parity, type of delivery, birth weight, gender, gestational age, nutritional adequacy, type of gastroschisis, fasting time, parenteral nutrition time, time until achieving full enteral nutrition, hospitalization time, weight gain and outcome. The results were expressed in percentage, average, and median. RESULTS: A total of 101 newborns with gastroschisis were admitted, of which 59.4% were premature (80.7% of late preterm infants). From the maternal data, the mean age was 21.2 years and 68.3% were primiparous. Regarding childbirth: 80% were cesarean sections. From newborns: the average birth weight was 2137 g, 56.6% were female, the average gestational age was 34.8 weeks, the average weight gain was 20.8 g/day during hospitalization and 83.3% were discharged from the hospital. CONCLUSION: The growth analysis by weight gain (grams/day) during hospitalization in the intensive care unit showed that more than 90% of the sample presented acceptable or adequate weight gain.

scholarly journals Emphasise the role of kangaroo mother care in stabilisation of vitals and weight gain pattern in low birth weight babies inside neonatal intensive care unit

2021 ◽  
Vol 8 (4) ◽  
pp. 721
Author(s):  
Shwetal M. Bhatt ◽  
Khushboo N. Mehta ◽  
Ankita Maheshwari ◽  
Priyanka C. Parmar
Keyword(s):  
Intensive Care Unit ◽  
Intensive Care ◽  
Weight Gain ◽  
Birth Weight ◽  
Low Birth Weight ◽  
Neonatal Intensive Care Unit ◽  
Neonatal Intensive Care ◽  
Kangaroo Mother Care ◽  
P Value

Background: Kangaroo mother care (KMC) is routinely practiced in post-natal wards for care of stable low birth weight (LBW) infants. Objectives of the study were conducted to emphasize on the role of KMC in vitals stabilization and weight gain in LBW babies inside neonatal intensive care unit (NICU).Methods: Cross-sectional analytical quantitative study.Results: A total of 80 babies (48 males and 32 females) were enrolled and given KMC inside NICU. Mean birth weight was 1330 grams. Mean gestational age was 33 weeks (range 30-38 weeks). KMC was initiated within 72 hours of life in majority of babies (71%). Though 65% of them required oxygen support via prongs, KMC was started in them, with monitoring of vitals. No episode of apnea was observed during KMC sessions. Mean duration of KMC was 6 days (3-14 days). Heart rate dropped by 3-4 beats per minute (150+2.12 to 146+1.63, Respiratory rate decreased from 53+3.9 to 49+2.7, Oxygen saturation improved by 2-3% (93+0.42 to 96+0.71). Temperature rose from 36.78+0.01 to 37.07+0.02. P value for all vitals was 0.0001, which is considered significant (<0.05). Average weight gain was 76 grams during the average 6 days of KMC inside NICU, (p value=0.0001).  Conclusions: KMC was found to be effective for stabilization of vitals in NICU, early initiation and upgradation of feeding, early achievement of weight gain pattern, and early shift to postnatal ward by mother’s side. Also, no adverse effects were noted on the babies.

scholarly journals Using Machine Learning Techniques to Aid Empirical Antibiotic Therapy Decisions in the Intensive Care Unit of a General Hospital in Greece

Antibiotics ◽  
2020 ◽  
Vol 9 (2) ◽  
pp. 50 ◽  
Author(s):  
Georgios Feretzakis ◽  
Evangelos Loupelis ◽  
Aikaterini Sakagianni ◽  
Dimitris Kalles ◽  
Maria Martsoukou ◽  
...  
Keyword(s):  
Machine Learning ◽  
Intensive Care Unit ◽  
Intensive Care ◽  
Antibiotic Therapy ◽  
Antibiotic Susceptibility ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
High Morbidity ◽  
Empirical Antibiotic Therapy ◽  
Gram Negative

Hospital-acquired infections, particularly in the critical care setting, have become increasingly common during the last decade, with Gram-negative bacterial infections presenting the highest incidence among them. Multi-drug-resistant (MDR) Gram-negative infections are associated with high morbidity and mortality with significant direct and indirect costs resulting from long hospitalization due to antibiotic failure. Time is critical to identifying bacteria and their resistance to antibiotics due to the critical health status of patients in the intensive care unit (ICU). As common antibiotic resistance tests require more than 24 h after the sample is collected to determine sensitivity in specific antibiotics, we suggest applying machine learning (ML) techniques to assist the clinician in determining whether bacteria are resistant to individual antimicrobials by knowing only a sample’s Gram stain, site of infection, and patient demographics. In our single center study, we compared the performance of eight machine learning algorithms to assess antibiotic susceptibility predictions. The demographic characteristics of the patients are considered for this study, as well as data from cultures and susceptibility testing. Applying machine learning algorithms to patient antimicrobial susceptibility data, readily available, solely from the Microbiology Laboratory without any of the patient’s clinical data, even in resource-limited hospital settings, can provide informative antibiotic susceptibility predictions to aid clinicians in selecting appropriate empirical antibiotic therapy. These strategies, when used as a decision support tool, have the potential to improve empiric therapy selection and reduce the antimicrobial resistance burden.

Sign in / Sign up

Export Citation Format

Share Document