Artificial intelligence in acute respiratory distress syndrome: A systematic review

Rashid, Muhammed; Ramakrishnan, Manasvini; Chandran, Vijay; Nandish, S; Nair, Sreekumaran; Shanbhag, Vishal; Thunga, Girish

doi:10.1016/j.artmed.2022.102361

Cited by 13 publications

(15 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We used a five-fold cross-validation for splitting randomly the 1,000-patient dataset as 800 patients for training and 200 patients for validating and repeating this process 100 times (Supplemental File, http://links.lww.com/CCM/H413). We evaluated the final minimum number of variables model using two ML approaches: random forest (RF) and extreme gradient boosting (XGBoost) (32, 33) (Supplemental File, http://links.lww.com/CCM/H413). Calculations were performed using the R Core Team 2022 software (R Version 4.2.2) (https://www.r-project.org) (R Foundation for Statistical Computing, Vienna, Austria), package “randomForest” and “xgboost” (34, 35).…”

Section: Methodsmentioning

confidence: 99%

Predicting ICU Mortality in Acute Respiratory Distress Syndrome Patients Using Machine Learning: The Predicting Outcome and STratifiCation of severity in ARDS (POSTCARDS) Study*

Villar,

González-Martín,

Hernández-González

et al. 2023

Critical Care Medicine

View full text Add to dashboard Cite

Objectives: To assess the value of machine learning approaches in the development of a multivariable model for early prediction of ICU death in patients with acute respiratory distress syndrome (ARDS). Design: A development, testing, and external validation study using clinical data from four prospective, multicenter, observational cohorts. Setting: A network of multidisciplinary ICUs. Patients: A total of 1,303 patients with moderate-to-severe ARDS managed with lung-protective ventilation. Interventions: None. Measurements and Main Results: We developed and tested prediction models in 1,000 ARDS patients. We performed logistic regression analysis following variable selection by a genetic algorithm, random forest and extreme gradient boosting machine learning techniques. Potential predictors included demographics, comorbidities, ventilatory and oxygenation descriptors, and extrapulmonary organ failures. Risk modeling identified some major prognostic factors for ICU mortality, including age, cancer, immunosuppression, Pao 2/Fio 2, inspiratory plateau pressure, and number of extrapulmonary organ failures. Together, these characteristics contained most of the prognostic information in the first 24 hours to predict ICU mortality. Performance with machine learning methods was similar to logistic regression (area under the receiver operating characteristic curve [AUC], 0.87; 95% CI, 0.82–0.91). External validation in an independent cohort of 303 ARDS patients confirmed that the performance of the model was similar to a logistic regression model (AUC, 0.91; 95% CI, 0.87–0.94). Conclusions: Both machine learning and traditional methods lead to promising models to predict ICU death in moderate/severe ARDS patients. More research is needed to identify markers for severity beyond clinical determinants, such as demographics, comorbidities, lung mechanics, oxygenation, and extrapulmonary organ failure to guide patient management.

show abstract

Section: Methodsmentioning

confidence: 99%

Predicting ICU Mortality in Acute Respiratory Distress Syndrome Patients Using Machine Learning: The Predicting Outcome and STratifiCation of severity in ARDS (POSTCARDS) Study*

Villar,

González-Martín,

Hernández-González

et al. 2023

Critical Care Medicine

View full text Add to dashboard Cite

show abstract

“…SI‐ALI is a lethal illness worldwide 38 . During its progression, neutrophils can release NETs, web‐like structures comprised of DNA, histones, MPO, cathepsin G and other antibacterial substances, to neutralise and kill pathogens 39 .…”

Section: Discussionmentioning

confidence: 99%

METTL3‐mediated N6‐methyladenosine exacerbates ferroptosis via m6A‐IGF2BP2‐dependent mitochondrial metabolic reprogramming in sepsis‐induced acute lung injury

Zhang,

Wu,

Wang

et al. 2023

Clinical & Translational Med

View full text Add to dashboard Cite

Neutrophil extracellular traps (NETs), released by polymorphonuclear neutrophils (PMNs), exert a robust antimicrobial function in infectious diseases such as sepsis. NETs also contribute to the pathogenesis and exacerbation of sepsis. Although the lung is highly vulnerable to infections, few studies have explored the role of NETs in sepsis‐induced acute lung injury (SI‐ALI). We demonstrate that NETs induce SI‐ALI via enhanced ferroptosis in alveolar epithelial cells. Our findings reveal that the excessive release of NETs in patients and mice with SI‐ALI is accompanied by upregulation of ferroptosis depending on METTL3‐induced m6A modification of hypoxia‐inducible factor‐1α (HIF‐1α) and subsequent mitochondrial metabolic reprogramming. In addition to conducting METTL3 overexpression and knockdown experiments in vitro, we also investigated the impact of ferroptosis on SI‐ALI caused by NETs in a caecum ligation and puncture (CLP)‐induced SI‐ALI model using METTL3 condition knockout (CKO) mice and wild‐type mice. Our results indicate the crucial role of NETs in the progression of SI‐ALI via NET‐activated METTL3 m6A‐IGF2BP2‐dependent m6A modification of HIF‐1α, which further contributes to metabolic reprogramming and ferroptosis in alveolar epithelial cells.

show abstract

“…Rashid et al [14] provided a comprehensive review of an application of AI in ARDS. The authors reveal 19 studies between year 2002 and 2020 that AI has been applied in various ARDS for diagnosis (53%), risk stratification (5%), prediction of severity (17%), management (10%), prediction of mortality (10%), and decision making (5%).…”

Section: Related Workmentioning

confidence: 99%

“…More studies could improve MLA readiness to obtain a more insightful process of MLA in clinical practice [58]. Moreover, studies MLA related to patient demographic is highly needed since its factor have a higher impact than physiological factors in the MLA [14].…”

Section: Research Question 3: How To Develop the ML For Prediction/de...mentioning

confidence: 99%

Mechanical ventilation intervention based on machine learning from vital signs monitoring: a scoping review

Baidillah

Busono

Riyanto

2023

Meas. Sci. Technol.

View full text Add to dashboard Cite

BackgroundAsynchronous breathing (AB) during mechanical ventilation (MV) may lead to a detrimental effect on the patient's condition. Due to the massive amount of data displayed in a large ICU, a machine learning algorithm (MLA) was proposed extensively to extract the patterns within the multiple continuous-in-time vital signs, to determine which are the variables that will predict the AB, to intervene in the MV as an early warning system, and finally to replace a highly demand of clinician's cognition. ObjectiveThis study reviews the MLA for prediction and detection models from vital signs monitoring data for MV intervention. MethodsPublication on MLA development on MV intervention based on vital signs monitoring to support clinicians' decision-making process was extracted from the three electronic academic research databases Web of Science Core Collection (WoSCC), ScienceDirect, and PUBMED Central to February 2023. Findings838 papers from the electronic academic research databases are extracted. There are 14 review papers, while 25 related papers that pass with the Quality Assessments (QA). ConclusionsFew studies have been published that considered VS monitoring data along with the MV parameters waveforms for MV intervention. Vital signs monitoring data is not the only predictor in the developed MLA. Most studies suggested that developing the MLA for direct MV intervention requires more concern in the pre-processing of real-time data to avoid false positive and false detection than developing MLA itself. 

show abstract

Artificial intelligence in acute respiratory distress syndrome: A systematic review

Cited by 13 publications

References 38 publications

Predicting ICU Mortality in Acute Respiratory Distress Syndrome Patients Using Machine Learning: The Predicting Outcome and STratifiCation of severity in ARDS (POSTCARDS) Study*

Predicting ICU Mortality in Acute Respiratory Distress Syndrome Patients Using Machine Learning: The Predicting Outcome and STratifiCation of severity in ARDS (POSTCARDS) Study*

METTL3‐mediated N6‐methyladenosine exacerbates ferroptosis via m6A‐IGF2BP2‐dependent mitochondrial metabolic reprogramming in sepsis‐induced acute lung injury

Mechanical ventilation intervention based on machine learning from vital signs monitoring: a scoping review

Contact Info

Product

Resources

About