April 17, 2018AnesthesiologyOpen Access

The deep neural network with a reduced feature set and ASA Physical Status had an AUC of 0.91 (95% CI, 0.88 to 0.93), while the Risk Stratification Index had the highest AUC at 0.97 (95% CI, 0.94 to 0.99).

View Full Paper

Ask AI

Bookmark

Why the study?

Does a deep neural network model trained on intraoperative features improve prediction of postoperative in-hospital mortality compared to existing clinical scores?

Population

59,985 surgical patients with 87 features extracted at the end of surgery

Comparison

Deep neural network model trained on… vs ASA Physical Status, logistic regression…

Design

Cohort

Follow-up

in-hospital

Authors

CLChristine K. Lee IHIra Hofer EGEilon Gabel

Discussion

Member takes

Overview

Intraoperative DNNs predict in-hospital mortality; leaves open whether they can outperform established scores like the Risk Stratification Index in practice.

Structured PICO

Does a deep neural network model trained on intraoperative features improve prediction of postoperative in-hospital mortality compared to existing clinical scores?

Population

59,985 surgical patients with 87 features extracted at the end of surgery

Intervention

Deep neural network model trained on intraoperative features (with or without ASA Physical Status Classification)

Comparator

ASA Physical Status, logistic regression, Surgical Apgar, Preoperative Score to Predict Postoperative Mortality, Risk Quantification Index, and Risk Stratification Index

Outcome

Postoperative in-hospital mortalityhard clinical

Deep neural networks can predict postoperative in-hospital mortality using intraoperative data but are not yet superior to existing clinical scores like the Risk Stratification Index.

Cite This Study

Lee et al. (2018) studied this question.

synapsesocial.com/papers/6a07b43feb3303cf0479f8b7 https://doi.org/10.1097/aln.0000000000002186

View Full Paper

Ask AI

Bookmark

April 17, 2018AnesthesiologyOpen Access

The deep neural network with a reduced feature set and ASA Physical Status had an AUC of 0.91 (95% CI, 0.88 to 0.93), while the Risk Stratification Index had the highest AUC at 0.97 (95% CI, 0.94 to 0.99).

View Full Paper

Ask AI

Bookmark

Why the study?

Does a deep neural network model trained on intraoperative features improve prediction of postoperative in-hospital mortality compared to existing clinical scores?

Population

59,985 surgical patients with 87 features extracted at the end of surgery

Comparison

Deep neural network model trained on… vs ASA Physical Status, logistic regression…

Design

Cohort

Follow-up

in-hospital

Authors

CLChristine K. Lee IHIra Hofer EGEilon Gabel

Discussion

Member takes

Overview

Intraoperative DNNs predict in-hospital mortality; leaves open whether they can outperform established scores like the Risk Stratification Index in practice.

Structured PICO

Does a deep neural network model trained on intraoperative features improve prediction of postoperative in-hospital mortality compared to existing clinical scores?

Population

59,985 surgical patients with 87 features extracted at the end of surgery

Intervention

Deep neural network model trained on intraoperative features (with or without ASA Physical Status Classification)

Comparator

ASA Physical Status, logistic regression, Surgical Apgar, Preoperative Score to Predict Postoperative Mortality, Risk Quantification Index, and Risk Stratification Index

Outcome

Postoperative in-hospital mortalityhard clinical

Deep neural networks can predict postoperative in-hospital mortality using intraoperative data but are not yet superior to existing clinical scores like the Risk Stratification Index.

Cite This Study

Lee et al. (2018) studied this question.

synapsesocial.com/papers/6a07b43feb3303cf0479f8b7 https://doi.org/10.1097/aln.0000000000002186

View Full Paper

Ask AI

Bookmark