CoChat

Predicting Abnormalities in Laboratory Values of Patients in the Intensive Care Unit Using Different Deep Learning Models: Comparative Study

Open access

The DOI matches its Crossref record; 32 of the 32 references it cites were found in a registry or an index.

Abstract

Background: In recent years, the volume of medical knowledge and health data has increased rapidly. For example, the increased availability of electronic health records (EHRs) provides accurate, up-to-date, and complete information about patients at the point of care and enables medical staff to have quick access to patient records for more coordinated and efficient care. With this increase in knowledge, the complexity of accurate, evidence-based medicine tends to grow all the time. Health care workers must deal with an increasing amount of data and documentation. Meanwhile, relevant patient data are frequently overshadowed by a layer of less relevant data, causing medical staff to often miss important values or abnormal trends and their importance to the progression of the patient’s case. Objective: The goal of this work is to analyze the current laboratory results for patients in the intensive care unit (ICU) and classify which of these lab values could be abnormal the next time the test is done. Detecting near-future abnormalities can be useful to support clinicians in their decision-making process in the ICU by drawing their attention to the important values and focus on future lab testing, saving them both time and money. Additionally, it will give doctors more time to spend with patients, rather than skimming through a long list of lab values. Methods: We used Structured Query Language to extract 25 lab values for mechanically ventilated patients in the ICU from the MIMIC-III and eICU data sets. Additionally, we applied time-windowed sampling and holding, and a support vector machine to fill in the missing values in the sparse time series, as well as the Tukey range to detect and delete anomalies. Then, we used the data to train 4 deep learning models for time series classification, as well as a gradient boosting–based algorithm and compared their performance on both data sets. Results: The models tested in this work (deep neural networks and gradient boosting), combined with the preprocessing pipeline, achieved an accuracy of at least 80% on the multilabel classification task. Moreover, the model based on the multiple convolutional neural network outperformed the other algorithms on both data sets, with the accuracy exceeding 89%. Conclusions: In this work, we show that using machine learning and deep neural networks to predict near-future abnormalities in lab values can achieve satisfactory results. Our system was trained, validated, and tested on 2 well-known data sets to ensure that our system bridged the reality gap as much as possible. Finally, the model can be used in combination with our preprocessing pipeline on real-life EHRs to improve patients’ diagnosis and treatment.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.2196/37658

Cites

32 of 32 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Overall abnormality detection system in practice. DNN: deep neural network.
Figure 1 Overall abnormality detection system in practice. DNN: deep neural network.CC BY · PubMed Central
Figure 2 Overall system model used in our study when trained on the MIMIC-III data set and tested on the eICU data set. ICU: intensive care unit; Sigmoid is an activation function; L: lab value; t: time step.
Figure 2 Overall system model used in our study when trained on the MIMIC-III data set and tested on the eICU data set. ICU: intensive care unit; Sigmoid is an activation function; L: lab value; t: time step.CC BY · PubMed Central
Figure 3 Moving window technique to extract sequences from intensive care unit stays. X and Y represent the input and output data respectively; W represents the windows extracted from the input sequences.
Figure 3 Moving window technique to extract sequences from intensive care unit stays. X and Y represent the input and output data respectively; W represents the windows extracted from the input sequences.CC BY · PubMed Central
Figure 4 LSTM architecture used in our experiments. LSTM: long short-term memory; ReLU: rectified linear unit; Tanh, ReLU and Sigmoid are activation functions.
Figure 4 LSTM architecture used in our experiments. LSTM: long short-term memory; ReLU: rectified linear unit; Tanh, ReLU and Sigmoid are activation functions.CC BY · PubMed Central
Figure 5 Multiple convolutional neural network model architecture used in our experiments. Conv1D: 1D convolutional layer; LeakyReLU: leaky rectified linear unit; ReLU: rectified linear unit; Sigmoid, LeakyReLU, and ReLU are activation functions.
Figure 5 Multiple convolutional neural network model architecture used in our experiments. Conv1D: 1D convolutional layer; LeakyReLU: leaky rectified linear unit; ReLU: rectified linear unit; Sigmoid, LeakyReLU, and ReLU are activation functions.CC BY · PubMed Central
Figure 6 Transformer architecture used in our experiments. Conv1D: 1D convolutional layer; Time2Vec: time to vector transformation; ReLU: rectified linear unit; ReLU and Sigmoid are activation functions.
Figure 6 Transformer architecture used in our experiments. Conv1D: 1D convolutional layer; Time2Vec: time to vector transformation; ReLU: rectified linear unit; ReLU and Sigmoid are activation functions.CC BY · PubMed Central
Figure 7 TCN architecture used in our experiments. LeakyReLU: leaky rectified linear unit; ReLU: rectified linear unit; TCN: temporal convolutional network; LeakyReLU, ReLU, and Sigmoid are activation functions.
Figure 7 TCN architecture used in our experiments. LeakyReLU: leaky rectified linear unit; ReLU: rectified linear unit; TCN: temporal convolutional network; LeakyReLU, ReLU, and Sigmoid are activation functions.CC BY · PubMed Central
Figure 8 Validation loss of the different models. LSTM: long short-term network; M-CNN: multiple convolutional neural network; TCN: temporal convolutional network; Val.: validation; ICU: intensive care unit.
Figure 8 Validation loss of the different models. LSTM: long short-term network; M-CNN: multiple convolutional neural network; TCN: temporal convolutional network; Val.: validation; ICU: intensive care unit.CC BY · PubMed Central
Figure 9 Validation F1 score of the different models. LSTM: long short-term network; M-CNN: multiple convolutional neural network; TCN: temporal convolutional network; Val.: validation; ICU: intensive care unit.
Figure 9 Validation F1 score of the different models. LSTM: long short-term network; M-CNN: multiple convolutional neural network; TCN: temporal convolutional network; Val.: validation; ICU: intensive care unit.CC BY · PubMed Central
Figure 10 Validation accuracy of the different models. LSTM: long short-term network; M-CNN: multiple convolutional neural network; TCN: temporal convolutional network; Val.: validation; ICU: intensive care unit.
Figure 10 Validation accuracy of the different models. LSTM: long short-term network; M-CNN: multiple convolutional neural network; TCN: temporal convolutional network; Val.: validation; ICU: intensive care unit.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 32 references

Research works printed but not linked: 6

  1. Moor, M, 2019, Proceedings of the 4th Machine Learning for Healthcare Conference, PMLR, 106, 2
  2. VaswaniAShazeerNParmarNUszkoreitJJonesLGomezANKaiserLPolosukhinIAttention is all you needProceedings of the Advances in Neural Information Processing Systems201731st Conference on Neural Information Processing Systems (NIPS 2017)December 4-9, 2017Long Beach, CA, United StatesCurran Associates, Inc
  3. Kazemi, M, ArXiv
  4. Lea, C, 2016, Computer Vision – ECCV 2016 Workshops. ECCV 2016. Lecture Notes in Computer Science
  5. KeGMengQFinleyTWangTChenWMaWLiuTYLightGBM: a highly efficient gradient boosting decision treeProceedings of the Advances in Neural Information Processing Systems201731st Conference on Neural Information Processing Systems (NIPS 2017)December 4-9, 2017Long Beach, CA, United StatesCurran Associates, Inc
  6. MaasALHannunAYNgAYRectifier nonlinearities improve neural network acoustic modelsProceedings of the 30th International Conference on Machine Learning20130630th International Conference on Machine LearningJune 17-19, 2013Atlanta, Georgia, United States

Cited by

Metadata

Authors
Ayad, A., Hallawa, A., Peine, A., Martin, L., Fazlic, L. B., Dartmann, G., Marx, G., & Schmeink, A.
Affiliations
  • RWTH Aachen University (DE)
  • Universitätsklinikum Aachen (DE)
  • Trier University of Applied Sciences (DE)
Corresponding author
Ayad, Ahmad
Venue
JMIR Medical Informatics
Published
2022-08-24
Volume
10
Issue
8
Pages
e37658
ISSN
2291-9694 (ISSN-L)
OpenAlex
W4292953849
Cited by
14
References
31
Citations by year
2026: 3 · 2025: 5 · 2024: 5 · 2023: 1
FWCI
0.91
Type
journal article
Language
en
Open access
Licence
cc-by
Locations
  • JMIR Medical Informatics (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
  • RWTH Publications (RWTH Aachen) (submittedVersion, other-oa)
  • RWTH Publications (RWTH Aachen)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
  • Physical Sciences › Computer Science › Artificial Intelligence › Anomaly Detection Techniques and Applications
  • Physical Sciences › Computer Science › Signal Processing › Time Series Analysis and Forecasting
Keywords
intensive care unit patients, mechanical ventilation, time series classification, multi-label classification, convolutional neural networks, gradient boosting, electronic health records, MIMIC-III, eICU, missing data imputation, clinical decision support, deep learning model comparison, data preprocessing pipeline

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.