This repository contains code to reproduce the paper Predicting intraoperative hypotension using deep learning with waveforms of arterial blood pressure, electroencephalogram, and electrocardiogram: Retrospective study by Y-Y Jo et al. (2022) [1]
Team members, alphabetically by first name:
- Aaron Barrie (abarrie2@illinois.edu)
- Maciej Wieczorek (maciejw2@illinois.edu)
- Shaun Phillips (shaunap2@illinois.edu)
The Project Draft submission can be found in the DL4H_Team_24_Project_Draft.ipnyb notebook. The development version can be found in the DL4H_Team_24.ipnyb notebook.
The code can be run locally, or in Google Colab. Complete instructions are contained inside of the notebook.
The fastest way to run a demo version of the code is in Google Colab.
- Open the DL4H_Team_24_Project_Draft.ipnyb notebook.
- Click on "Open in Colab" in the top left.
- Set the runtime type to "T4 GPU". The code will also run in the default "CPU" instance, but training will take ~10x longer. To do this, in the top right click on the down arrow (Additional connection options) next to "Reconnect", select "Change runtime type", select "T4 GPU" and hit "Save".
- In the Runtime menu, click on "Run all".
In the default demo mode, the notebook should take approximately 5 minutes to run on a T4 GPU instance, including 6 training epochs.
The fastest way to run a demo version of the code is in Google Colab.
- Open the DL4H_Team_24_Project_Final.ipnyb notebook.
- Click on "Open in Colab" in the top left.
- Set the runtime type to "T4 GPU". The code will also run in the default "CPU" instance, but training will take ~10x longer. To do this, in the top right click on the down arrow (Additional connection options) next to "Reconnect", select "Change runtime type", select "T4 GPU" and hit "Save".
- In the Runtime menu, click on "Run all".
In the default demo mode, the notebook should take approximately 8 minutes to run on a T4 GPU instance, including 6 training epochs.
Open Dataset https://vitaldb.net/dataset/
Github https://github.com/vitaldb
Open source code for processing vital files. https://github.com/vitaldb/vitalutils
Jupyter Notebook Examples https://github.com/vitaldb/examples
- Jo Y-Y, Jang J-H, Kwon J-m, Lee H-C, Jung C-W, Byun S, et al. “Predicting intraoperative hypotension using deep learning with waveforms of arterial blood pressure, electroencephalogram, and electrocardiogram: Retrospective study.” PLoS ONE, (2022) 17(8): e0272055 https://doi.org/10.1371/journal.pone.0272055
- Hatib, Feras, Zhongping J, Buddi S, Lee C, Settels J, Sibert K, Rhinehart J, Cannesson M “Machine-learning Algorithm to Predict Hypotension Based on High-fidelity Arterial Pressure Waveform Analysis” Anesthesiology (2018) 129:4 https://doi.org/10.1097/ALN.0000000000002300
- Bao, X., Kumar, S.S., Shah, N.J. et al. "AcumenTM hypotension prediction index guidance for prevention and treatment of hypotension in noncardiac surgery: a prospective, single-arm, multicenter trial." Perioperative Medicine (2024) 13:13 https://doi.org/10.1186/s13741-024-00369-9
- Lee, HC., Park, Y., Yoon, S.B. et al. VitalDB, a high-fidelity multi-parameter vital signs database in surgical patients. Sci Data 9, 279 (2022). https://doi.org/10.1038/s41597-022-01411-5
- Li Q., Mark R.G. & Clifford G.D. "Artificial arterial blood pressure artifact models and an evaluation of a robust blood pressure and heart rate estimator." BioMed Eng OnLine. (2009) 8:13. pmid:19586547 https://doi.org/10.1186/1475-925X-8-13
- Park H-J, "VitalDB Python Example Notebooks" GitHub Repository https://github.com/vitaldb/examples/blob/master/hypotension_art.ipynb
- Vest A, Da Poian G, Li Q, Liu C, Nemati S, Shah A, Clifford GD, "An Open Source Benchmarked Toolbox for Cardiovascular Waveform and Interval Analysis", Physiological measurement 39, no. 10 (2018): 105004. DOI:10.5281/zenodo.1243111; 2018.