{"id":21575850,"url":"https://github.com/deib-geco/ards_h-cpap_ml","last_synced_at":"2026-01-04T06:14:48.166Z","repository":{"id":239797357,"uuid":"739303846","full_name":"DEIB-GECO/ARDS_H-CPAP_ML","owner":"DEIB-GECO","description":"Machine learning-based forecast of Helmet-CPAP therapy failure in Acute Respiratory Distress Syndrome patients","archived":false,"fork":false,"pushed_at":"2024-11-09T16:13:21.000Z","size":6154,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":4,"default_branch":"main","last_synced_at":"2025-01-24T13:34:41.149Z","etag":null,"topics":["ards","covid-19","h-cpap","machine-learning","predicting-h-cpap-therapy-failure-in-ards-patients"],"latest_commit_sha":null,"homepage":"","language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/DEIB-GECO.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-01-05T08:43:28.000Z","updated_at":"2024-11-09T16:13:25.000Z","dependencies_parsed_at":"2024-05-15T12:32:53.448Z","dependency_job_id":"505e2f6a-2700-4371-891a-84ee7e12e47e","html_url":"https://github.com/DEIB-GECO/ARDS_H-CPAP_ML","commit_stats":null,"previous_names":["deib-geco/ards_h-cpap_ml"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/DEIB-GECO%2FARDS_H-CPAP_ML","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/DEIB-GECO%2FARDS_H-CPAP_ML/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/DEIB-GECO%2FARDS_H-CPAP_ML/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/DEIB-GECO%2FARDS_H-CPAP_ML/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/DEIB-GECO","download_url":"https://codeload.github.com/DEIB-GECO/ARDS_H-CPAP_ML/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":244173554,"owners_count":20410300,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ards","covid-19","h-cpap","machine-learning","predicting-h-cpap-therapy-failure-in-ards-patients"],"created_at":"2024-11-24T12:14:26.862Z","updated_at":"2026-01-04T06:14:48.129Z","avatar_url":"https://github.com/DEIB-GECO.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"# Machine learning-based forecast of Helmet-CPAP therapy failure in Acute Respiratory Distress Syndrome patients\n\n\u003cbr\u003e\n\n## Authors\nRiccardo Campi\u003csup\u003ea\u003c/sup\u003e, Antonio De Santis\u003csup\u003ea\u003c/sup\u003e, Paolo Colombo\u003csup\u003eb\u003c/sup\u003e, Paolo Scarpazza\u003csup\u003eb\u003c/sup\u003e, Marco Masseroli\u003csup\u003ea\u003c/sup\u003e\n\n\u003csup\u003ea\u003c/sup\u003eDepartment of Electronics Information and Bioengineering, Politecnico di Milano, Piazza L. Da Vinci 32, Milano, MI, 20133, Italy\u003cbr\u003e\n\u003csup\u003eb\u003c/sup\u003eAzienda Socio-Sanitaria Territoriale (ASST) della Brianza, Via Santi Cosma e Damiano 10, Vimercate, MB, 20871, Italy\n\nEmail addresses: riccardo.campi@mail.polimi.it (Riccardo Campi), antonio.desantis@polimi.it (Antonio De Santis), paolo.colombo@asst-brianza.it (Paolo Colombo), paolo.scarpazza@asst-brianza.it (Paolo Scarpazza), marco.masseroli@polimi.it (Marco Masseroli)\n\n\u003cbr\u003e\n\n## Abstract\n### Background and Objective:\nHelmet-Continuous Positive Airway Pressure (H-CPAP) is a non-invasive respiratory support that is used for the treatment of Acute Respiratory Distress Syndrome (ARDS), a severe medical condition diagnosed when symptoms like profound hypoxemia, pulmonary opacities on radiography, or unexplained respiratory failure are present.\nIt can be classified as mild, moderate or severe.\nH-CPAP therapy is recommended as the initial treatment approach for mild ARDS.\nEven though the efficacy of H-CPAP in managing patients with moderate-to-severe hypoxemia remains unclear, its use has increased for these cases in response to the emergence of the COVID-19 Pandemic.\n\nUsing the electronic medical records (EMR) from the Pulmonology Department of Vimercate Hospital, in this study we develop and evaluate a Machine Learning (ML) system able to predict the failure of H-CPAP therapy on ARDS patients.\n\n### Methods:\nThe Vimercate Hospital EMR provides demographic information, blood tests, and vital parameters of all hospitalizations of patients who are treated with H-CPAP and diagnosed with ARDS.\nThis data is used to create a dataset of 622 records and 38 features, with 70-30% split between training and test set.\nDifferent ML models such as SVM, XGBoost, Neural Network, Random Forest, and Logistic Regression are iteratively trained in a cross-validation fashion.\nWe also apply a feature selection algorithm to improve predictions quality and reduce the number of features.\n\n### Results and Conclusions:\nThe SVM and Neural Network models proved to be the most effective, achieving final accuracies of 95.19% and 94.65%, respectively. In terms of F1-score, the models scored 88.61% and 87.18%, respectively. Additionally, the SVM and XGBoost models performed well with a reduced number of features (23 and 13, respectively).\nThe PaO2/FiO2 Ratio, C-Reactive Protein, and O2 Saturation resulted as the most important features, followed by Heartbeats, White Blood Cells, and D-Dimer, in accordance with the clinical scientific literature.\n\n\u003cbe\u003e\n\n## Code Availability\n\nPlease refer to [ARDS_H-CPAP_ML](https://github.com/riccamper/ARDS_H-CPAP_ML)\n\n\n\n## Reproducibility\n\n| File/folder        | Description                                                                    |\n|:------------------ |:------------------------------------------------------------------------------ |\n| /main.ipynb        | code used to clean the datasets and train the models                           |\n| /functions.py      | accessory functions and classes                                                |\n| /models_vimercate/ | folder containing the trained models                                           |\n| /datasets/         | folder containing the dataset \"dataset_vimercate.csv\" used to train the models |\n| /images/           | folder containing the obtained images                                          |\n\n\u003cbr\u003e\n\n## Institutional Review Board Statement\nHuman participants were involved in this research; the study was conducted in accordance with the Declaration of Helsinki. Our study was approved by the local institution, Vimercate Hospital, ASST-Brianza, according to the legal requirements concerning observational studies (Resolutions 0000573 27/07/2021 and 0000133 22/02/2023).\n\n## Informed Consent Statement\nDue to the nature of the present observational study and data anonymization, the patients' consent to participate was not required, as declared by the ASST Brianza Ethics Committee.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdeib-geco%2Fards_h-cpap_ml","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fdeib-geco%2Fards_h-cpap_ml","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdeib-geco%2Fards_h-cpap_ml/lists"}