Real-time prediction of HFNC treatment failure in acute hypoxemic respiratory failure using machine learning
Abstract Accurate and timely prediction of high-flow nasal cannula (HFNC) treatment failure in patients with acute hypoxemic respiratory failure (AHRF) can lower patient mortality. Previous studies have highlighted inconsistencies in the predictive performance of existing indices, such as ROX and mR...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-08-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-16061-x |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849226417573724160 |
|---|---|
| author | Xiaojie Li Chunliang Jiang Qingyan Xie Huiquan Wang Jiameng Xu Guanjun Liu Panpan Chang Guang Zhang |
| author_facet | Xiaojie Li Chunliang Jiang Qingyan Xie Huiquan Wang Jiameng Xu Guanjun Liu Panpan Chang Guang Zhang |
| author_sort | Xiaojie Li |
| collection | DOAJ |
| description | Abstract Accurate and timely prediction of high-flow nasal cannula (HFNC) treatment failure in patients with acute hypoxemic respiratory failure (AHRF) can lower patient mortality. Previous studies have highlighted inconsistencies in the predictive performance of existing indices, such as ROX and mROX, which are limited by their reliance on oxygenation parameters alone. To address this, we developed a machine learning-based predictive model using temporal data from AHRF patients, aimed at facilitating quicker development of individualized treatment plans and intervention strategies for healthcare professionals. We extracted 15 non-invasive and 15 laboratory features, including patient demographic characteristics, Glasgow Coma Scale, blood gas analysis, chemical assay, and complete blood cell count features. In addition to five machine learning models and an ensemble classifier, an long short-term memory (LSTM) network was included to assess deep learning performance on time-series data. Our study enrolled 427 patients with 498 treatment records. The soft-voting ensemble algorithm achieved an optimal predictive performance with an AUC of 0.839 (95% CI 0.786–0.889) for the all-features model, while logistic regression using common features achieved an AUC of 0.767 (95% CI 0.704–0.825), outperforming ROX and mROX indices. Incorporating blood gas analysis features improved the non-invasive model’s performance by 0.104. This study introduces a machine learning model integrated with a dynamic real-time alert system for predicting HFNC treatment failure in AHRF patients, demonstrating improved performance over traditional indices in internal validation and showing potential for decision support in select healthcare settings. |
| format | Article |
| id | doaj-art-464a8fe6c3344776ab9dbf3d4346f78a |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-08-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-464a8fe6c3344776ab9dbf3d4346f78a2025-08-24T11:21:58ZengNature PortfolioScientific Reports2045-23222025-08-0115111710.1038/s41598-025-16061-xReal-time prediction of HFNC treatment failure in acute hypoxemic respiratory failure using machine learningXiaojie Li0Chunliang Jiang1Qingyan Xie2Huiquan Wang3Jiameng Xu4Guanjun Liu5Panpan Chang6Guang Zhang7School of Life Sciences, Tiangong UniversitySchool of Control Science and Engineering, Tiangong UniversitySchool of Life Sciences, Tiangong UniversitySchool of Life Sciences, Tiangong UniversitySchool of Life Sciences, Tiangong UniversitySystems Engineering Institute, Academy of Military Sciences, People’s Liberation ArmyTrauma Medicine Center of Peking University People’s Hospital, Key Laboratory of Trauma and Neural Regeneration (Peking University) Ministry of Education, National Center for Trauma Medicine of ChinaSystems Engineering Institute, Academy of Military Sciences, People’s Liberation ArmyAbstract Accurate and timely prediction of high-flow nasal cannula (HFNC) treatment failure in patients with acute hypoxemic respiratory failure (AHRF) can lower patient mortality. Previous studies have highlighted inconsistencies in the predictive performance of existing indices, such as ROX and mROX, which are limited by their reliance on oxygenation parameters alone. To address this, we developed a machine learning-based predictive model using temporal data from AHRF patients, aimed at facilitating quicker development of individualized treatment plans and intervention strategies for healthcare professionals. We extracted 15 non-invasive and 15 laboratory features, including patient demographic characteristics, Glasgow Coma Scale, blood gas analysis, chemical assay, and complete blood cell count features. In addition to five machine learning models and an ensemble classifier, an long short-term memory (LSTM) network was included to assess deep learning performance on time-series data. Our study enrolled 427 patients with 498 treatment records. The soft-voting ensemble algorithm achieved an optimal predictive performance with an AUC of 0.839 (95% CI 0.786–0.889) for the all-features model, while logistic regression using common features achieved an AUC of 0.767 (95% CI 0.704–0.825), outperforming ROX and mROX indices. Incorporating blood gas analysis features improved the non-invasive model’s performance by 0.104. This study introduces a machine learning model integrated with a dynamic real-time alert system for predicting HFNC treatment failure in AHRF patients, demonstrating improved performance over traditional indices in internal validation and showing potential for decision support in select healthcare settings.https://doi.org/10.1038/s41598-025-16061-xHigh-flow nasal cannulaAcute hypoxemic respiratory failureMachine learning methodsReal-time dynamic alertROX indexmROX index |
| spellingShingle | Xiaojie Li Chunliang Jiang Qingyan Xie Huiquan Wang Jiameng Xu Guanjun Liu Panpan Chang Guang Zhang Real-time prediction of HFNC treatment failure in acute hypoxemic respiratory failure using machine learning Scientific Reports High-flow nasal cannula Acute hypoxemic respiratory failure Machine learning methods Real-time dynamic alert ROX index mROX index |
| title | Real-time prediction of HFNC treatment failure in acute hypoxemic respiratory failure using machine learning |
| title_full | Real-time prediction of HFNC treatment failure in acute hypoxemic respiratory failure using machine learning |
| title_fullStr | Real-time prediction of HFNC treatment failure in acute hypoxemic respiratory failure using machine learning |
| title_full_unstemmed | Real-time prediction of HFNC treatment failure in acute hypoxemic respiratory failure using machine learning |
| title_short | Real-time prediction of HFNC treatment failure in acute hypoxemic respiratory failure using machine learning |
| title_sort | real time prediction of hfnc treatment failure in acute hypoxemic respiratory failure using machine learning |
| topic | High-flow nasal cannula Acute hypoxemic respiratory failure Machine learning methods Real-time dynamic alert ROX index mROX index |
| url | https://doi.org/10.1038/s41598-025-16061-x |
| work_keys_str_mv | AT xiaojieli realtimepredictionofhfnctreatmentfailureinacutehypoxemicrespiratoryfailureusingmachinelearning AT chunliangjiang realtimepredictionofhfnctreatmentfailureinacutehypoxemicrespiratoryfailureusingmachinelearning AT qingyanxie realtimepredictionofhfnctreatmentfailureinacutehypoxemicrespiratoryfailureusingmachinelearning AT huiquanwang realtimepredictionofhfnctreatmentfailureinacutehypoxemicrespiratoryfailureusingmachinelearning AT jiamengxu realtimepredictionofhfnctreatmentfailureinacutehypoxemicrespiratoryfailureusingmachinelearning AT guanjunliu realtimepredictionofhfnctreatmentfailureinacutehypoxemicrespiratoryfailureusingmachinelearning AT panpanchang realtimepredictionofhfnctreatmentfailureinacutehypoxemicrespiratoryfailureusingmachinelearning AT guangzhang realtimepredictionofhfnctreatmentfailureinacutehypoxemicrespiratoryfailureusingmachinelearning |