Artificial intelligence sepsis prediction algorithm learns to say “I don’t know”
Abstract Sepsis is a leading cause of morbidity and mortality worldwide. Early identification of sepsis is important as it allows timely administration of potentially life-saving resuscitation and antimicrobial therapy. We present COMPOSER (COnformal Multidimensional Prediction Of SEpsis Risk), a de...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2021-09-01
|
| Series: | npj Digital Medicine |
| Online Access: | https://doi.org/10.1038/s41746-021-00504-6 |
| _version_ | 1827608659513835520 |
|---|---|
| author | Supreeth P. Shashikumar Gabriel Wardi Atul Malhotra Shamim Nemati |
| author_facet | Supreeth P. Shashikumar Gabriel Wardi Atul Malhotra Shamim Nemati |
| author_sort | Supreeth P. Shashikumar |
| collection | DOAJ |
| description | Abstract Sepsis is a leading cause of morbidity and mortality worldwide. Early identification of sepsis is important as it allows timely administration of potentially life-saving resuscitation and antimicrobial therapy. We present COMPOSER (COnformal Multidimensional Prediction Of SEpsis Risk), a deep learning model for the early prediction of sepsis, specifically designed to reduce false alarms by detecting unfamiliar patients/situations arising from erroneous data, missingness, distributional shift and data drifts. COMPOSER flags these unfamiliar cases as indeterminate rather than making spurious predictions. Six patient cohorts (515,720 patients) curated from two healthcare systems in the United States across intensive care units (ICU) and emergency departments (ED) were used to train and externally and temporally validate this model. In a sequential prediction setting, COMPOSER achieved a consistently high area under the curve (AUC) (ICU: 0.925–0.953; ED: 0.938–0.945). Out of over 6 million prediction windows roughly 20% and 8% were identified as indeterminate amongst non-septic and septic patients, respectively. COMPOSER provided early warning within a clinically actionable timeframe (ICU: 12.2 [3.2 22.8] and ED: 2.1 [0.8 4.5] hours prior to first antibiotics order) across all six cohorts, thus allowing for identification and prioritization of patients at high risk for sepsis. |
| first_indexed | 2024-03-09T07:17:48Z |
| format | Article |
| id | doaj.art-6852a8b842cb4d289b750afc458515bb |
| institution | Directory Open Access Journal |
| issn | 2398-6352 |
| language | English |
| last_indexed | 2024-03-09T07:17:48Z |
| publishDate | 2021-09-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | npj Digital Medicine |
| spelling | doaj.art-6852a8b842cb4d289b750afc458515bb2023-12-03T08:22:40ZengNature Portfolionpj Digital Medicine2398-63522021-09-01411910.1038/s41746-021-00504-6Artificial intelligence sepsis prediction algorithm learns to say “I don’t know”Supreeth P. Shashikumar0Gabriel Wardi1Atul Malhotra2Shamim Nemati3Division of Biomedical Informatics, University of California San DiegoDepartment of Emergency Medicine, University of California San DiegoDivision of Pulmonary, Critical Care and Sleep Medicine, University of California San DiegoDivision of Biomedical Informatics, University of California San DiegoAbstract Sepsis is a leading cause of morbidity and mortality worldwide. Early identification of sepsis is important as it allows timely administration of potentially life-saving resuscitation and antimicrobial therapy. We present COMPOSER (COnformal Multidimensional Prediction Of SEpsis Risk), a deep learning model for the early prediction of sepsis, specifically designed to reduce false alarms by detecting unfamiliar patients/situations arising from erroneous data, missingness, distributional shift and data drifts. COMPOSER flags these unfamiliar cases as indeterminate rather than making spurious predictions. Six patient cohorts (515,720 patients) curated from two healthcare systems in the United States across intensive care units (ICU) and emergency departments (ED) were used to train and externally and temporally validate this model. In a sequential prediction setting, COMPOSER achieved a consistently high area under the curve (AUC) (ICU: 0.925–0.953; ED: 0.938–0.945). Out of over 6 million prediction windows roughly 20% and 8% were identified as indeterminate amongst non-septic and septic patients, respectively. COMPOSER provided early warning within a clinically actionable timeframe (ICU: 12.2 [3.2 22.8] and ED: 2.1 [0.8 4.5] hours prior to first antibiotics order) across all six cohorts, thus allowing for identification and prioritization of patients at high risk for sepsis.https://doi.org/10.1038/s41746-021-00504-6 |
| spellingShingle | Supreeth P. Shashikumar Gabriel Wardi Atul Malhotra Shamim Nemati Artificial intelligence sepsis prediction algorithm learns to say “I don’t know” npj Digital Medicine |
| title | Artificial intelligence sepsis prediction algorithm learns to say “I don’t know” |
| title_full | Artificial intelligence sepsis prediction algorithm learns to say “I don’t know” |
| title_fullStr | Artificial intelligence sepsis prediction algorithm learns to say “I don’t know” |
| title_full_unstemmed | Artificial intelligence sepsis prediction algorithm learns to say “I don’t know” |
| title_short | Artificial intelligence sepsis prediction algorithm learns to say “I don’t know” |
| title_sort | artificial intelligence sepsis prediction algorithm learns to say i don t know |
| url | https://doi.org/10.1038/s41746-021-00504-6 |
| work_keys_str_mv | AT supreethpshashikumar artificialintelligencesepsispredictionalgorithmlearnstosayidontknow AT gabrielwardi artificialintelligencesepsispredictionalgorithmlearnstosayidontknow AT atulmalhotra artificialintelligencesepsispredictionalgorithmlearnstosayidontknow AT shamimnemati artificialintelligencesepsispredictionalgorithmlearnstosayidontknow |