How Consistent are Clinicians? Evaluating the Predictability of Sepsis Disease Progression with Dynamics Models
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914748173385728 |
|---|---|
| author | Park, Unnseo Sivaraman, Venkatesh Perer, Adam |
| author_facet | Park, Unnseo Sivaraman, Venkatesh Perer, Adam |
| contents | Reinforcement learning (RL) is a promising approach to generate treatment policies for sepsis patients in intensive care. While retrospective evaluation metrics show decreased mortality when these policies are followed, studies with clinicians suggest their recommendations are often spurious. We propose that these shortcomings may be due to lack of diversity in observed actions and outcomes in the training data, and we construct experiments to investigate the feasibility of predicting sepsis disease severity changes due to clinician actions. Preliminary results suggest incorporating action information does not significantly improve model performance, indicating that clinician actions may not be sufficiently variable to yield measurable effects on disease progression. We discuss the implications of these findings for optimizing sepsis treatment. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_07148 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | How Consistent are Clinicians? Evaluating the Predictability of Sepsis Disease Progression with Dynamics Models Park, Unnseo Sivaraman, Venkatesh Perer, Adam Machine Learning Human-Computer Interaction Reinforcement learning (RL) is a promising approach to generate treatment policies for sepsis patients in intensive care. While retrospective evaluation metrics show decreased mortality when these policies are followed, studies with clinicians suggest their recommendations are often spurious. We propose that these shortcomings may be due to lack of diversity in observed actions and outcomes in the training data, and we construct experiments to investigate the feasibility of predicting sepsis disease severity changes due to clinician actions. Preliminary results suggest incorporating action information does not significantly improve model performance, indicating that clinician actions may not be sufficiently variable to yield measurable effects on disease progression. We discuss the implications of these findings for optimizing sepsis treatment. |
| title | How Consistent are Clinicians? Evaluating the Predictability of Sepsis Disease Progression with Dynamics Models |
| topic | Machine Learning Human-Computer Interaction |
| url | https://arxiv.org/abs/2404.07148 |