Data-driven distillation and precision prognosis in traumatic brain injury with interpretable machine learning
Abstract Traumatic brain injury (TBI) affects how the brain functions in the short and long term. Resulting patient outcomes across physical, cognitive, and psychological domains are complex and often difficult to predict. Major challenges to developing personalized treatment for TBI include distill...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2023-12-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-023-48054-z |
| _version_ | 1797415103093538816 |
|---|---|
| author | Andrew Tritt John K. Yue Adam R. Ferguson Abel Torres Espin Lindsay D. Nelson Esther L. Yuh Amy J. Markowitz Geoffrey T. Manley Kristofer E. Bouchard the TRACK-TBI Investigators |
| author_facet | Andrew Tritt John K. Yue Adam R. Ferguson Abel Torres Espin Lindsay D. Nelson Esther L. Yuh Amy J. Markowitz Geoffrey T. Manley Kristofer E. Bouchard the TRACK-TBI Investigators |
| author_sort | Andrew Tritt |
| collection | DOAJ |
| description | Abstract Traumatic brain injury (TBI) affects how the brain functions in the short and long term. Resulting patient outcomes across physical, cognitive, and psychological domains are complex and often difficult to predict. Major challenges to developing personalized treatment for TBI include distilling large quantities of complex data and increasing the precision with which patient outcome prediction (prognoses) can be rendered. We developed and applied interpretable machine learning methods to TBI patient data. We show that complex data describing TBI patients' intake characteristics and outcome phenotypes can be distilled to smaller sets of clinically interpretable latent factors. We demonstrate that 19 clusters of TBI outcomes can be predicted from intake data, a ~ 6× improvement in precision over clinical standards. Finally, we show that 36% of the outcome variance across patients can be predicted. These results demonstrate the importance of interpretable machine learning applied to deeply characterized patients for data-driven distillation and precision prognosis. |
| first_indexed | 2024-03-09T05:43:12Z |
| format | Article |
| id | doaj.art-75b5b33a7a9c4decb1237d41001a7084 |
| institution | Directory Open Access Journal |
| issn | 2045-2322 |
| language | English |
| last_indexed | 2024-03-09T05:43:12Z |
| publishDate | 2023-12-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj.art-75b5b33a7a9c4decb1237d41001a70842023-12-03T12:23:14ZengNature PortfolioScientific Reports2045-23222023-12-0113111610.1038/s41598-023-48054-zData-driven distillation and precision prognosis in traumatic brain injury with interpretable machine learningAndrew Tritt0John K. Yue1Adam R. Ferguson2Abel Torres Espin3Lindsay D. Nelson4Esther L. Yuh5Amy J. Markowitz6Geoffrey T. Manley7Kristofer E. Bouchard8the TRACK-TBI InvestigatorsApplied Math and Computational Research Division, Lawrence Berkeley National LaboratoryBrain and Spinal Injury Center, Zuckerberg San Francisco General Hospital and Trauma CenterBrain and Spinal Injury Center, Zuckerberg San Francisco General Hospital and Trauma CenterBrain and Spinal Injury Center, Zuckerberg San Francisco General Hospital and Trauma CenterDepartments of Neurosurgery and Neurology, Medical College of WisconsinBrain and Spinal Injury Center, Zuckerberg San Francisco General Hospital and Trauma CenterBrain and Spinal Injury Center, Zuckerberg San Francisco General Hospital and Trauma CenterBrain and Spinal Injury Center, Zuckerberg San Francisco General Hospital and Trauma CenterWeill Neurohub, University of California BerkeleyAbstract Traumatic brain injury (TBI) affects how the brain functions in the short and long term. Resulting patient outcomes across physical, cognitive, and psychological domains are complex and often difficult to predict. Major challenges to developing personalized treatment for TBI include distilling large quantities of complex data and increasing the precision with which patient outcome prediction (prognoses) can be rendered. We developed and applied interpretable machine learning methods to TBI patient data. We show that complex data describing TBI patients' intake characteristics and outcome phenotypes can be distilled to smaller sets of clinically interpretable latent factors. We demonstrate that 19 clusters of TBI outcomes can be predicted from intake data, a ~ 6× improvement in precision over clinical standards. Finally, we show that 36% of the outcome variance across patients can be predicted. These results demonstrate the importance of interpretable machine learning applied to deeply characterized patients for data-driven distillation and precision prognosis.https://doi.org/10.1038/s41598-023-48054-z |
| spellingShingle | Andrew Tritt John K. Yue Adam R. Ferguson Abel Torres Espin Lindsay D. Nelson Esther L. Yuh Amy J. Markowitz Geoffrey T. Manley Kristofer E. Bouchard the TRACK-TBI Investigators Data-driven distillation and precision prognosis in traumatic brain injury with interpretable machine learning Scientific Reports |
| title | Data-driven distillation and precision prognosis in traumatic brain injury with interpretable machine learning |
| title_full | Data-driven distillation and precision prognosis in traumatic brain injury with interpretable machine learning |
| title_fullStr | Data-driven distillation and precision prognosis in traumatic brain injury with interpretable machine learning |
| title_full_unstemmed | Data-driven distillation and precision prognosis in traumatic brain injury with interpretable machine learning |
| title_short | Data-driven distillation and precision prognosis in traumatic brain injury with interpretable machine learning |
| title_sort | data driven distillation and precision prognosis in traumatic brain injury with interpretable machine learning |
| url | https://doi.org/10.1038/s41598-023-48054-z |
| work_keys_str_mv | AT andrewtritt datadrivendistillationandprecisionprognosisintraumaticbraininjurywithinterpretablemachinelearning AT johnkyue datadrivendistillationandprecisionprognosisintraumaticbraininjurywithinterpretablemachinelearning AT adamrferguson datadrivendistillationandprecisionprognosisintraumaticbraininjurywithinterpretablemachinelearning AT abeltorresespin datadrivendistillationandprecisionprognosisintraumaticbraininjurywithinterpretablemachinelearning AT lindsaydnelson datadrivendistillationandprecisionprognosisintraumaticbraininjurywithinterpretablemachinelearning AT estherlyuh datadrivendistillationandprecisionprognosisintraumaticbraininjurywithinterpretablemachinelearning AT amyjmarkowitz datadrivendistillationandprecisionprognosisintraumaticbraininjurywithinterpretablemachinelearning AT geoffreytmanley datadrivendistillationandprecisionprognosisintraumaticbraininjurywithinterpretablemachinelearning AT kristoferebouchard datadrivendistillationandprecisionprognosisintraumaticbraininjurywithinterpretablemachinelearning AT thetracktbiinvestigators datadrivendistillationandprecisionprognosisintraumaticbraininjurywithinterpretablemachinelearning |