Abstract | ||
---|---|---|
•Introduce Learned Binary Masks (LBM) to interpret an RNN’s ICU mortality predictions.•Attribute individual RNN predictions to their input features using LBM & KernelSHAP.•Aggregate attributions from KernelSHAP and LBM to interpret the RNN at various scales.•Introduce a patient data representation that facilitates use of LBM and KernelSHAP. |
Year | DOI | Venue |
---|---|---|
2021 | 10.1016/j.jbi.2021.103672 | Journal of Biomedical Informatics |
Keywords | DocType | Volume |
Model interpretation,Recurrent neural networks,Feature importance,Feature attribution,Electronic medical records,Deep learning | Journal | 114 |
ISSN | Citations | PageRank |
1532-0464 | 1 | 0.36 |
References | Authors | |
0 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Long V Ho | 1 | 1 | 0.36 |
Melissa Aczon | 2 | 1 | 1.37 |
David R Ledbetter | 3 | 2 | 1.08 |
Randall C. Wetzel | 4 | 182 | 11.24 |