# ICU Transfer Prediction on MIMIC-III Source: [ICU Transfer Prediction on MIMIC-III]() Collection index: [llms.txt]() Data updated: 2026-06-17T07:01:19.643Z Evaluation of models predicting patient transfer to the Intensive Care Unit (ICU) using the MIMIC-III clinical database, often evaluated using information-theoretic metrics like Mutual Information. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | ICU transfer prediction (ICU) | | Dataset (including any stated split) | MIMIC-III | | Metric | MI (x10^3) | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 16 of 16 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [MI (x10^3)](): Higher is better; unit: SCALAR. ## Recorded results | Method | MI (x10^3) | Result date | Report paper | | --- | --- | --- | --- | | DeepSets | 32.42 | Not recorded | 6a322a39d33dfb23e783ec5b | | Deep MLP Bag-of-Codes | 29.89 | Not recorded | 6a322a39d33dfb23e783ec5b | | Set Transformer | 29.64 | Not recorded | 6a322a39d33dfb23e783ec5b | | Star-GAT | 29.28 | Not recorded | 6a322a39d33dfb23e783ec5b | | FusedLogits LR | 28.19 | Not recorded | 6a322a39d33dfb23e783ec5b | | Factorization Machine | 27.58 | Not recorded | 6a322a39d33dfb23e783ec5b | | Complement Naive Bayes | 27.53 | Not recorded | 6a322a39d33dfb23e783ec5b | | Deep and Cross Network | 27.2 | Not recorded | 6a322a39d33dfb23e783ec5b | | MLCI | 26.42 | Not recorded | 6a322a39d33dfb23e783ec5b | | Pure MIL Attention Pooling | 26.14 | Not recorded | 6a322a39d33dfb23e783ec5b | | Gradient Boosted Trees | 24.55 | Not recorded | 6a322a39d33dfb23e783ec5b | | Multinomial Naive Bayes | 22.29 | Not recorded | 6a322a39d33dfb23e783ec5b | | kNN | 20.65 | Not recorded | 6a322a39d33dfb23e783ec5b | | CCI+ECI | 16.52 | Not recorded | 6a322a39d33dfb23e783ec5b | | Elixhauser | 14.72 | Not recorded | 6a322a39d33dfb23e783ec5b | | Charlson | 14.6 | Not recorded | 6a322a39d33dfb23e783ec5b | ### DeepSets MI (x10^3): 32.42. Result ID: 6a324584fbc373592c2f3c64. [Reporting paper]() Recorded evaluation configuration: {"Pooling":"mean ⊕ max pool"}. This record does not assert independent reproduction. ### Deep MLP Bag-of-Codes MI (x10^3): 29.89. Result ID: 6a324584fbc373592c2f3c60. [Reporting paper]() Recorded evaluation configuration: {"Implementation":"EmbeddingBag"}. This record does not assert independent reproduction. ### Set Transformer MI (x10^3): 29.64. Result ID: 6a324584fbc373592c2f3c63. [Reporting paper]() Recorded evaluation configuration: {"Mode":"single-logit"}. This record does not assert independent reproduction. ### Star-GAT MI (x10^3): 29.28. Result ID: 6a324584fbc373592c2f3c61. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### FusedLogits LR MI (x10^3): 28.19. Result ID: 6a324584fbc373592c2f3c5c. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Factorization Machine MI (x10^3): 27.58. Result ID: 6a324584fbc373592c2f3c5d. [Reporting paper]() Recorded evaluation configuration: {"Metric":"FM score"}. This record does not assert independent reproduction. ### Complement Naive Bayes MI (x10^3): 27.53. Result ID: 6a324584fbc373592c2f3c5b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Deep and Cross Network MI (x10^3): 27.2. Result ID: 6a324584fbc373592c2f3c5f. [Reporting paper]() Recorded evaluation configuration: {"Acronym":"DCN"}. This record does not assert independent reproduction. ### MLCI MI (x10^3): 26.42. Result ID: 6a324584fbc373592c2f3c65. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Pure MIL Attention Pooling MI (x10^3): 26.14. Result ID: 6a324584fbc373592c2f3c62. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Gradient Boosted Trees MI (x10^3): 24.55. Result ID: 6a324584fbc373592c2f3c5e. [Reporting paper]() Recorded evaluation configuration: {"Evaluation":"Score/Logit"}. This record does not assert independent reproduction. ### Multinomial Naive Bayes MI (x10^3): 22.29. Result ID: 6a324584fbc373592c2f3c5a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### kNN MI (x10^3): 20.65. Result ID: 6a324584fbc373592c2f3c59. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### CCI+ECI MI (x10^3): 16.52. Result ID: 6a324584fbc373592c2f3c58. [Reporting paper]() Recorded evaluation configuration: {"Mode":"scalar"}. This record does not assert independent reproduction. ### Elixhauser MI (x10^3): 14.72. Result ID: 6a324584fbc373592c2f3c57. [Reporting paper]() Recorded evaluation configuration: {"Acronym":"ECI"}. This record does not assert independent reproduction. ### Charlson MI (x10^3): 14.6. Result ID: 6a324584fbc373592c2f3c56. [Reporting paper]() Recorded evaluation configuration: {"Acronym":"CCI"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.