# ICD-9 Code Prediction on MIMIC-III 8922 labels (full)

Source: [ICD-9 Code Prediction on MIMIC-III 8922 labels (full)](<https://www.sota2.com/research/sota/icd-9-code-prediction-on-mimic-iii-8922-labels-full>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:25:24.723Z

Evaluation of model performance on predicting ICD-9 codes using the full set of 8922 labels derived from the MIMIC-III clinical database, typically measured using various F1 scores (Macro, Micro, Diagnosis, Procedure) and Precision at K (P@8, P@15).

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 | ICD-9 code prediction |
| Dataset (including any stated split) | MIMIC-III 8922 labels (full) |
| Metric | AUC Macro |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 18 of 18 |
| 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

- [AUC Macro](<https://www.sota2.com/research/sota/icd-9-code-prediction-on-mimic-iii-8922-labels-full.md?metric=AUC%20Macro>): Higher is better; unit: SCALAR.

- [AUC Micro](<https://www.sota2.com/research/sota/icd-9-code-prediction-on-mimic-iii-8922-labels-full.md?metric=AUC%20Micro>): Higher is better; unit: SCALAR.

- [F1 Macro](<https://www.sota2.com/research/sota/icd-9-code-prediction-on-mimic-iii-8922-labels-full.md?metric=F1%20Macro>): Higher is better; unit: SCALAR.

- [F1 Micro](<https://www.sota2.com/research/sota/icd-9-code-prediction-on-mimic-iii-8922-labels-full.md?metric=F1%20Micro>): Higher is better; unit: SCALAR.

- [F1 Diag](<https://www.sota2.com/research/sota/icd-9-code-prediction-on-mimic-iii-8922-labels-full.md?metric=F1%20Diag>): Higher is better; unit: SCALAR.

- [F1 Proc](<https://www.sota2.com/research/sota/icd-9-code-prediction-on-mimic-iii-8922-labels-full.md?metric=F1%20Proc>): Higher is better; unit: SCALAR.

- [P@8](<https://www.sota2.com/research/sota/icd-9-code-prediction-on-mimic-iii-8922-labels-full.md?metric=P%408>): Higher is better; unit: SCALAR.

- [P@15](<https://www.sota2.com/research/sota/icd-9-code-prediction-on-mimic-iii-8922-labels-full.md?metric=P%4015>): Higher is better; unit: SCALAR.

- [Macro P](<https://www.sota2.com/research/sota/icd-9-code-prediction-on-mimic-iii-8922-labels-full.md?metric=Macro%20P>): Higher is better; unit: SCALAR.

- [Macro R](<https://www.sota2.com/research/sota/icd-9-code-prediction-on-mimic-iii-8922-labels-full.md?metric=Macro%20R>): Higher is better; unit: SCALAR.

- [Micro P](<https://www.sota2.com/research/sota/icd-9-code-prediction-on-mimic-iii-8922-labels-full.md?metric=Micro%20P>): Higher is better; unit: SCALAR.

- [Micro R](<https://www.sota2.com/research/sota/icd-9-code-prediction-on-mimic-iii-8922-labels-full.md?metric=Micro%20R>): Higher is better; unit: SCALAR.

- [Micro F1-diag](<https://www.sota2.com/research/sota/icd-9-code-prediction-on-mimic-iii-8922-labels-full.md?metric=Micro%20F1-diag>): Higher is better; unit: SCALAR.

- [Micro F1-proc](<https://www.sota2.com/research/sota/icd-9-code-prediction-on-mimic-iii-8922-labels-full.md?metric=Micro%20F1-proc>): Higher is better; unit: SCALAR.

## Recorded results

| Method | AUC Macro | Result date | Report paper |
| --- | --- | --- | --- |
| MultiResCNN | 0.91 | Not recorded | 68fa66cb1e209131cea12551 |
| CNN+att | 0.902 | Not recorded | 68fa6a621e209131cea27e97 |
| DR-CAML | 0.897 | Not recorded | 68fa644a1e209131cea03c15 |
| DR-CAML | 0.897 | Not recorded | 68fa66cb1e209131cea12551 |
| CAML | 0.895 | Not recorded | 68fa644a1e209131cea03c15 |
| CAML | 0.895 | Not recorded | 68fa66cb1e209131cea12551 |
| CNN+att | 0.886 | Not recorded | 68fa6a621e209131cea27e97 |
| HAN (HLAN) | 0.885 | Not recorded | 68fa6a621e209131cea27e97 |
| HAN (HLAN) | 0.882 | Not recorded | 68fa6a621e209131cea27e97 |
| HAN (HLAN) | 0.874 | Not recorded | 68fa6a621e209131cea27e97 |
| Bi-GRU | 0.849 | Not recorded | 68fa6a621e209131cea27e97 |
| Bi-GRU | 0.835 | Not recorded | 68fa6a621e209131cea27e97 |
| CNN | 0.824 | Not recorded | 68fa6a621e209131cea27e97 |
| Bi-GRU | 0.822 | Not recorded | 68fa644a1e209131cea03c15 |
| CNN | 0.818 | Not recorded | 68fa6a621e209131cea27e97 |
| CNN | 0.806 | Not recorded | 68fa644a1e209131cea03c15 |
| Logistic Regression | 0.561 | Not recorded | 68fa644a1e209131cea03c15 |

### MultiResCNN

AUC Macro: 0.91. Result ID: 698d07618b517eb4331e7794.

[Reporting paper](<https://www.sota2.com/research/paper/68fa66cb1e209131cea12551.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66cb1e209131cea12551.md>)

[Implementation](<https://github.com/foxlf823/Multi-Filter-Residual-Convolutional-Neural-Network>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### CNN+att

AUC Macro: 0.902. Result ID: 698e3152ffc3901de09b8b5a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a621e209131cea27e97.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a621e209131cea27e97.md>)

[Implementation](<https://github.com/acadTags/Explainable-Automated-Medical-Coding>)

Recorded evaluation configuration: {"label embedding initialization":true}. This record does not assert independent reproduction.

### DR-CAML

AUC Macro: 0.897. Result ID: 698c2ac0112aeb9460efe7df.

[Reporting paper](<https://www.sota2.com/research/paper/68fa644a1e209131cea03c15.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa644a1e209131cea03c15.md>)

[Implementation](<https://www.sota2.com/github.com/jamesmullenbach/caml-mimic>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### DR-CAML

AUC Macro: 0.897. Result ID: 698d07618b517eb4331e7793.

[Reporting paper](<https://www.sota2.com/research/paper/68fa66cb1e209131cea12551.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa644a1e209131cea03c15.md>)

[Implementation](<https://www.sota2.com/github.com/jamesmullenbach/caml-mimic>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### CAML

AUC Macro: 0.895. Result ID: 698c2ac0112aeb9460efe7de.

[Reporting paper](<https://www.sota2.com/research/paper/68fa644a1e209131cea03c15.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa644a1e209131cea03c15.md>)

[Implementation](<https://www.sota2.com/github.com/jamesmullenbach/caml-mimic>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### CAML

AUC Macro: 0.895. Result ID: 698d07618b517eb4331e7792.

[Reporting paper](<https://www.sota2.com/research/paper/68fa66cb1e209131cea12551.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa644a1e209131cea03c15.md>)

[Implementation](<https://www.sota2.com/github.com/jamesmullenbach/caml-mimic>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### CNN+att

AUC Macro: 0.886. Result ID: 698e3152ffc3901de09b8b59.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a621e209131cea27e97.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a621e209131cea27e97.md>)

[Implementation](<https://github.com/acadTags/Explainable-Automated-Medical-Coding>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### HAN (HLAN)

AUC Macro: 0.885. Result ID: 698e3152ffc3901de09b8b5b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a621e209131cea27e97.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a621e209131cea27e97.md>)

[Implementation](<https://github.com/acadTags/Explainable-Automated-Medical-Coding>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### HAN (HLAN)

AUC Macro: 0.882. Result ID: 698e3152ffc3901de09b8b5c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a621e209131cea27e97.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a621e209131cea27e97.md>)

[Implementation](<https://github.com/acadTags/Explainable-Automated-Medical-Coding>)

Recorded evaluation configuration: {"label embedding initialization":true}. This record does not assert independent reproduction.

### HAN (HLAN)

AUC Macro: 0.874. Result ID: 698e3152ffc3901de09b8b5d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a621e209131cea27e97.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a621e209131cea27e97.md>)

[Implementation](<https://github.com/acadTags/Explainable-Automated-Medical-Coding>)

Recorded evaluation configuration: {"sentence split":true}. This record does not assert independent reproduction.

### Bi-GRU

AUC Macro: 0.849. Result ID: 698e3152ffc3901de09b8b58.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a621e209131cea27e97.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a621e209131cea27e97.md>)

[Implementation](<https://github.com/acadTags/Explainable-Automated-Medical-Coding>)

Recorded evaluation configuration: {"label embedding initialization":true}. This record does not assert independent reproduction.

### Bi-GRU

AUC Macro: 0.835. Result ID: 698e3152ffc3901de09b8b57.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a621e209131cea27e97.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a621e209131cea27e97.md>)

[Implementation](<https://github.com/acadTags/Explainable-Automated-Medical-Coding>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### CNN

AUC Macro: 0.824. Result ID: 698e3152ffc3901de09b8b56.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a621e209131cea27e97.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a621e209131cea27e97.md>)

[Implementation](<https://github.com/acadTags/Explainable-Automated-Medical-Coding>)

Recorded evaluation configuration: {"label embedding initialization":true}. This record does not assert independent reproduction.

### Bi-GRU

AUC Macro: 0.822. Result ID: 698c2ac0112aeb9460efe7dd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa644a1e209131cea03c15.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### CNN

AUC Macro: 0.818. Result ID: 698e3152ffc3901de09b8b55.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a621e209131cea27e97.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a621e209131cea27e97.md>)

[Implementation](<https://github.com/acadTags/Explainable-Automated-Medical-Coding>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### CNN

AUC Macro: 0.806. Result ID: 698c2ac0112aeb9460efe7dc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa644a1e209131cea03c15.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa617c1e209131ce9f4631.md>)

[Implementation](<https://code.google.com/p/word2vec/>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Logistic Regression

AUC Macro: 0.561. Result ID: 698c2ac0112aeb9460efe7db.

[Reporting paper](<https://www.sota2.com/research/paper/68fa644a1e209131cea03c15.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.