# Sepsis prediction on Real-world ICU datasets

Source: [Sepsis prediction on Real-world ICU datasets](<https://www.sota2.com/research/sota/sepsis-prediction-on-real-world-icu-datasets>)

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

Data updated: 2026-03-18T06:41:19.350Z

Evaluation of models for Sepsis prediction using a collection of real-world Intensive Care Unit (ICU) datasets, assessed across clinical efficacy, privacy protection, communication efficiency, and scalability.

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 | Sepsis prediction |
| Dataset (including any stated split) | Real-world ICU datasets |
| Metric | Clinical Efficacy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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

- [Clinical Efficacy](<https://www.sota2.com/research/sota/sepsis-prediction-on-real-world-icu-datasets.md?metric=Clinical%20Efficacy>): Higher is better; unit: PERCENT.

- [Privacy Protection Score](<https://www.sota2.com/research/sota/sepsis-prediction-on-real-world-icu-datasets.md?metric=Privacy%20Protection%20Score>): Higher is better; unit: PERCENT.

- [Communication Efficiency](<https://www.sota2.com/research/sota/sepsis-prediction-on-real-world-icu-datasets.md?metric=Communication%20Efficiency>): Higher is better; unit: PERCENT.

- [Scalability Score](<https://www.sota2.com/research/sota/sepsis-prediction-on-real-world-icu-datasets.md?metric=Scalability%20Score>): Higher is better; unit: PERCENT.

- [Overall Score](<https://www.sota2.com/research/sota/sepsis-prediction-on-real-world-icu-datasets.md?metric=Overall%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Clinical Efficacy | Result date | Report paper |
| --- | --- | --- | --- |
| Our Full Method | 94 | Not recorded | 69ba316b0636e5bf4afc05f2 |
| Temporal FL | 86 | Not recorded | 69ba316b0636e5bf4afc05f2 |
| Knowledge-Enhanced FL | 83 | Not recorded | 69ba316b0636e5bf4afc05f2 |
| Standard FL | 79 | Not recorded | 69ba316b0636e5bf4afc05f2 |
| Centralized LSTM | 72 | Not recorded | 69ba316b0636e5bf4afc05f2 |

### Our Full Method

Clinical Efficacy: 94. Result ID: 69ba482547f34ce924a374ce.

[Reporting paper](<https://www.sota2.com/research/paper/69ba316b0636e5bf4afc05f2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ba316b0636e5bf4afc05f2.md>)

Recorded evaluation configuration: {"Learning Paradigm":"Federated","Components":"Knowledge Graph + Temporal Transformer + Meta-learning"}. This record does not assert independent reproduction.

### Temporal FL

Clinical Efficacy: 86. Result ID: 69ba482547f34ce924a374cd.

[Reporting paper](<https://www.sota2.com/research/paper/69ba316b0636e5bf4afc05f2.md>)

Recorded evaluation configuration: {"Learning Paradigm":"Federated","Components":"Temporal Transformer"}. This record does not assert independent reproduction.

### Knowledge-Enhanced FL

Clinical Efficacy: 83. Result ID: 69ba482547f34ce924a374cc.

[Reporting paper](<https://www.sota2.com/research/paper/69ba316b0636e5bf4afc05f2.md>)

Recorded evaluation configuration: {"Learning Paradigm":"Federated","Components":"Knowledge Graph"}. This record does not assert independent reproduction.

### Standard FL

Clinical Efficacy: 79. Result ID: 69ba482547f34ce924a374cb.

[Reporting paper](<https://www.sota2.com/research/paper/69ba316b0636e5bf4afc05f2.md>)

Recorded evaluation configuration: {"Learning Paradigm":"Federated"}. This record does not assert independent reproduction.

### Centralized LSTM

Clinical Efficacy: 72. Result ID: 69ba482547f34ce924a374ca.

[Reporting paper](<https://www.sota2.com/research/paper/69ba316b0636e5bf4afc05f2.md>)

Recorded evaluation configuration: {"Learning Paradigm":"Centralized"}. This record does not assert independent reproduction.

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