# Per-patient AHI Class Prediction on Stroke Unit 1.0 (All patients)

Source: [Per-patient AHI Class Prediction on Stroke Unit 1.0 (All patients)](<https://www.sota2.com/research/sota/per-patient-ahi-class-prediction-on-stroke-unit-1-0-all-patients>)

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

Data updated: 2026-02-26T05:26:56.070Z

Evaluation of models for predicting patient AHI class using the Stroke Unit 1.0 (All patients) dataset, reporting metrics such as Acc@1 and Acc@2.

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 | Per-patient AHI Class Prediction |
| Dataset (including any stated split) | Stroke Unit 1.0 (All patients) |
| Metric | Acc@1 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [Acc@1](<https://www.sota2.com/research/sota/per-patient-ahi-class-prediction-on-stroke-unit-1-0-all-patients.md?metric=Acc%401>): Higher is better; unit: PERCENT.

- [Acc@2](<https://www.sota2.com/research/sota/per-patient-ahi-class-prediction-on-stroke-unit-1-0-all-patients.md?metric=Acc%402>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Acc@1 | Result date | Report paper |
| --- | --- | --- | --- |
| CNN + LSTM | 66.6 | Not recorded | 68fa6e901e209131cea40a57 |
| CNN + LSTM | 43.3 | Not recorded | 68fa6e901e209131cea40a57 |
| LSTM | 30 | Not recorded | 68fa6e901e209131cea40a57 |
| ResNet | 23.3 | Not recorded | 68fa6e901e209131cea40a57 |

### CNN + LSTM

Acc@1: 66.6. Result ID: 699035743a6c314849feeb71.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e901e209131cea40a57.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e901e209131cea40a57.md>)

[Implementation](<https://github.com/dslab-uniud/OSAS>)

Recorded evaluation configuration: {"Model":"CNN + LSTM","Signal":"ECG + SpO2","Evaluation protocol":"Leave-one-patient-out CV"}. This record does not assert independent reproduction.

### CNN + LSTM

Acc@1: 43.3. Result ID: 699035743a6c314849feeb70.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e901e209131cea40a57.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e901e209131cea40a57.md>)

[Implementation](<https://github.com/dslab-uniud/OSAS>)

Recorded evaluation configuration: {"Model":"CNN + LSTM","Signal":"ECG","Evaluation protocol":"Leave-one-patient-out CV"}. This record does not assert independent reproduction.

### LSTM

Acc@1: 30. Result ID: 699035743a6c314849feeb6f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e901e209131cea40a57.md>)

Recorded evaluation configuration: {"Model":"LSTM","Signal":"SpO2","Evaluation protocol":"Leave-one-patient-out CV"}. This record does not assert independent reproduction.

### ResNet

Acc@1: 23.3. Result ID: 699035743a6c314849feeb6e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e901e209131cea40a57.md>)

Recorded evaluation configuration: {"Model":"ResNet","Signal":"ECG","Evaluation protocol":"Leave-one-patient-out CV"}. This record does not assert independent reproduction.

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