Personalised Basal Tuner (PBT) - Paper 2 Peaks Pipeline
Fuente:
Zenodo
Guardado en:
| Autor principal: | |
|---|---|
| Formato: | Recurso digital |
| Publicado: |
Zenodo
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866901919658672128 |
|---|---|
| author | Gasca García, Daniel |
| author_facet | Gasca García, Daniel |
| contents | <p>This deposit contains the exact code for the PBT — Paper 2 Peaks Pipeline. It parses and harmonises CGM data, applies low-pass filtering, detects and removes meal-related peaks (using prominence/width at the 95th percentile), handles windows crossing midnight, builds 24-h composites, and computes baseline-subtracted hourly relative change. It exports filtered daily tables, hourly stacks, last values per hour, a 24-h boxplot table, 24-h medians with a reliability flag, and a final merge without activity. The pipeline targets Python ≥ 3.10 with pandas/numpy/scipy/matplotlib/seaborn (see requirements.txt) and can be executed with a single runner script (17.ScriptforTestExperiment2.py).</p> <p>Dependency: This pipeline requires Paper 1 artifacts with the same `id`: `Boxplot{id}0-24total.csv` and `BasalSimulated{id}.csv`. Generate them by running Paper 1 or download the versioned release from Zenodo: <a target="_new" rel="noopener">https://doi.org/10.5281/zenodo.17392921</a>. Set `globals.path2` to the directory that contains these files.</p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_17393515 |
| institution | Zenodo |
| language | |
| publishDate | 2025 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | Personalised Basal Tuner (PBT) - Paper 2 Peaks Pipeline Gasca García, Daniel diabetes t1dm insulin cgm peaks python reproducibility <p>This deposit contains the exact code for the PBT — Paper 2 Peaks Pipeline. It parses and harmonises CGM data, applies low-pass filtering, detects and removes meal-related peaks (using prominence/width at the 95th percentile), handles windows crossing midnight, builds 24-h composites, and computes baseline-subtracted hourly relative change. It exports filtered daily tables, hourly stacks, last values per hour, a 24-h boxplot table, 24-h medians with a reliability flag, and a final merge without activity. The pipeline targets Python ≥ 3.10 with pandas/numpy/scipy/matplotlib/seaborn (see requirements.txt) and can be executed with a single runner script (17.ScriptforTestExperiment2.py).</p> <p>Dependency: This pipeline requires Paper 1 artifacts with the same `id`: `Boxplot{id}0-24total.csv` and `BasalSimulated{id}.csv`. Generate them by running Paper 1 or download the versioned release from Zenodo: <a target="_new" rel="noopener">https://doi.org/10.5281/zenodo.17392921</a>. Set `globals.path2` to the directory that contains these files.</p> |
| title | Personalised Basal Tuner (PBT) - Paper 2 Peaks Pipeline |
| topic | diabetes t1dm insulin cgm peaks python reproducibility |
| url | https://doi.org/10.5281/zenodo.17393515 |