Gradient Biotech
U

Menu

Research Areas

Computational BiologySingle-cell, bulk RNA-seq, spatial, and biomarkers
CardiologyCardiovascular signals, cohorts, and interpretationImmunologyImmune cell states, signaling networks, and repertoire dynamicsNeurologyBrain connectivity, functional imaging, and electrophysiologyOncologyTumor microenvironment, immune response, mutations, and survivalPathologyWhole-slide imaging, tissue quantification, and spatial pathology

Application

Documentation

Documentation

Back to studies

Quick start

This walkthrough covers a single ECG recording from upload to interpretation.

1. Create a study

Click Create study. Enter a name (for example ECG demo) and optional description.

2. Upload a recording

Open Upload and add either:

  • CSV — one file with time (seconds) and ECG amplitude columns
  • WFDB — .hea and .dat together (PhysioNet-style)

Generate a sample CSV locally:

conda activate biochem
python products/cardiology/experiments/example01/generate_ecg_csv.py

Upload the output file from products/cardiology/experiments/example01/.

3. Run preprocessing

Open the dataset from the study overview. On the Overview tab, click Run preprocessing. Wait for the job to complete — watch the Jobs panel in the navbar.

4. Compute HRV

When preprocessing completes, click Compute HRV on the same tab. Review time-domain metrics (SDNN, RMSSD, mean HR) and frequency metrics (LF, HF, LF/HF) in the HRV panel.

5. Explore the signal

Switch to the Explorer tab. Pan and zoom the downsampled waveform. R-peak markers overlay the trace; low-quality segments may be highlighted when SQI is poor.

6. Interpret (optional)

Open the HRV run from Runs or the recent-runs panel. On the run detail page, click Interpret for a plain-language summary that cites your computed metrics.

Requires Ollama configuration in backend-analysis — see the repo root ai.md.

Next steps