fyron 0.2.4

Fyron is an open-source Python toolkit for interoperable healthcare data and AI workflows. It provides unified access to FHIR data via REST APIs and relational (SQL-backed) FHIR servers, integrates DICOM imaging sources, and enables semantic exploration of clinical narratives using modern language models.

Fyron

Fyron is a pragmatic Python toolkit for interoperable healthcare data and AI workflows. It provides clean, testable primitives for FHIR, DICOM, imaging, documents, cohort construction, survival analysis, tabular machine learning, LLM-assisted analysis, and BOA visualization.

Full documentation lives at bitsandflames.github.io/fyron.

Install

uv add fyron

Useful extras:

Extra	Install	Adds
Excel	uv add "fyron[excel]"	Excel read/write via openpyxl
Survival	uv add "fyron[survival]"	KM, RMST, Cox, Weibull AFT
Survival ML	uv add "fyron[survival-ml]"	Gradient survival boosting via scikit-survival
ML	uv add "fyron[ml]"	Random Forest, XGBoost, metrics, plots
Boruta	uv add "fyron[ml,boruta]"	Boruta feature selection
Visualization	uv add "fyron[visualization]"	BOA segmentation collages

For a broad local analysis environment:

uv add "fyron[survival,survival-ml,ml,visualization,excel]"

Quick Examples

Console Easter Egg

Fyron keeps normal imports and CLI jobs quiet. The banner is opt-in:

fyron spark
fyron banner --theme mono --no-version

FHIR Query

from fyron import FHIRRestClient

client = FHIRRestClient("https://hapi.fhir.org/baseR4")
patients = client.search_df("Patient", params={"_count": 10}, max_pages=1)
patients.head()

Cohort Table

from fyron.cohort import build_survival_columns, validate_cohort_table

cohort = build_survival_columns(
    cohort,
    start_col="diagnosis_date",
    end_col="last_followup_or_event_date",
    event_col="event",
)
validate_cohort_table(cohort, required_columns=["patient_id", "time", "event"])

Survival Analysis

from fyron.survival import plot_kaplan_meier, fit_multivariate_cox

plot_kaplan_meier(
    cohort,
    duration_col="time",
    event_col="event",
    group_col="treatment_group",
    at_risk_counts=True,
)

cox = fit_multivariate_cox(
    cohort,
    duration_col="time",
    event_col="event",
    covariates=["age", "stage", "risk_score"],
)

Tabular ML

from fyron import ml

result = ml.run_classification_pipeline(
    X,
    y,
    model="random_forest",
    n_estimators=300,
    random_state=42,
    plot=True,
)

result["metrics"]

Clinical Plots

from fyron import ml

fig_corr, _, corr = ml.plot_correlation_heatmap(X, method="spearman")
fig_metrics, _ = ml.plot_metric_bars({"Random Forest": result["metrics"]})

BOA Visualization

from fyron.visualization import create_boa_segmentation_collage

create_boa_segmentation_collage(
    ["/data/patient_a", "/data/patient_b"],
    "boa_collage.png",
    orientation="axial",
    segmentation_layers=["body_regions", "tissues", "total"],
    slice_fraction=0.5,
)

Modules

Module	Purpose
fyron.fhir	FHIR REST, SQL, authentication, and resource utilities
fyron.dicom	DICOMweb download workflows
fyron.imaging	DICOM/NIfTI I/O and CT/MR normalization
fyron.documents	Authenticated document downloads
fyron.llm	LLM-assisted prompts over DataFrames and documents
fyron.core	Environment loading, local/S3 table I/O, JSON sidecars, Teable cohorts
fyron.cohort	Patient-level joins, validation, and survival columns
fyron.survival	KM, RMST, Cox, Weibull AFT, survival boosting
fyron.ml	Classification models, metrics, plots, feature selection
fyron.boa_extraction	BOA cohort feature extraction from measurement JSON and NIfTI masks
fyron.visualization	BOA segmentation collages

See the module documentation for examples for every module.

Development

uv sync --all-extras --dev
uv run pytest

Preview documentation locally:

uv run python scripts/build_docs.py
uv run python -m http.server 8001 --directory site

GitHub Pages is built from the Markdown files in docs/ through the BF-native generator in scripts/build_docs.py. The Pages workflow publishes the generated site/ folder when changes land on main.

License

MIT. See LICENSE.