Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Removed the support for older MEDS versions. #114

Merged
merged 1 commit into from
Aug 23, 2024
Merged

Removed the support for older MEDS versions. #114

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
53 changes: 0 additions & 53 deletions src/aces/predicates.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -223,56 +223,6 @@ def direct_load_plain_predicates(
)


def unnest_meds(data: pl.DataFrame) -> pl.DataFrame:
"""Unnest MEDS data.

Single-nested MEDS has a row per patient, with all the observations aggregated into a list of "event"
structs. The events column needs to be exploded and unnested.

Args:
data: The Polars DataFrame containing the single-nested MEDS data.

Returns:
The Polars DataFrame with the events column exploded and unnested.

Example:
>>> data = pl.DataFrame({
... "subject_id": [1, 2, 3],
... "events": [
... [{"timestamp": None, "code": "EYE_COLOR//HAZEL", "numerical_value": None},
... {"timestamp": None, "code": "HEIGHT", "numerical_value": 160},
... {"timestamp": "3/9/1978 00:00", "code": "DOB", "numerical_value": None}],
... [{"timestamp": None, "code": "EYE_COLOR//BROWN", "numerical_value": None},
... {"timestamp": None, "code": "HEIGHT", "numerical_value": 175},
... {"timestamp": "12/28/1980 00:00", "code": "DOB", "numerical_value": None}],
... [{"timestamp": None, "code": "EYE_COLOR//BROWN", "numerical_value": None},
... {"timestamp": None, "code": "HEIGHT", "numerical_value": 166},
... {"timestamp": "12/19/1988 00:00", "code": "DOB", "numerical_value": None}],
... ],
... })
>>> unnest_meds(data)
shape: (9, 4)
┌────────────┬──────────────────┬──────────────────┬─────────────────┐
│ subject_id ┆ timestamp ┆ code ┆ numerical_value │
│ --- ┆ --- ┆ --- ┆ --- │
│ i64 ┆ str ┆ str ┆ i64 │
╞════════════╪══════════════════╪══════════════════╪═════════════════╡
│ 1 ┆ null ┆ EYE_COLOR//HAZEL ┆ null │
│ 1 ┆ null ┆ HEIGHT ┆ 160 │
│ 1 ┆ 3/9/1978 00:00 ┆ DOB ┆ null │
│ 2 ┆ null ┆ EYE_COLOR//BROWN ┆ null │
│ 2 ┆ null ┆ HEIGHT ┆ 175 │
│ 2 ┆ 12/28/1980 00:00 ┆ DOB ┆ null │
│ 3 ┆ null ┆ EYE_COLOR//BROWN ┆ null │
│ 3 ┆ null ┆ HEIGHT ┆ 166 │
│ 3 ┆ 12/19/1988 00:00 ┆ DOB ┆ null │
└────────────┴──────────────────┴──────────────────┴─────────────────┘
"""
logger.info("Found single-nested MEDS data, unnesting...")

return data.explode("events").unnest("events")


def generate_plain_predicates_from_meds(data_path: Path, predicates: dict) -> pl.DataFrame:
"""Generate plain predicate columns from a MEDS dataset.

Expand Down Expand Up @@ -318,9 +268,6 @@ def generate_plain_predicates_from_meds(data_path: Path, predicates: dict) -> pl
logger.info("Loading MEDS data...")
data = pl.read_parquet(data_path).rename({"patient_id": "subject_id", "time": "timestamp"})

if data.columns == ["subject_id", "events"]:
data = unnest_meds(data)

# generate plain predicate columns
logger.info("Generating plain predicate columns...")
for name, plain_predicate in predicates.items():
Expand Down
Loading