Skip to main content

functions

Fovea-specific GA calculation functions.

The shared low-level primitives (convert_nan_to_zero, get_max_ga_bscan_index, get_lesion_sizes, get_shortest_distance_from_image_centre, parse_bscan_predictions, compute_ga_metrics_for_scan) live in bitfount.steps.data_utils.ga_metrics and are used by both the with-fovea and without-fovea steps.

This module provides the fovea-specific helpers and the high-level compute_ga_metrics_with_fovea_for_scan function.

Module​

Functions​

compute_ga_metrics_with_fovea_for_scan​

def compute_ga_metrics_with_fovea_for_scan(    bscan_predictions: tuple[str, ...],    slice_thickness: float,    pixel_spacing_column: float,    all_segmentation_labels: dict[str, int],    ga_area_include_segmentations: list[str],    ga_area_exclude_segmentations: list[str],    central_slice_fovea: int | None,    landmarks_fovea: FoveaLandmarks | None,    fovea_landmark_idx: int = 2,    n_scan_biomarker_thresholds: Mapping[str, float] | None = None,    include_raw_pathology_probabilities: bool = False,) ‑> GAMetricsWithFovea:

Compute GA metrics for a single scan, including fovea distance.

This encapsulates the per-file computation loop body from _BaseWorkerSideWithFovea.run() (lines 865-987 in the original).

Arguments

  • bscan_predictions: The raw per-B-scan prediction JSON strings.
  • slice_thickness: The distance between B-scans in mm.
  • pixel_spacing_column: The pixel spacing along the B-scan column axis.
  • all_segmentation_labels: All segmentation label names to class index.
  • ga_area_include_segmentations: Segmentation labels to include when computing the GA area column mask.
  • ga_area_exclude_segmentations: Segmentation labels to exclude when computing the GA area column mask.
  • central_slice_fovea: Central slice of the fovea prediction.
  • landmarks_fovea: List of 3-element lists of fovea landmarks.
  • fovea_landmark_idx: Index of the middle landmark in the tuple.
  • n_scan_biomarker_thresholds: Per-biomarker >= probability threshold used to compute n_scan_run_lengths. Keys define which N-scan biomarkers are evaluated; when None, every label in N_SCAN_BIOMARKER_LABELS defaults to 0.5.
  • include_raw_pathology_probabilities: Whether to include the raw per-B-scan pathology probability arrays on the returned metrics.

Raises

  • Exception: Propagates any exception from prediction parsing or metric computation so that the caller can handle/skip.

get_shortest_distance_from_fovea​

def get_shortest_distance_from_fovea(    fovea_coordinates: NDArray[Any] | None,    labeled_array: NDArray[Any],    num_lesions: int,    slice_thickness: float,    pixel_spacing_column: float,    na_bscan_indices: tuple[int, ...] = (),) ‑> float | None:

Calculate the distance from the fovea to the nearest lesion.

Arguments

  • fovea_coordinates: Numpy array of fovea coordinates in the image, as (bscan, column, row) in the ORIGINAL (uncompacted) B-scan index space — the space a fovea landmark arrives in.
  • labeled_array: Numpy array of shape (num_bscans, num_cols) where each pixel is labelled with the lesion number it belongs to. Its first axis is compacted: parse_bscan_predictions gives it a row only for a B-scan that carried model output.
  • num_lesions: Number of lesions in the image.
  • slice_thickness: Thickness of each B-scan in mm.
  • pixel_spacing_column: Spacing between columns in mm.
  • na_bscan_indices: The dropped input B-scan indices, from ParsedBScanPredictions.na_bscan_indices. Used to shift fovea_coordinates' B-scan component into labeled_array's compacted space. Defaults to (), i.e. no compaction, for a caller that genuinely has none.

Returns Distance from the fovea to the nearest lesion in mm, or None.