Segment cohort recordings while preserving subject and session identity

Reuses the PhysioCore cohort hierarchy and segmentMultimodalCycles(). Sessions must be MultiPhysioExperiment objects with at least two consistently named streams. No resampling or synchronization is performed.

Usage

segmentCohortCycles(
  x,
  cycles,
  assay = "raw",
  incomplete = c("error", "retain")
)

Arguments

x: A PhysioCore PhysioCohort.
cycles: Nonempty data frame with character subject_id, session_id, trial_id, side, cycle_id, and numeric start_time, end_time. Times are seconds relative to each session's master-clock t0, as in PhysioExperiment::streamTimeIndex(). Trial IDs describe intervals within a session; they are never inferred from session IDs. Additional columns are retained. An optional participant_id must equal subject_id.
assay: A single assay name used in every stream (default "raw"). Assays must be numeric time-by-channel matrices.
incomplete: Passed to segmentMultimodalCycles().

Value

A multimodal_cycles object accepted by summarizeCycleFeatures(). Canonical keys include subject, session, trial, side and cycle. The participant_id alias is derived from subject_id. cohort_design contains the source cohort's subject/session table; recordings contains one entry per selected subject/session/trial context, including clock and per-modality acquisition metadata. Cycle rows retain input order.

Details

Subject-level splits can be made using existing cohort subsetting before extraction, together with the corresponding cycle rows. This bridge does not enforce statistical independence or downstream training splits. Units, anatomical channel meaning, and event detection remain the caller's responsibility. The selected sessions must share stream names; feature summarization additionally checks feature names within each stream. Feature callbacks receive matrices, not recording metadata. For rate-dependent features, process contexts with their own rate settings before pooling; a single fixed-rate callback is inappropriate when session rates differ.

Operates on a PhysioCohort multi-subject container (package PhysioCohort), which is not among this package's dependencies, so the example cannot run offline and is not executed.

Examples

if (FALSE) { # \dontrun{
# `cohort` is a PhysioCohort object; see the PhysioCohort package.
segmentCohortCycles(cohort, cycles = cohort_cycles)
} # }