Notice. Illustration and research prototype. Not clinically validated, not a medical device, not for patient care. Example data are fictional.
Live demonstration. The sliders and test cases work — every score recomputes in your browser. The data is generated, no patient is involved, and nothing here has been verified by a clinician.

waveform.yaml

Configuration · waveform.yaml · 95 lines · editable in place
meta
version0.1.0
ownerUNASSIGNED โ€” set to the responsible clinician
verified bynot set
note'Thresholds are geometric and were calibrated against generated waveforms, not real
patients. They must be checked against real recordings from the unit''s own ventilators
before use.
'
sampling
default hz50
min breaths to assess8
window seconds60
segmentation
min inspiratory time s0.15
flow threshold lpm3.0
min breath interval s0.8
asynchrony
double trigger
enabledtrue
nameDouble triggering
geometry'Two inspiratory cycles separated by less than the stated fraction of the median
expiratory time, and by less than the absolute floor.
'
why'Delivers stacked breaths, so the true tidal volume is double what is set.
'
max gap fraction of expiry0.25
max gap absolute s0.25
ineffective effort
enabledtrue
nameIneffective effort
geometry'Expiratory flow rises against its decay for at least the stated duration, by
at least the stated fraction of the expiratory peak, with no machine breath following.
'
why'The patient tried to breathe and the ventilator did not respond.
'
skip initial fraction0.35
min rise duration s0.12
min amplitude fraction0.55
short cycle
enabledtrue
namePremature cycling
geometry'Inspiratory time below the stated fraction of the median AND below the absolute
floor. Half of all breaths sit below the median by definition, so the fraction alone is
not a rule.
'
whyThe breath ends while the patient is still inhaling.
max fraction of median0.45
max absolute s0.35
auto trigger
enabledtrue
nameAuto-triggering
geometry'Observed respiratory rate exceeds the set rate by the stated factor.
'
why'The ventilator is cycling on artefact โ€” cardiac oscillation or water in the circuit.
'
rate ratio1.3
index
significant threshold0.1
note'Asynchrony index above 10% is the conventional threshold for clinical significance.
Only meaningful if the false positive rate on normal breathing is near zero.
'
arterial
ppv
enabledtrue
responsive threshold percent13
min beats6
min pulse pressure mmhg15
min beat interval s0.35
caveat'Valid only in a passively ventilated patient in sinus rhythm with tidal volume
of at least 8 mL/kg. Meaningless with spontaneous effort, arrhythmia, or an open chest
โ€” and those conditions are common, so the caveat travels with the number.
'
hrv
enabledtrue
min beats20
plausible rr ms
300
2000
reference
sdnn normal ms50
note'One study reported a 25% wavelet-HRV drop at a mean of 35 hours before sepsis onset,
on 17 patients with 14 cases. Individual lead times ranged 0 to 227 hours. Directionally
interesting; not a threshold.
'