Role of automated detection of respiratory related heart rate changes in the diagnosis of sleep disordered breathing

Figures

• 

  Figure 1. A representative record from a participant in the Sleep Heart Health Study. In the RRI channel, red dots indicate RRI dips and black stars indicate artifacts. In the heart rate channel red dots indicate heart rate accelerations. In the SaO2 channel, the red triangles indicate desaturations that are 3% or greater. RRI, R-R interval; SaO2, pulse oximetry. *indicate artifcats.

• 

  Figure 2. A representative record from a participant in the Sleep Heart Health Study who has SDB (AHI = 16.9 events/h). The red dots represent the O2 desaturations (ODI = 11.5), heart rate accelerations (HRAI from pulse oximeter), and RRI dips (from ECG) throughout the duration of the PSG recording (9 h). Note the incremental increase of values from ODI, AHI to HRAI, and RRI dips. In the respiratory events channel, red lines indicate hypopneas, blue lines indicate obstructive apneas and yellow lines indicate central apneas. In the hypnogram, the wake-sleep stages are indicated as wake, N1, N2, N3, and REM. In the RRI channel, red dots indicate RRI dips. In the Heart Rate channel, red dots indicate heart rate accelerations. In the SaO2 channel, the red triangles indicate desaturations that are 3% or greater. N1, non-REM stage N1; N2, non-REM stage N2; N3, non-REM stage N3; RRI, R-R interval; SaO2, pulse oximetry.

• 

  Figure 3. A representative polygraph from a participant in the Sleep Heart Health Study. In the EEG channel, the dotted lines indicate the beginning and ends of arousals scored by SHHS staff. In the flow channel, the dotted lines indicate the beginning and end of respiratory events scored by SHHS staff. In the SaO2 channel, the gray arrow indicates a desaturation. In the heart rate channel, the black arrows indicate heart rate accelerations. In the RRI channel, the white arrow indicates RRI dips. EEG, electroencephalogram; EMG, electromyogram; SaO2, pulse oximetry; EKG, electrocardiogram; RRI, R-R interval.

• 

  Figure 4. Bland Altman Plots comparing (A) respiratory-related HRAI to AHI [ICC = 0.64 (0.61, 0.67)]; (B) respiratory-related RRDI to AHI [ICC = 0.38 (0.33, 0.42)]; (C) total HRAI to NPSG AHI [ICC = 0.22 (0.17, 0.27)]; and (D) total RRDI to NSPG AHI [ICC = 0.22 (0.16, 0.26)] in the discovery dataset. HRAI, heart rate acceleration index; RRDI, RR interval dips index; AHI, apnea–hypopnea index; NPSG, nocturnal polysomnography.

• 

  Figure 5. Bland Altman Plots comparing (A) respiratory-related HRAI to AHI [ICC =0.51 (0.45, 0.56)]; (B) respiratory-related RRDI to AHI [ICC = 0.18 (0.10, 0.25)]; (C) total HRAI to NPSG AHI [ICC = 0.19 (0.12, 0.26)]; and (D) total RRDI to NSPG AHI [ICC = 0.08 (0.003, 0.16)] in the validation dataset. HRAI, heart rate acceleration index; RRDI, RR interval dips index; AHI, apnea–hypopnea index; NPSG, nocturnal polysomnography.

• 

  Figure 6. Receiver operating characteristic curves for (A) respiratory-related HRAI to AHI; (B) respiratory-related RRDI to AHI; (C) total HRAI to NPSG AHI; and (D) total RRDI to NSPG AHI in the discovery dataset. HRAI, heart rate acceleration index; RRDI, RR interval dips index; AHI, apnea-hypopnea index; NPSG, nocturnal polysomnography.

• 

  Figure 7. Receiver operating characteristic curves for (A) respiratory-related HRAI to AHI; (B) respiratory-related RRDI to AHI; (C) total HRAI to NPSG AHI; and (D) total RRDI to NSPG AHI in the validation dataset. HRAI, heart rate acceleration index; RRDI, RR interval dips index; AHI, apnea-hypopnea index; NPSG, nocturnal polysomnography.