Accurate and continuous respiratory rate using touchless monitoring technology

PURPOSE

Respiratory rate is a commonly used vital sign with various clinical applications. It serves as a crucial marker of acute health issues and any significant alteration in respiratory rate may be an early warning sign of major issues such as infections in the respiratory tract, respiratory failure, or cardiac arrest. Timely recognition of changes in respiratory rate enables prompt medical action, while neglecting to detect a change may lead to adverse patient outcomes. Here, we report on the performance of respiratory rate determined using a depth sensing camera system (RRdepth) which allows for continuous, non-contact 'touchless' monitoring of this important vital sign.

METHODS

Thirty adult volunteers undertook a range of set breathing rates to cover a target breathing range of 4-40 breaths/min. Depth information was acquired from the torso region of the subjects using an Intel D415 RealSense camera positioned above the bed. The depth information was processed to generate a respiratory signal from which RRdepth was calculated. This was compared to a manually scored capnograph reference (RRcap).

RESULTS

An overall RMSD accuracy of 0.77 breaths/min was achieved across the target respiratory rate range with a corresponding bias of 0.05 breaths/min. This corresponded to a line of best fit given by RRdepth = 1.01 x RRcap - 0.22 breaths/min with an associated high degree of correlation (R = 0.997). A breakdown of the performance with respect to sub-ranges corresponding to respiratory rates or ≤7, >7-10, >10-20, >20-30, >30 breaths/min all exhibited RMSD accuracies of less than 1.00 breaths/min. We also had the opportunity to test the performance of spontaneous breathing of the subjects which occurred during the study and found an overall RMSD accuracy of 1.20 breaths/min with corresponding accuracies ≤1.30 breaths/min across each of the individual sub-ranges.

CONCLUSIONS

We have conducted an investigative study of a prototype depth sensing camera system for the non-contact monitoring of respiratory rate. The system achieved good performance with high accuracy across a wide range of rates including both clinically important high and low rates.

Robust Non-Contact Monitoring of Respiratory Rate using a Depth Camera

PURPOSE

Respiratory rate (RR) is one of the most common vital signs with numerous clinical uses. It is an important indicator of acute illness and a significant change in RR is often an early indication of a potentially serious complication or clinical event such as respiratory tract infection, respiratory failure and cardiac arrest. Early identification of changes in RR allows for prompt intervention, whereas failing to detect a change may result in poor patient outcomes. Here, we report on the performance of a depth-sensing camera system for the continuous non-contact 'touchless' monitoring of Respiratory Rate.

METHODS

Seven healthy subjects undertook a range of breathing rates from 4 to 40 breaths-per-minute (breaths/min). These were set rates of 4, 5, 6, 8, 10, 15, 20, 25, 30, 35 and 40 breaths/min. In total, 553 separate respiratory rate recordings were captured across a range of conditions including body posture, position within the bed, lighting levels and bed coverings. Depth information was acquired from the scene using an Intel D415 RealSense^TM camera. This data was processed in real-time to extract depth changes within the subject's torso region corresponding to respiratory activity. A respiratory rate RR_depth was calculated using our latest algorithm and output once-per-second from the device and compared to a reference.

RESULTS

An overall RMSD accuracy of 0.69 breaths/min with a corresponding bias of -0.034 was achieved across the target RR range of 4-40 breaths/min. Bland-Altman analysis revealed limits of agreement of -1.42 to 1.36 breaths/min. Three separate sub-ranges of low, normal and high rates, corresponding to < 12, 12-20, > 20 breaths/min, were also examined separately and each found to demonstrate RMSD accuracies of less than one breath-per-minute.

CONCLUSIONS

We have demonstrated high accuracy in performance for respiratory rate based on a depth camera system. We have shown the ability to perform well at both high and low rates which are clinically important.

Noncontact Respiratory Monitoring Using Depth Sensing Cameras: A Review of Current Literature

There is considerable interest in the noncontact monitoring of patients as it allows for reduced restriction of patients, the avoidance of single-use consumables and less patient-clinician contact and hence the reduction of the spread of disease. A technology that has come to the fore for noncontact respiratory monitoring is that based on depth sensing camera systems. This has great potential for the monitoring of a range of respiratory information including the provision of a respiratory waveform, the calculation of respiratory rate and tidal volume (and hence minute volume). Respiratory patterns and apneas can also be observed in the signal. Here we review the ability of this method to provide accurate and clinically useful respiratory information.

Continuous respiratory rate monitoring during an acute hypoxic challenge using a depth sensing camera

Respiratory rate is a well-known to be a clinically important parameter with numerous clinical uses including the assessment of disease state and the prediction of deterioration. It is frequently monitored using simple spot checks where reporting is intermittent and often prone to error. We report here on an algorithm to determine respiratory rate continuously and robustly using a non-contact method based on depth sensing camera technology. The respiratory rate of 14 healthy volunteers was studied during an acute hypoxic challenge where blood oxygen saturation was reduced in steps to a target 70% oxygen saturation and which elicited a wide range of respiratory rates. Depth sensing data streams were acquired and processed to generate a respiratory rate (RR_depth). This was compared to a reference respiratory rate determined from a capnograph (RR_cap). The bias and root mean squared difference (RMSD) accuracy between RR_depth and the reference RR_cap was found to be 0.04 bpm and 0.66 bpm respectively. The least squares fit regression equation was determined to be: RR_depth = 0.99 × RR_cap + 0.13 and the resulting Pearson correlation coefficient, R, was 0.99 (p < 0.001). These results were achieved with a 100% reporting uptime. In conclusion, excellent agreement was found between RR_depth and RR_cap. Further work should include a larger cohort combined with a protocol to further test algorithmic performance in the face of motion and interference typical of that experienced in the clinical setting.

Video-based heart rate monitoring across a range of skin pigmentations during an acute hypoxic challenge

The robust monitoring of heart rate from the video-photoplethysmogram (video-PPG) during challenging conditions requires new analysis techniques. The work reported here extends current research in this area by applying a motion tolerant algorithm to extract high quality video-PPGs from a cohort of subjects undergoing marked heart rate changes during a hypoxic challenge, and exhibiting a full range of skin pigmentation types. High uptimes in reported video-based heart rate (HRvid) were targeted, while retaining high accuracy in the results. Ten healthy volunteers were studied during a double desaturation hypoxic challenge. Video-PPGs were generated from the acquired video image stream and processed to generate heart rate. HRvid was compared to the pulse rate posted by a reference pulse oximeter device (HRp). Agreement between video-based heart rate and that provided by the pulse oximeter was as follows: Bias = - 0.21 bpm, RMSD = 2.15 bpm, least squares fit gradient = 1.00 (Pearson R = 0.99, p < 0.0001), with a 98.78% reporting uptime. The difference between the HRvid and HRp exceeded 5 and 10 bpm, for 3.59 and 0.35% of the reporting time respectively, and at no point did these differences exceed 25 bpm. Excellent agreement was found between the HRvid and HRp in a study covering the whole range of skin pigmentation types (Fitzpatrick scales I-VI), using standard room lighting and with moderate subject motion. Although promising, further work should include a larger cohort with multiple subjects per Fitzpatrick class combined with a more rigorous motion and lighting protocol.

Video-Based Physiologic Monitoring During an Acute Hypoxic Challenge: Heart Rate, Respiratory Rate, and Oxygen Saturation

BACKGROUND

The physiologic information contained in the video photoplethysmogram is well documented. However, extracting this information during challenging conditions requires new analysis techniques to capture and process the video image streams to extract clinically useful physiologic parameters. We hypothesized that heart rate, respiratory rate, and oxygen saturation trending can be evaluated accurately from video information during acute hypoxia.

METHODS

Video footage was acquired from multiple desaturation episodes during a porcine model of acute hypoxia using a standard visible light camera. A novel in-house algorithm was used to extract photoplethysmographic cardiac pulse and respiratory information from the video image streams and process it to extract a continuously reported video-based heart rate (HRvid), respiratory rate (RRvid), and oxygen saturation (SvidO2). This information was then compared with HR and oxygen saturation references from commercial pulse oximetry and the known rate of respiration from the ventilator.

RESULTS

Eighty-eight minutes of data were acquired during 16 hypoxic episodes in 8 animals. A linear mixed-effects regression showed excellent responses relative to a nonhypoxic reference signal with slopes of 0.976 (95% confidence interval [CI], 0.973-0.979) for HRvid; 1.135 (95% CI, 1.101-1.168) for RRvid, and 0.913 (95% CI, 0.905-0.920) for video-based oxygen saturation. These results were obtained while maintaining continuous uninterrupted vital sign monitoring for the entire study period.

CONCLUSIONS

Video-based monitoring of HR, RR, and oxygen saturation may be performed with reasonable accuracy during acute hypoxic conditions in an anesthetized porcine hypoxia model using standard visible light camera equipment. However, the study was conducted during relatively low motion. A better understanding of the effect of motion and the effect of ambient light on the video photoplethysmogram may help refine this monitoring technology for use in the clinical environment.

Gradient adjustment method for better discriminating correlating and non-correlating regions of physiological signals: application to the partitioning of impaired and intact zones of cerebral autoregulation

Cerebral blood flow (CBF) is regulated over a range of systemic blood pressures by the cerebral autoregulation (CA) control mechanism. This range lies within the lower and upper limits of autoregulation (LLA, ULA), beyond which blood pressure drives CBF, and CA function is considered impaired. A standard method to determine autoregulation limits noninvasively using NIRS technology is via the COx measure: a moving correlation index between mean arterial pressure and regional oxygen saturation. In the intact region, there should be no correlation between these variables whereas in the impaired region, the correlation index should approximate unity. In practice, however, the data may be noisy and/or the intact region may often exhibit a slightly positive relationship. This positive relationship may render traditional autoregulation limit calculations difficult to perform, resulting in the need for manual interpretation of the data using arbitrary thresholds. Further, the underlying mathematics of the technique are asymmetric in terms of the results produced for impaired and intact regions and are, in fact, not computable for the ideal case within the intact region. In this work, we propose a novel gradient adjustment method (GACOx) to enhance the differences in COx values observed in the intact and impaired regions. Results from a porcine model (N = 8) are used to demonstrate that GACOx is successful in determining LLA values where traditional methods fail. It is shown that the derived GACOx indices exhibit a mean difference between the intact/impaired regions of 1.54 ± 0.26 (mean ± SD), compared to 0.14 ± 0.10 for the traditional COx method. The GACOx effectively polarizes the COx data in order to better differentiate the intact and impaired zones and, in doing so, makes the determination of the LLA and ULA points a simpler and more consistent task. The method lends itself to the automation of the robust determination of autoregulation zone limits.

Veno-venous extracorporeal lung assist with concurrent distal aortic perfusion: repair of ruptured aorta in a patient with dense ARDS

Extracorporeal lung assist (ECLA) allowed surgical repair of a ruptured descending thoracic aorta to be performed in a patient with profound respiratory failure. Dense acute respiratory distress syndrome (ARDS) developed during his 15-day hospitalization at a regional trauma center. After transfer to a Level I facility, an additional injury was diagnosed: traumatic rupture of the aorta, contained within a pseudoaneurysm. ECLA by the veno-venous route was required immediately preoperatively and distal aortic perfusion was performed during the aortic repair. Despite deflation of the left lung, the patient was oxygenated and ventilated adequately during surgery. Cross-clamp time was 48 minutes. The patient was weaned from ECLA by the fifth postoperative day. To our knowledge, this is the first report of concurrent veno-venous pulmonary support with distal aortic perfusion.

Low blood flow extracorporeal carbon dioxide removal (ECCO2R): a review of the concept and a case report

Despite advances in respiratory and critical care medicine, the mortality from ARDS remains unchanged. Recent research suggests current ventilatory therapy may produce additional lung injury, retarding the recovery process of the lung. Alternative supportive therapies, such as ECMO and ECCO2R, ultimately may result in less ventilator induced lung injury. Due to the invasiveness of ECMO/ECCO2R, these modalities are initiated reluctantly and commonly not until patients suffer from terminal or near-terminal respiratory failure. Low flow ECCO2R may offer advantages of less invasiveness and be suitable for early institution before ARDS becomes irreversible. We describe a patient with ARDS and severe macroscopic barotrauma supported with low flow ECCO2R resulting in significant CO2 clearance, reduction of peak, mean airway pressures and minute ventilation.

An in vitro physiologic model for cardiopulmonary simulation: a system for ECMO training

Extracorporeal life support (ELS) systems may be run by certified perfusionists, specially trained nurses or respiratory therapy staff. Guidelines for the training, certification and retraining of ELS operators have been established by the Extracorporeal Life Support Organization. Recommendations include "... a well defined program for staff training, certification, and retraining". Some clinicians have suggested that ELS operators be certified and recertified in an animal laboratory. But such practice involves veterinary expenses, animal use issues and considerable clean-up and disposal. We describe an alternative method of training, using an in vitro physiologic model designed to simulate various pathophysiologic states. In addition, the in vitro physiologic model may be used to evaluate membrane lung characteristics. This model's ease of construction, maintenance and use for training compared with live animal techniques are discussed. Research capabilities may be more flexible than with the use of the live animal technique. The in vitro physiologic model can be a useful and convenient asset to an extracorporeal membrane oxygenation/extracorporeal carbon dioxide removal (ECMO/ECCO2R) program.

Prospective evaluation of combined high-frequency ventilation in post-traumatic patients with adult respiratory distress syndrome refractory to optimized conventional ventilatory management

This study explores the value of combined high-frequency ventilation (CHFV) in a prospective clinical trial of 35 patients suffering from severe post-traumatic and/or septic adult respiratory distress syndrome (ARDS) who were refractory to conventional controlled mechanical ventilatory (CMV) support. The severity of ARDS was quantified by lung mechanics and gas exchange variables and the patients were classified on clinical grounds as well as on the basis of their respiratory index/pulmonary shunt relationship [RI/(Qsp/Qt)]. During the same time period as the CHFV study, data from these patients were compared to those from 88 ARDS patients who had quantitatively similar degrees of respiratory insufficiency, but who were treated only with controlled mechanical ventilation (CMV). The use of CHFV in the 35 CMV refractory patients resulted in an increase in expired tidal volume (VTE) by reducing the CMV inspired tidal volume (VTI) while increasing the volume component derived from high-frequency ventilation (HFV). This procedure appeared to reveal potentially salvageable ARDS patients who were refractory to CMV. In these patients, CHFV significantly reduced pulmonary mean airway pressure (Paw). The RI also decreased significantly and it was possible to reduce significantly the FIO2. In surviving ARDS patients treated with CHFV, an improvement in blood gases at reduced FIO2, without decreased cardiac output, was produced. The CHFV technique was used for less than or equal to 25 days and resulted in 23% survival of patients who were clinically and physiologically indistinguishable from the patients in the ARDS nonsurvivor group who were treated by CMV only. In surviving CHFV patients the decrease in Paw permitted a sustained, or increased, cardiac output with a rise in the oxygen delivery/oxygen consumption ratio, thus allowing for a higher PaO2 for any given level of pulmonary shunt.

Respiratory index/pulmonary shunt relationship: quantification of severity and prognosis in the post-traumatic adult respiratory distress syndrome

The relationship between the respiratory index (RI = alveolar-arterial oxygen gradient [P(A-a)O2] normalized by PaO2) and the pulmonary shunt (Qsp/Qt) has been examined in 929 studies from 240 critically ill post-traumatic patients. Of these, 88 patients (443 studies) were individuals who developed post-traumatic adult respiratory distress syndrome (ARDS) and 152 were patients (486 studies) who did not develop ARDS. This study demonstrates that the RI to Qsp/Qt [RI/(Qsp/Qt)] relationship was significantly (p less than .0001) increased in patients who developed fatal ARDS compared with those who did not develop ARDS, or with those whose ARDS resolved. Because of the increased oxygen consumption (VO2) in ARDS patients in association with their severe limitations in gas exchange (RI) and increased Qsp/Qt, surviving ARDS patients had a significant increase in the cardiac index which resulted in a higher oxygen delivery to VO2 ratio. ARDS patients showed significant (p less than .0001) evidence of increased pulmonary vascular tone, correlated with the increase in the RI/(Qsp/Qt) relationship. In addition, those patients with high RI/(Qsp/Qt) also had increased right ventricular (RVSW) to left ventricular work (LVSW) ratios which were shown to be a direct function of the rise in RI. This increase in both RVSW/LVSW and RI/(Qsp/Qt) ratios was significantly (p less than .0001) correlated with an increased mortality. Thus, the RI/(Qsp/Qt) relationship, which can be obtained from arterial and mixed venous blood gases and saturations only, can be used to predict the severity of the ARDS process as well as important pulmonary vascular and right ventricular overload consequences.

Gas exchange in low-compression HFPPV is maintained at low distending pressures in the pig

The fact that collateral ventilation normally occurs in the human lung has led to the suggestion that it might contribute to the successful clinical effects of low-compression high-frequency positive-pressure ventilation (HFPPV). As the pig has poor collateral ventilation, pulmonary vasoconstriction has to be part of the regulatory mechanisms matching ventilation-perfusion. A study was made on nine pigs anesthetized with ketamine hydrochloride intravenously to elucidate the maintenance of ventilation-perfusion balance during mechanical ventilation. Comparisons were made between the ventilatory patterns provided by a conventional ventilator (Servo-Ventilator 900C) and an improved prototype of a low-compression system for volume-controlled ventilation (system H). A ventilatory frequency of 20 breaths per min (bpm) with SV-900C (SV-20) and system H (H-20) and of 60 bpm with system H (H-60) was used. The experimental conditions were otherwise identical. Positive end-expiratory pressures (PEEP) were applied to maintain the same mean airway pressure with the three systems. The tidal volume required for normoventilation differed significantly between the three ventilatory patterns, but there were no differences in circulatory and oxygen-transport variables. By measurements of airway pressure and intrapleural liquid surface pressure, it was demonstrated that the distending pressure (at end-inspiration) was significantly lower with a low-compression system (H-20 versus SV-20), especially at a high ventilatory frequency (H-60 versus H-20). Consequently, although the mean airway pressure was set at the same level for the three different ventilatory modalities, the distending pressures required for the same alveolar ventilation and arterial oxygenation differed significantly.(ABSTRACT TRUNCATED AT 250 WORDS)

Comparative studies of IPPV and HFPPV with PEEP in critical care patients. II: Studies on intrapulmonary gas distribution

The ventilatory patterns of a conventional ventilator for volume-controlled ventilation (SV-900) and a low-compression ventilator utilizing the pneumatic valve principle for pressure/flow-generated, volume-controlled ventilation (system H) were studied in a lung model and in 10 patients with respiratory failure. System H was used at frequencies of 20 (H-20) and 60 (H-60 = high-frequency positive-pressure ventilation [HFPPV]) breath/min, and SV-900 at a frequency of 20 (SV-20) breath/min. With system H, inspiration constituted 22% (no inspiratory pause) and with SV-900, 25% (with 10% inspiratory pause). System H delivers an instantaneous accelerating flow which rapidly decelerates during the second part of the inspiratory phase. Maximal flow rates studied were 1.3 (H-60), 0.9 (H-20), and 0.7 (SV-20) L/sec. Thus, HFPPV delivers an effective tidal volume with highest linear velocity. This increased velocity increases gas mixing by increasing turbulence in conducting airways. In the 10 patients with respiratory failure, intrapulmonary gas distribution (measured as the nitrogen washout delay) was improved from 106% during SV-20 to 74% with H-60 (p less than 0.05). H-60 also increased carbon dioxide elimination in the 2 patients with the most severe pulmonary dysfunction.

Comparative studies of IPPV and HFPPV with PEEP in critical care patients. I: A clinical evaluation

The effects of the ventilatory patterns of a conventional ventilator (SV-900) and a low-compression ventilator (system H) were studied in 12 patients with respiratory failure (RF). Volume-controlled ventilation at frequencies (f) of 20 breath/min (SV-20) with SV-900, and 20 (H-20) and 60 (H-60 = high-frequency positive-pressure ventilation, HFPPV) breath/min with system H was given. Inspiration constituted 25% (with an inspiratory pause of 10%) of the ventilatory cycle with SV-900 and 22% with system H. Intratracheal (ITP), intrapleural, systemic and pulmonary arterial (PAP), and central venous (CVP) pressures were measured at normoventilation. During H-60, normoventilation was provided with smaller tidal volumes and lower mean intratracheal pressures than during SV-20 and H-20. Cardiac index and oxygen transport were not affected by changes in ventilatory pattern. The respiration-synchronous variations in CVP, PAP, and pulmonary capillary wedge pressure (WP) during ventilation at 20 breath/min were abolished during HFPPV. In the most severely ill patients, long-term HFPPV was uneventful. Airway suctioning during ventilation with oxygen was an important feature of the pneumatic valve principle (system H). The results of this study indicate that volume-controlled HFPPV is as efficient and as well accepted by the patient as conventional ventilation (SV-20).