Recognition of correct finger placement for photoplethysmographic imaging

Calculation of Heartbeat Rate and SpO2 Parameters Using a Smartphone Camera: Analysis and Testing

Mathematical and signal-processing methods were used to obtain reliable measurements of the heartbeat pulse rate and information on oxygen concentration in the blood using short video recordings of the index finger attached to a smartphone built-in camera. Various types of smartphones were used with different operating systems (e.g., iOS, Android) and capabilities. A range of processing algorithms were applied to the red-green-blue (RGB) component signals, including mean intensity calculation, moving average smoothing, and quadratic filtering based on the Savitzky-Golay filter. Two approaches-gradient and local maximum methods-were used to determine the pulse rate, which provided similar results. A fast Fourier transform was applied to the signal to correlate the signal's frequency components with the pulse rate. We resolved the signal into its DC and AC components to calculate the ratio-of-ratios of the AC and DC components of the red and green signals, a method which is often used to estimate the oxygen concentration in blood. A series of measurements were performed on healthy human subjects, producing reliable data that compared favorably to benchmark data obtained by commercial and medically approved oximeters. Furthermore, the effect of the video recording duration on the accuracy of the results was investigated.

HRV4Training: Large-scale longitudinal training load analysis in unconstrained free-living settings using a smartphone application

We describe an approach to support athletes at various fitness levels in their training load analysis using heart rate (HR) and heart rate variability (HRV). A smartphone-based application (HRV4Training) was developed that captures heart activity over one to five minutes using photoplethysmography (PPG) and derives HR and HRV features. HRV4Training integrated a guide for an early morning spot measurement protocol and a questionnaire to capture self-reported training activity. The smartphone application was made publicly available for interested users to quantify training effect. Here we analyze data acquired over a period of 3 weeks to 5 months, including 797 users, breaking down results by gender and age group. Our results suggest a strong relation between HR, HRV and self-reported training load independent of gender and age group. HRV changes due to training were larger than those of HR. We conclude that smartphone-based training monitoring is feasible and a can be used as a practical tool to support large populations outside controlled laboratory environments.

Estimation of respiratory rate from photoplethysmographic imaging videos compared to pulse oximetry

We present a study evaluating two respiratory rate estimation algorithms using videos obtained from placing a finger on the camera lens of a mobile phone. The two algorithms, based on Smart Fusion and empirical mode decomposition (EMD), consist of previously developed signal processing methods to detect features and extract respiratory induced variations in photoplethysmographic signals to estimate respiratory rate. With custom-built software on an Android phone, photoplethysmographic imaging videos were recorded from 19 healthy adults while breathing spontaneously at respiratory rates between 6 to 32 breaths/min. Signals from two pulse oximeters were simultaneously recorded to compare the algorithms' performance using mobile phone data and clinical data. Capnometry was recorded to obtain reference respiratory rates. Two hundred seventy-two recordings were analyzed. The Smart Fusion algorithm reported 39 recordings with insufficient respiratory information from the photoplethysmographic imaging data. Of the 232 remaining recordings, a root mean square error (RMSE) of 6 breaths/min was obtained. The RMSE for the pulse oximeter data was lower at 2.3 breaths/min. RMSE for the EMD method was higher throughout all data sources as, unlike the Smart Fusion, the EMD method did not screen for inconsistent results. The study showed that it is feasible to estimate respiratory rates by placing a finger on a mobile phone camera, but that it becomes increasingly challenging at respiratory rates greater than 20 breaths/min, independent of data source or algorithm tested.

Respiratory rate assessment from photoplethysmographic imaging

We present a study investigating the suitability of a respiratory rate estimation algorithm applied to photoplethysmographic imaging on a mobile phone. The algorithm consists of a cascade of previously developed signal processing methods to detect features and extract respiratory induced variations in photoplethysmogram signals to estimate respiratory rate. With custom-built software on an Android phone (Camera Oximeter), contact photoplethysmographic imaging videos were recorded using the integrated camera from 19 healthy adults breathing spontaneously at respiratory rates between 6 and 40 breaths/min. Capnometry was simultaneously recorded to obtain reference respiratory rates. Two hundred and ninety-eight Camera Oximeter recordings were available for analysis. The algorithm detected 22 recordings with poor photoplethysmogram quality and 46 recordings with insufficient respiratory information. Of the 232 remaining recordings, a root mean square error of 5.9 breaths/min and a median absolute error of 2.3 breaths/min was obtained. The study showed that it is feasible to estimate respiratory rates by placing a finger on a mobile phone camera, but that it becomes increasingly challenging at respiratory rates higher than 20 breaths/min.

Detection of the optimal region of interest for camera oximetry

The estimation of heart rate and blood oxygen saturation with an imaging array on a mobile phone (camera oximetry) has great potential for mobile health applications as no additional hardware other than a camera and LED flash enabled phone are required. However, this approach is challenging as the configuration of the camera can negatively influence the estimation quality. Further, the number of photons recorded with the photo detector is largely dependent on the optical path length, resulting in a non-homogeneous image. In this paper we describe a novel method to automatically detect the optimal region of interest (ROI) for the captured image to extract a pulse waveform. We also present a study to select the optimal camera settings, notably the white balance. The experiments show that the incandescent white balance mode is the preferable setting for camera oximetry applications on the tested mobile phone (Samsung Galaxy Ace). Also, the ROI algorithm successfully identifies the frame regions which provide waveforms with the largest amplitudes.