sponsored FeATUre u Cover story

the LED pulses, timing, and transimpedance gain within a specific range to achieve the same power consumption on both systems to get a fair comparison (see Table 1). Table 1 shows an ADPD188GG LED

current, which is twice as high as the LED current in the ADPD107 setup. The reason is that the photodiode surface of the integrated solution is smaller than the surface of the discrete solution, so we had to compensate for this. Having two LEDs, driven from a 3 V supply, adds 156 μW to the overall power dissipation, which is almost negligible compared to the total power dissipation. We sampled the ADC at a rate of 100 Hz, which is common in a wearable system. In addition, we measured at a 500 Hz sampling rate, which is often used for systems with clinical performance. The data recording was taken under

Figure 3. Individual PPG waveforms (±125 data points around the local maximum) were extracted and plotted on top of each other (blue dotted line). The ensembled averages of the waveforms are shown by the red lines. The figure demonstrates the fundamental similarity between the PPG signals recorded by the ADPD188GG and the ADPD107 discrete solution.

considered a gold standard among the optical front ends on the market, and due to its good performance, it is being used in many medical products. DataSenseLabs Ltd. has a lot of experience with the ADPD107. However, because a fully integrated optical module in certain use cases has advantages, it started to work with them and did a comparative analysis, comparing the ADPD107’s performance with the ADPD188GG integrated optical module. In the following sections you will read more about the test setup, configurations, and results.

TesT seTup and daTa ColleCTion

For the optical comparison, raw PPG readings were recorded, simultaneously with the ADPD188GG and ADPD107, over a period of two minutes. For the ADPD188GG setup, the standard evaluation board was used, where the ADPD107 was part of the optical system inside the wearable demo platform (EVAL- HCRWATCH). Both systems have been controlled by Analog Devices’ user interface applications wavetool software. For the test, configuration settings were optimised to achieve the highest signal quality. We kept the configuration of the AFEs, including


the same circumstances as a normal smartwatch or fitness tracker, with the optical sensor attached to the top of the wrist. Because the microcirculation and vasoconstriction properties can be slightly different between the subdermal layers of the dominant hand and the nondominant hand, the recordings were repeated on both wrists by both optical systems. Datasets collected from the left and right wrists were analysed and compared carefully to avoid position- specific influence on the signal quality. PPG datasets were recorded on 11 different users (subjects) while they were seated and under the same ambient light intensity conditions.

daTa analysis and sTaTisTiCs

It is very important to take a comparative approach because signal quality validation does not only mean hard science signal processing, data analysis, and statistics, but it also has to do with what the market and end user are expecting. To be successful in the wearable market you need well-defined use cases and a clear objective of what outcome you want to get from the optical signal. Optical heart rate monitors are strongly connected to fitness tracking and wellness monitoring applications, but there are many use cases where optical technology can be found in medical grade systems. In a fitness, health informatics, or medical related use case, the accuracy of the peak detection algorithm depends primarily on the raw data quality around the local maximum of the PPG signal. Accurate peak detection is not only the principal of heart rate or HRV measurement, it is also extremely important for PPG-based blood pressure estimation detection. So, the designer must choose the sensor platform that gives the best physical signal quality if the finally extracted and calculated PPG signals are supposed to support health related applications. Comparative measurement configuration and data analysis was designed and carried out based on the Biosignal Metrology patent (pending ID: P1900302) owned by János Pálhalmi.1

The Final ResulTs

To support a peak detection algorithm, baseline fluctuations within the PPG raw data can easily be subtracted and filtered. In parallel, high signal quality around the peaks is needed at the raw data level to extract the targeted results as mentioned above. That is why in this study we focused on the comparative analysis of the

Figure 4. The magnitude squared coherence between the ensembled averages of the two compared PPG signals is shown by the colour intensity plot in the time and frequency domains. Direction of the arrowheads is proportional with the phase difference between the signals. The horizontally pointing arrowhead to the right refers to no phase difference between the signals.1

April 2021 Instrumentation Monthly

1 János Pálhalmi. Biosignal Metrology patent, P1900302.

Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48  |  Page 49  |  Page 50  |  Page 51  |  Page 52  |  Page 53  |  Page 54  |  Page 55  |  Page 56  |  Page 57  |  Page 58  |  Page 59  |  Page 60  |  Page 61  |  Page 62  |  Page 63  |  Page 64  |  Page 65  |  Page 66  |  Page 67  |  Page 68  |  Page 69  |  Page 70  |  Page 71  |  Page 72  |  Page 73  |  Page 74