A Telespirometer for the Developing World
Round 1
Reviewer 1 Report
Telemedicine was originally created as a way to treat patients who were located in remote places, far away from local health facilities or in areas of with insufficiency of medical professionals. The typical concept realization of mobile patient telemonitoring system consists of portable ultralow power device with battery supply that is worn during the study. It records one or several biomedical signals and/or parameters. The device includes standard analogue and digital circuits – sensors/electrodes, amplifiers, filters, analogue-to-digital converter, microcontroller. The data transfer from the patient module to the end recipient is performed through the mobile networks (GSM, GPRS, UMTS, 4G).
In the focus of this paper is one non-widely used approach for out of hospital diagnostic of chronic obstructive pulmonary disease (COPD). The presented constructive solution (telespirometer) is simple and relative inexpensive and allows to perform diagnostics even from non-specialist. The author has described in detail the construction and functionality of the sensor element - Fleisch pneumotachograph. It should be underline the correct theoretical analysis of the physical processes and parameters relating to the functionality of the spirometer, as well as for the accuracy of breathing volumes measurement during inhalation and exhalation.
One innovative and very practical solution is the use of standard voice communication via cellphone for transmission of the obtained diagnostic data. In this way, various telemedicine devices could be used, especially in the rural and isolated regions.
The content of the paper is well structured. The abstract is clear, well written and correctly presents the content of the article.
Critical comments, questions and proposals to the authors:
The reference sources cited are relatively old - from 8-9 years ago. The current state of telemedicine applications, especially those using sound transformation (sonification, audiofication) of biomedical signals and audio communication through mobile devices (GSM), should be presented; Ln157-Ln164: It is not clear how the calculated maximum differential pressure of 6.8 inH2O, could be measured with pressure transducer with a full scale range +/- 2.5 inH2O; How will the environmental parameters, in particular temperature, affect the measurement accuracy? Is it necessary to perform periodic calibration of the device (in situ), e.g. when evaporating liquid from the tube? It would be better to provide more parallel data from measurements with proposed device compared to measurements with other commonly used in clinical practice certified devices.
Author Response
The reference sources cited are relatively old - from 8-9 years ago. The current state of telemedicine applications, especially those using sound transformation (sonification, audiofication) of biomedical signals and audio communication through mobile devices (GSM), should be presented; Ln157-Ln164:
A number of new, more recent references have been included in the paper, they are highlighted in red. [6], [11], [16], [17], [19]. Of particular interest is
[6] C. Hernandez, J. Mallow, and G. Narsavage, "Delivering telemedicineinterventions in chronicrespiratory disease," Breathe, vol. 10, pp. 199-212, 2014.
[11] F. Burgos, C. Disdier, E. deSantamaria, B. Galdiz, N. Roger, M. Rivera, et al., "Telemedicine enhances quality of forced spirometry in primary care," European Respiratory Journal, vol. 39, pp. 1313-1318, 2012.
[16] J. Ong. (2019) Medical Detection and Diagnosis with Smartwatches. Smartwatch News. Available: https://www.smartwatches4u.com/medical-detection-and-diagnosis/
[17] J. Nosta. (March 26, 2018) A Smart Device, Ultrasound, And Telemedicine Combine To Drive A New Level of Collaboration And Care. Forbes. Available: https://www.forbes.com/sites/johnnosta/2018/03/26/a-smart-device-ultrasound-and-telemedicine-combine-to-drive-a-new-level-of-collaboration-and-care/#7d2772f83a2c
[19] P. Zhou, L. Yang, and Y.-X. Huang, "A Smart Phone Based Handheld Wireless Spirometer with Functions and Precision Comparable to Laboratory Spirometers," Sensors (Basel), vol. 19, 2019.
It is not clear how the calculated maximum differential pressure of 6.8 inH2O, could be measured with pressure transducer with a full scale range +/- 2.5 inH2O;
During calibration, the differential pressure would only be a fraction of this maximum because we didn’t have any means to generate faster flow rates with the required stability. For this reason a more sensitive transducer with a smaller full-scale range was considered more appropriate.
How will the environmental parameters, in particular temperature, affect the measurement accuracy?
The overall accuracy of the device will be affected somewhat by changes in temperature and the resulting condensation in the Fleisch cell. However, as these effects are proportional to the differential temperature between that of the expelled air and the ambient, operation in the tropics where the differential is smaller than it would be in a more temperate climate reduces the error. Measurements conducted by Miller [41] show an over-reading of 5% at an ambient temperature of 14°C reducing to 3% at 26°C. By extrapolation, the error will have reduced to an acceptable 2% for an ambient temperature of 32°C.
[41] M. Miller and T. Sigsgaard, "Prevention of thermal and condensation errors in pheumatachographic recordings of the maximal forced expiratory manoeuvre," Europeran Respiratory Journal, vol. 7, pp. 198-201, 1994.
Is it necessary to perform periodic calibration of the device (in situ), e.g. when evaporating liquid from the tube?
The proposed operational procedure is as follows:
The liquid in the manometer is topped up (or ideally replaced) so that the levels are half way up the coils.
It would be better to provide more parallel data from measurements with proposed device compared to measurements with other commonly used in clinical practice certified devices.
Unfortunately when the paper was written, we didn’t have access to other certified devices. However, we ran a comparison between our device and that described by Zhou et al. [19]
[19] P. Zhou, L. Yang, and Y.-X. Huang, "A Smart Phone Based Handheld Wireless Spirometer with Functions and Precision Comparable to Laboratory Spirometers," Sensors (Basel), vol. 19, 2019.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
First of all, I want to congratulate the authors for their efforts done in the manuscript. Their work and a good understanding of the current problems in the application of e-health in the developing countries. The authors present an innovative solution to perform an essential test in a low-cost version. Following I include some comments aimed to enhance the quality of the paper:
1 In the introduction, authors should include a short paragraph describing the aim of the paper and a short paragraph detailing the structure of the paper (the sections).
2- At the beginning of each section, the authors must add a short paragraph that outlines the content of the section.
3- To the extent possible, I suggest the authors include in the pictures of their design a ruler, a coin, or other elements to facilitate the readers the comprehension of the size of the created items.
4 Section 4 should be renamed as conclusions
5- Figure 12 must be explained in more detail
6- I suggest to the authors to include a new section describing the related work In this section authors can compare their system with current systems for measuring the
Author Response
I would like to thank this reviewer particularly for the positive suggestions which I believe has improved the paper significantly
1 In the introduction, authors should include a short paragraph describing the aim of the paper and a short paragraph detailing the structure of the paper (the sections).
1.3 Aim and Structure
The aim of this research is to develop the mechanical components, the electronics and the relevant cellphone interfaces that could be used to provide a blueprint for a tele-connected Fleisch pneumotachograph to allow its manufacture in the developing world for as little as $1.
Section 2 starts with a description of the development of a simple Fleisch cell made from syringe needles. Subsequent analysis suggests that the flow regime would remain below the critical Reynolds number for flows in excess of 10 lit/s allowing for accurate flow rates to be measured. The method of using two radio frequency oscillators to generate an acoustic signal with a frequency proportional to flow rate using a simple manometer is then described.
Section 3 describes the calibration process using a simple-to-make spirometer and a sensitive differential pressure transducer. This section confirms that the measured performance of the Fleisch pneumotachograph is within a factor of two of that derived theoretically. It goes on to examine the expiratory flow data with the objective of identifying COPD. The complete system including the low cost pressure transducer and acoustic link are shown to generate waveform data identical to those generated by the expensive differential pressure transducer.
Section 4 concludes that it is possible to design and manufacture a telespirometer from “junk box” medical and electronic components that are easily available in the developing world.
2- At the beginning of each section, the authors must add a short paragraph that outlines the content of the section.
This was done in some cases, but I felt that it would clutter the paper with too much detail
3- To the extent possible, I suggest the authors include in the pictures of their design a ruler, a coin, or other elements to facilitate the readers the comprehension of the size of the created items.
For Fig 3 and Fig 4, the text is clear that the Fleisch cell has a diameter of 25mm, so it was considered to be unnecessary to include a scale. However in the case of Fig 9, a cm scale has been included as there is no easy reference to the size of the circuit
4 Section 4 should be renamed as conclusions
This has been done, and the section has been updated with any changes highlighted in red.
5- Figure 12 must be explained in more detail
This has been done with the updates highlighted in red.
6- I suggest to the authors to include a new section describing the related work In this section authors can compare their system with current systems for measuring the
3.2 Comparison with a Lilly based Spirometer
It is interesting to compare the calibration results in this paper with those obtained using a precision smart-phone based spirometer [19]. Conceptually, the two systems are similar with Zhou et al. selecting to use a commercial Lilly screen with a flow restrictor (R=3.3×104 Pa per m3/s) matched to the differential pressure sensor dynamic range of 0 to 500 Pa, to cater for flow rates up to 15 lit/s. Whereas in this paper the Fleisch cell provided a fixed pressure drop (R=8.01×104 Pa per m3/s ) with a differential pressure sensor +/- 2.5 inH2O (+/- 622 Pa) to cater for flow rates up to a maximum of 7.7 lit/s. Both of the differential pressure transducers provided analogue outputs that could be digitized to the required precision.
The Zhou et al. system measured the analogue output with a precision of 1 mV corresponding to a pressure differential of 0.133 Pa or a flow rate of 4 ml/s, whereas our system uses the Labjack 12 bit ADC over a range of +/-5 V, corresponding to a precision of 2.4 mV (0.6 Pa) and hence a flow rate of 7.4 ml/s.
The Lilly based system provided a measured linear response for flow rates from 1 lit/s up to 15 lit/s whereas the Fleisch was only measured for flow rate between 0.1 and 2.5 l/s due to so a comparison is not possible.
[19] P. Zhou, L. Yang, and Y.-X. Huang, "A Smart Phone Based Handheld Wireless Spirometer with Functions and Precision Comparable to Laboratory Spirometers," Sensors (Basel), vol. 19, 2019.
Author Response File: Author Response.pdf
Reviewer 3 Report
The paper presents a low-cost solution for COPD detection and is of interest to many in the developing world.
The equation (5) has pressure drop([M]/[L]2) in left hand side and flow rate([L]3/[T]) in right hand side. They do not have the dimensional match. Please check your equations above and correct. Fig. 11 : Vertical Axis: Flow(V) & Volume(V.s) - need to be corrected. How do you filter out the noise generated during the transmission from cell phone to smart phone? Conclusion ? in general the paper needs a careful major review focusing on the points the author wants to highlight. The theoretical analysis needs to be developed carefully to match the operation of the physical device made.
Author Response
The equation (5) has pressure drop([M]/[L]2) in left hand side and flow rate([L]3/[T]) in right hand side. They do not have the dimensional match. Please check your equations above and correct
If you examine equation (3), the standard form of the Poiseuille equation which relates the change in pressure to the flow rate. There are a number of constants that ensure that the dimensions match
The constant of proportionality in equation (5) lumps all of the constants together
. Fig. 11 : Vertical Axis: Flow(V) & Volume(V.s) - need to be corrected.
These axes are correctly labelled as per the units in which the variables were measured (volts), or after integration (V.s). It is important that they remain as such because the final integrated output (equivalent to Volume) is used to generate the required calibration constant.
How do you filter out the noise generated during the transmission from cell phone to smart phone?
These are filtered using a 3rd order Butterworth filter with a cutoff frequency of 15 Hz to produce an estimate of the signal frequency as a function of time.
The paper has been updated to reflect this change.
Conclusion ?
Done
in general the paper needs a careful major review focusing on the points the author wants to highlight.
The paper has had a major review to include a section describing the Aim and contents of the various sections. This goes somewhat towards addressing the important contributions of the paper. Please see all of the new sections highlighted in red.
The conclusions section has been renamed and updated
The theoretical analysis needs to be developed carefully to match the operation of the physical device made.
I believe the theoretical analysis has been developed as well as possible, given the complexity of analysing the airflow through both the needle capillaries and the spaces surrounding them. The fact that theory and practise match too within a factor of 2 is considered to be an excellent result given this complexity.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The corrections made are adequate and contribute to the quality and readability of the article.
Reviewer 3 Report
Most of the previous recommendations have been addressed.
