A Phantom-Based Study of the X-Ray Fluorescence Detectability of Iron, Copper, Zinc, and Selenium in the Human Blood of Superficial and Cutaneous Vasculature

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

In the paper authors present the possibilities of non-invasive detection of certain essential trace elements in cutaneous blood using the XRF technique and a phantom model simulating skin vasculature.

Comments and suggestions to the authors

Section 1.  Introduction

1. The authors used the XRF technique to analyze four essential trace elements, although more sensitive techniques are available, e.g. TXRF. Please comment.
2. Lines 48–69: Diagnostic methods of iron deficiency or overload are discussed in detail in this section. The aim of this study is to assess the detectability of the four essential elements. Therefore, this section should be shortened. In addition, relevant data on physiological concentrations of all the mentioned essential elements in human whole blood are provided in Table 1.
3. Line 85: "X-ray fluorescence elemental concentration measurements are simpler, faster, and less costly than ICP-MS, and they typically involve a low radiation dose."
   The authors should elaborate further on the advantages of the XRF technique compared to ICP-MS but also mention its limitations.
4. Lines 136–137: "Distribution and expected concentrations of the four trace elements in human skin were not simulated in this study."
   Detailed information on the determination of the four trace essential elements in human skin is given in the Introduction section (lines 112–126). Why? No detailed information is required if the distribution and expected concentrations of these trace elements in human skin are not simulated in this study.
5. Lines 149–150: " Low cost, easy-to-use, and access are important characteristics of novel instruments for clinical applications".                                                                                Why is this technique considered low cost when according to this study only measurements of Fe concentration in cutaneous blood have the possibility of clinical applications.
6. Lines 151–153: "Our results indicate that in vivo rapid, low-cost, and low-dose XRF measurement of Fe concentration in cutaneous blood is possible but requires additional instrumental optimization and the development of accurate calibration methods."

         Instead of this conclusion, the authors should state the objective of this study.

Subsection 2.2. Standard Solutions and POM Phantoms

Since this study represents a novel method with potential clinical application, additional validation parameters such as specificity, accuracy, and precision need to be assessed. Certified reference material should be used to assess the accuracy of the method. These validation data are missing. Please comment.

Subsection 3.1. Calibration Lines and Detection Limits

• Lines 421–428: Fig. 6 and Fig. 8
  The calibration lines for Cu show low correlation coefficient values (R² = 0.28, 0.57, and 0.52).
• Figure 7: correlation coefficient value for Zn is R² = 0.49.
  Please explain these low R² values for Cu and Zn.

Author Response

Comment 1: Section 1.  Introduction.

The authors used the XRF technique to analyze four essential trace elements, although more sensitive techniques are available, e.g. TXRF. Please comment.

Response 1: TXRF is the most sensitive XRF technique, but similarly to ICP-MS can only be applied on ex vivo blood samples. This was highlighted in the first sentence of the 8-th paragraph of the Introduction section and references were provided. 

Comment 2: Lines 48–69: Diagnostic methods of iron deficiency or overload are discussed in detail in this section. The aim of this study is to assess the detectability of the four essential elements. Therefore, this section should be shortened. In addition, relevant data on physiological concentrations of all the mentioned essential elements in human whole blood are provided in Table 1.

Response 2: This section is needed to articulate the motivation for this study.

Comment 3: Line 85: "X-ray fluorescence elemental concentration measurements are simpler, faster, and less costly than ICP-MS, and they typically involve a low radiation dose."
The authors should elaborate further on the advantages of the XRF technique compared to ICP-MS but also mention its limitations.

Response 3: Strengths and weaknesses of ICP-MS were included in the 6-th paragraph of the Introduction section. The following sentence states the main advantage of XRF: its capability to perform in vivo measurements, whereas most other mass spectrometric techniques cannot. We also added the following sentence later in the same paragraph: “Also, determination of accurate elemental concentrations from experimental data requires a robust calibration method”.

Comment 4: Lines 136–137: "Distribution and expected concentrations of the four trace elements in human skin were not simulated in this study."
Detailed information on the determination of the four trace essential elements in human skin is given in the Introduction section (lines 112–126). Why? No detailed information is required if the distribution and expected concentrations of these trace elements in human skin are not simulated in this study.

Response 4: The elemental composition of the four elements in the skin is relevant because separation of the XRF signal of elements in the skin from those in the cutaneous blood from spectral data obtained from in vivo measurements would be difficult.

Comment 5: Lines 149–150: " Low cost, easy-to-use, and access are important characteristics of novel instruments for clinical applications".                                                                               

Why is this technique considered low cost when according to this study only measurements of Fe concentration in cutaneous blood have the possibility of clinical applications.

Response 5: This technique is low-cost in comparison to other lab-based methods. It is not just the instrumental cost, but also the maintenance and operational cost associated with ICP-MS as mentioned in the 5-th paragraph of the Introduction section.

Comment 6: Lines 151–153: "Our results indicate that in vivo rapid, low-cost, and low-dose XRF measurement of Fe concentration in cutaneous blood is possible but requires additional instrumental optimization and the development of accurate calibration methods." Instead of this conclusion, the authors should state the objective of this study.

Response 6: We agree with the reviewer that stating an objective is more appropriate in this section. We replaced this phrase with “Our objective was to measure the detection limits of four elements (Fe, Cu, Zn, and Se) in human blood using phantoms that simulated the x-ray attenuation of skin and blood tis-sues expected during an in vivo measurement.”

Comment 7: Subsection 2.2. Standard Solutions and POM Phantoms

Since this study represents a novel method with potential clinical application, additional validation parameters such as specificity, accuracy, and precision need to be assessed. Certified reference material should be used to assess the accuracy of the method. These validation data are missing. Please comment.

Response 7: The validation parameters mentioned are applicable to a calibration method which remains to be determined in future research. This paper is concerned only with detectability and these parameters cannot be determined. While there is a wide range of certified reference materials, to the best of our knowledge, there are no reference materials or tissues of the skin and its vasculature, hence, this phantom study.

Comment 8: Subsection 3.1. Calibration Lines and Detection LimitsLines 421–428: Fig. 6 and Fig. 8
The calibration lines for Cu show low correlation coefficient values (R² = 0.28, 0.57, and 0.52).Figure 7: correlation coefficient value for Zn is R² = 0.49.
Please explain these low R² values for Cu and Zn.

Response 8: The low correlation coefficients for Cu are due to large variations in the Cu data. In turn, these were caused by the large external Cu contamination mentioned in subsection 4.1. The XRF signal of Al collimator Cu exceeded that of Cu in the prepared aqueous solutions. This contamination also inadvertently affected Zn detection. As can be noticed in the inset plot of the right-hand side of figure 4, the Zn Ka overlapped with the right-hand side and left-hand side tails of the larger Cu Ka and Cu Kb peaks. The R² value of 0.28 was a typo that was now corrected . Subsequently, all linear fitting parameters and correlation coefficients were verified.

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript metrology-3519545 titled “A phantom-based study of X-ray fluorescence detectability of iron, copper, zinc, and selenium in the human blood of superficial and cutaneous vasculature” shows the determination of four elements (iron, copper, zinc and selenium) by using ED-XRF in cutaneous blood as alternative to ICP-MS.

The authors report the analytical methodology used, the calibration and the detection limit intervals.

Moreover, they studied the radiation dose and equivalent dose to skin, and they show that the doses were below those of common radiological procedures.

The authors conclude their manuscript by asserting that the applications will require further instrumental development and finding a calibration method.In my opinion, the research is innovative and merits to be published because this will improve knowledge in the field studied.

Therefore, the manuscript is a topic original and relevant to the studied field. The research could full specific gaps in the determination of iron in blood.The authors report a simpler, faster and cheaper method to determine iron in blood compared to the ICP-MS method.

The authors are aware of the analytical and safety difficulties, but scientific research is for this very purpose, that is, to try new ways and optimize them together with the final protocol.The conclusions are consistent with the evidence and arguments presented (in the introduction, in the results and in the discussions).

The References are appropriate with the arguments presented by authors.

I want to congratulate the authors for the excellent research work carried out.

Therefore, I believe that the manuscript can be accepted for publication after the following minor revisions:

Introduction. The authors write “… Essential trace elements are chemical elements required in very small or trace concentrations for the development and physiology of all organisms.”. The sentence is correct, but the authors should add a reference.

Lines 42-44. The authors write “More recent investigations also found links between essential trace elements levels in the blood and other conditions such as autism spectrum disorders, neurodegenerative conditions”. The sentence is correct, but the authors should add a reference such as 10.3390/ijms22042038.

Lines 85-86. The authors write “X-ray fluorescence elemental concentrations measurements are simpler, faster, and 85 less costly than ICP-MS and they typically have a low radiation dose”. The sentence is correct, but the authors should add some references.

Table 7, Table 8 and table 9 can be merged into one table.

The authors should replace "x-ray" with "X-ray" throughout the manuscript.

The authors should replace "water solutions” with "aqueous solutions" throughout the manuscript.

Keywords. The authors must add “Selenium”.

Abstract. The authors write “Inductively coupled plasma mass spectrometry (ICP-MS) has been employed for such measurement…..”. In my opinion, the authors must write “Inductively coupled plasma mass spectrometry (ICP-MS), inductively coupled plasma atomic emission spectroscopy (ICP-AES) and graphite furnace atomic absorption spectrometry (GF-AAS) have been employed for such measurement…..

Abstract. The authors write “Detection limit intervals in mg/L were: (36, 100), (14, 40), (3.7, 10), and (2.1, 3.4) for….”. In my opinion the authors should write “Detection limit intervals in mg/L were: (36 - 100), (14 - 40), (3.7 - 10), and (2.1 - 3.4) for…..”.

Line 220. Replace “Five solutions containing unique Fe, Cu, Zn, and Fe concentrations were prepared …” with “Five solutions containing unique Fe, Cu, Zn, and Se concentrations were prepared … ”.

Line 232. Replace “1 ml” with “1 mL”.

Line 233. Replace “25 ml” with “25 mL”.

Line 239. Replace “polyoxymethylene (POM) plastic” with “polyoxymethylene plastic”. In my opinion, POM is superfluous.

Line 262. Replace “water (H₂O)” with “water”. H₂O is superfluous.

Conclusions. The authors should replace “Detection limit range, in mg/L, were: (36, 100), (14, 40), (3.7, 10), and (2.1, 3.4) for….” with “Detection limit range, in mg/L, were: (36 - 100), (14 - 40), (3.7 - 10), and (2.1 - 3.4) for….”.

Author Response

Comment 1: Introduction. The authors write “… Essential trace elements are chemical elements required in very small or trace concentrations for the development and physiology of all organisms.”. The sentence is correct, but the authors should add a reference.

Response 1: References 1 to 3 are already included in the manuscript at the end of the sentence following the first sentence of the Introduction section.

Comment 2: Lines 42-44. The authors write “More recent investigations also found links between essential trace elements levels in the blood and other conditions such as autism spectrum disorders, neurodegenerative conditions”. The sentence is correct, but the authors should add a reference such as 10.3390/ijms22042038.

Response 2: Two papers (references 14 and 15) are already included in the current version of the manuscript. We limited strictly to papers suggesting correlations between blood levels and the mentioned conditions.   

Comment 3: Lines 85-86. The authors write “X-ray fluorescence elemental concentrations measurements are simpler, faster, and less costly than ICP-MS and they typically have a low radiation dose”. The sentence is correct, but the authors should add some references.

Response 3: We added four references as suggested. Since our reference software (Zotero) was inactivated in the current digital version of our paper, we had to update all references manually based on an updated references of our original Word version.

Comment 4: Table 7, Table 8 and table 9 can be merged into one table.

Response 4: This revision was included, and the subsequent two tables (now 8 and 9) were renumbered and references in the text were modified accordingly.

Comment 5: The authors should replace "x-ray" with "X-ray" throughout the manuscript.

Response 5: The manuscript was modified to include this revision.

Comment 6: The authors should replace "water solutions” with "aqueous solutions" throughout the manuscript.

Response 6: The manuscript was modified to include this revision.

Comment 7: Keywords. The authors must add “Selenium”.

Response 7: The manuscript was modified to include this revision.

Comment 8: Abstract. The authors write “Inductively coupled plasma mass spectrometry (ICP-MS) has been employed for such measurement…..”. In my opinion, the authors must write “Inductively coupled plasma mass spectrometry (ICP-MS), inductively coupled plasma atomic emission spectroscopy (ICP-AES) and graphite furnace atomic absorption spectrometry (GF-AAS) have been employed for such measurement…..

Response 8: The abstract was modified to include this revision.

Comment 9: Abstract. The authors write “Detection limit intervals in mg/L were: (36, 100), (14, 40), (3.7, 10), and (2.1, 3.4) for….”. In my opinion the authors should write “Detection limit intervals in mg/L were: (36 - 100), (14 - 40), (3.7 - 10), and (2.1 - 3.4) for…..”.

Response 9: The abstract was modified to include this revision.

Comment 10: Line 220. Replace “Five solutions containing unique Fe, Cu, Zn, and Fe concentrations were prepared …” with “Five solutions containing unique Fe, Cu, Zn, and Se concentrations were prepared … ”.

Response 10: The line was modified to include this revision.

Comment 11: Line 232. Replace “1 ml” with “1 mL”.

Response: The correction was included.

Comment 12: Line 233. Replace “25 ml” with “25 mL”.

Response 12: The correction was included.

Comment 13: Line 239. Replace “polyoxymethylene (POM) plastic” with “polyoxymethylene plastic”. In my opinion, POM is superfluous.

Response 13: We removed POM appearing in this sentence, but we kept the abbreviation since it appears throughout the manuscript.

Comment 14: Line 262. Replace “water (H2O)” with “water”. H2O is superfluous.

Response 14: The line was modified to include this revision.

Comment 15: Conclusions. The authors should replace “Detection limit range, in mg/L, were: (36, 100), (14, 40), (3.7, 10), and (2.1, 3.4) for….” with “Detection limit range, in mg/L, were: (36 - 100), (14 - 40), (3.7 - 10), and (2.1 - 3.4) for….”.

Response 15: The Conclusions section was mo

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript explores the feasibility of measuring the concentrations of four elements—iron, copper, zinc, and selenium—typically found in human blood, using in vivo X-ray fluorescence (XRF) analysis. The authors estimate the detection limits through experimental measurements using custom phantoms that imitate vasculature. These phantoms are made of polyoxymethylene (POM) plastic and are filled with solutions of varying concentrations. A custom tabletop XRF spectrometer, equipped with an X-ray source and a polycapillary optical element, is used for the analysis.

While the topic is of significant interest, and the study is well-designed from a metrological perspective – fully aligning with the focus of the journal – several critical remarks can be made regarding the manuscript.

1. Line 175. “The integrated PXL was 10 cm in length with a 1 cm outer diameter…” As I understand, a semi-lens was used in this X-ray source to modify the divergent beam into a quasi-parallel one, rather than a focusing lens. Please clarify this point in the text.
2. Line 190. “… 106 photons/s were given by the manufacturer.” It seems that "10^6" should be used instead of "106." Please verify and correct this.
3. Figure 2. Please indicate the objects in the figure (e.g., detector, phantom, lens, etc.) using arrows or circles for clarity.
4. Lines 265-269. “Using the x-ray linear attenuation coefficient values of table 5, one can compute that blood is, on average, about 6.7% more attenuating than water solution 5, and POM is about 7.4% more attenuating than human skin. Therefore, the more attenuating POM is approximately compensated by the less attenuating water solutions in the 5 to 20 keV photon energy range”. This statement implies that the attenuation effects of POM and the water solution counterbalance each other, which is not entirely accurate. Is this compensation correct for solutions with different concentrations as well? It would be clearer to state that the phantom closely mimics the attenuation properties of skin and blood, rather than suggesting an exact compensation between POM and the water solutions.
5. Please check the font size throughout the manuscript. For example, the text in lines 322-345 appears to differ noticeably from the surrounding text. Ensure uniformity across the manuscript.
6. Table 6 and Figure 4. It would be much more informative to include two additional plots in Figure 4. The first should show the spectrum for the same phantom (0.6 mm POM cup with insert) infilled with pure water (formatted like the current right plot in Figure 4). The second should present the differential spectrum between the spectrum measured for solution 5 and pure water. Based on the calibration lines presented below in the text, it seems that the authors measured the spectrum for a 0 concentration of elements (pure water). Including these two additional plots would provide a clearer understanding of the contribution of scattered X-ray fluorescence from the setup components. This would also help to explain why the detection limit (DL) for copper is so low that is discussed later in the manuscript.
7. Line 479. “As seen in the figure 9 plot …” There is no Figure 9.
8. Lines 490-499. It is not entirely clear which specific lower peak area is being discussed here. Is this referring to the drop in signal at the last point on the right in the middle plot panels of Figure 6? If so, I would suggest a simpler explanation. Upon closer inspection of the middle plots in Figures 5–8, you can observe a similar drop at this specific point in almost all of them (6 out of 8 plots), including for the Fe and Zn Kα lines. Have you checked the deadtime of the detector during measurements at these points? Since the middle plots correspond to the phantom with the highest "yield" of fluorescent X-rays (due to the thin wall and full solution fill) and the highest concentration of the solution, the intensity of X-rays reaching the detector is highest here. It seems that, at this point, the detector starts to saturate, and the integral deadtime becomes too high, leading to undercounting of the X-ray quanta hitting the detector. If you were to continue measurements with solutions of even higher concentration, you would likely observe a continuous drop in the signal. This explanation seems more plausible than assuming a sudden and unusual malfunction of the analyzer. The chemical reaction is also unlikely to be the cause, as it would not change the concentration of atoms in the solution. The only exception would be if a reaction led to the formation of a precipitate in the solution, but in that case, you would have observed the precipitate. I recommend revisiting this point for further verification.
9. Table 12. Please consider reformatting the top row of the table to improve clarity. This will help readers more easily interpret the data presented.
10. Lines 546-553. Please remove this section from the article, as it lacks clarity and is unnecessary. First, the limitation discussed refers to the equivalent dose, not the effective dose. Therefore, it should not be multiplied by the skin weighting factor but only by the radiation weighting factor (which equals 1 for X-rays). The effects on specific organs are already accounted for in these limitations (e.g., the whole-body dose limit is 1 mSv, compared to 50 mSv for the skin). Consequently, in your approach, you should compare 2.1 Gy * 1 = 2.1 Sv with the 50 mSv limit. However, this comparison is not necessary because the ICRP does not regulate exposure during medical procedures, and these exposures are often much higher than the limits set by the ICRP. Therefore, your comparison with other medical investigations in the first paragraph of this section is sufficient. The calculations in lines 546-553 are purely mathematical manipulations. You already have an upper estimate of the dose for your session (42-48 mGy), so you can compare it with other studies. For instance, you mentioned 8.45 mSv and 9.37 mSv for in vivo XRF tibia bone lead studies. This shows that your estimate is of a similar order of magnitude, though few times higher. However, since you are estimating the upper limit of the dose and not optimizing the scheme, it is possible to reduce this value in future iterations and should be addresses in future researches. Please be more precise when discussing radiation dose limits, as it an issue of public health.

In summary, I have mixed feelings about this manuscript. On one hand, it presents a thorough experimental study with a strong methodological approach, with only a few critical remarks. As such, it is well-suited for the Metrology journal. The detailed description of the experiment, peak fitting, and calibration processes will likely be of interest to researchers working with etalon-based quantitative XRF, particularly those dealing with solutions.

On the other hand, I have significant doubts regarding the applicability of in vivo XRF for measuring elemental concentrations in human blood. Any serious medical examination typically involves blood collection and analysis at an early stage, making ex vivo XRF of blood a more practical and viable option. Ex vivo XRF would avoid issues such as X-ray scattering and absorption in the skin, the contribution of X-ray fluorescence from elements like Zn and Fe in the skin (as noted by the authors), and unnecessary additional radiation exposure to the patient. While the ~45 mGy dose in this study is not enormous, it is concentrated in a very small volume (a spot approximately 1.5 mm in size), which could lead to significant consequences. However, this is a question best addressed by radiobiology specialists, as I am not an expert in this field. In the case of XRF tibia bone studies, the in vivo approach is justified since ex vivo analysis is impossible. The same holds true for CT (µCT) studies. This approach could also be useful for skin investigations, as the authors propose. However, for blood, I do not believe in vivo XRF is a viable or promising approach.

Additionally, the conclusions require substantial revision as they do not fully align with the results presented.

Lines 633-635. “… and the equivalent dose to skin was below the annual limit for planned radiation exposures of the general public.” This is not applicable, as noted previously. Once again, when accounting for organ weighting factors, you are calculating the effective dose, not the equivalent dose.

Lines 636-649: “Fe was the only element with detection limits significantly lower than the median Fe human blood concentration of ~480 mg/L indicating the possibility of clinical applications.”  This does not indicate the possibility of clinical application, particularly due to the issue of signal contamination from skin-deposited Fe. It suggests the potential for further development of instrumentation for medical applications rather than immediate clinical applicability.

Lines 643-645: “Therefore, in vivo measurements of their concentrations in cutaneous blood will depend on the success of instrumental modifications tailored to mitigate external XRF signal contamination and x-ray scatter.” This is not entirely accurate. It is not just a matter of instrumental modification. At least for Zn, it also depends on the ability to separate signals from the blood and the skin. For copper, the experimental results do not reliably support the feasibility of such an application. One can only hope that ongoing developments and modifications in

Finally, I would conclude that the article can be considered for publication in the Metrology journal after thorough revision. Remarks 8 and 10 are particularly crucial and should be addressed carefully, along with a strict revision of the conclusions to ensure they accurately reflect the results presented in the manuscript.

Author Response

We would like to thank the reviewer for hi/her careful and pertinent comments, suggestions, and corrections. 

Our responses to reviewer's comments are below. 

Comment 1: Line 175. “The integrated PXL was 10 cm in length with a 1 cm outer diameter…” As I understand, a semi-lens was used in this X-ray source to modify the divergent beam into a quasi-parallel one, rather than a focusing lens. Please clarify this point in the text.

Response 1: No, polycapillary x-ray lens (PXL) it was a focusing one; the x-rays are convergent at distances below the focal distance and divergent at distances larger than the focal distance with a provided divergence of 76 milliradians. We added a clarifying sentence in this paragraph: “The x-rays converge towards a focal point at distances smaller than the focal length, and are divergent at larger distances.” We also moved the statement regarding focal distance value for a better flow in this paragraph.  

Comment 2: Line 190. “… 106 photons/s were given by the manufacturer.” It seems that "10^6" should be used instead of "106." Please verify and correct this.

Response 2: Yes, it was 10^6, in the original draft, the subscript was changed in subsequent manipulations of the manuscript. This was changed accordingly and verified in the final version.

Comment 3: Figure 2. Please indicate the objects in the figure (e.g., detector, phantom, lens, etc.) using arrows or circles for clarity.

Response 3: We modified figure 2 as follows: (1) we added a slightly better digital photograph of the experimental setup, (2) we added numbers on the figure which along with a legend identify the objects in the image.

Comment 4: Lines 265-269. “Using the x-ray linear attenuation coefficient values of table 5, one can compute that blood is, on average, about 6.7% more attenuating than water solution 5, and POM is about 7.4% more attenuating than human skin. Therefore, the more attenuating POM is approximately compensated by the less attenuating water solutions in the 5 to 20 keV photon energy range”. This statement implies that the attenuation effects of POM and the water solution counterbalance each other, which is not entirely accurate. Is this compensation correct for solutions with different concentrations as well? It would be clearer to state that the phantom closely mimics the attenuation properties of skin and blood, rather than suggesting an exact compensation between POM and the water solutions.

Response 4: As we mentioned in the text, the compensation is only approximate, not accurate. As indicated in table 5, the linear attenuation coefficient was computed for aqueous solution 5 which has the highest elemental concentrations (see table 3).  

Comment 5: Please check the font size throughout the manuscript. For example, the text in lines 322-345 appears to differ noticeably from the surrounding text. Ensure uniformity across the manuscript.

Response 5: Yes, we noticed that as well and corrected this error. The same font was used in the original manuscript. However, the original manuscript was edited by the journal editing team to fit their precise format. It is possible that in the process, the obvious font mishap might have occurred. We will verify the final format of the manuscript.  

Comment 6: Table 6 and Figure 4. It would be much more informative to include two additional plots in Figure 4. The first should show the spectrum for the same phantom (0.6 mm POM cup with insert) infilled with pure water (formatted like the current right plot in Figure 4). The second should present the differential spectrum between the spectrum measured for solution 5 and pure water. Based on the calibration lines presented below in the text, it seems that the authors measured the spectrum for a 0 concentration of elements (pure water). Including these two additional plots would provide a clearer understanding of the contribution of scattered X-ray fluorescence from the setup components. This would also help to explain why the detection limit (DL) for copper is so low that is discussed later in the manuscript.

Response 6: Yes, that is a good suggestion. Two additional plots were added as suggested and the figure caption was modified accordingly.  

Comment 7: Line 479. “As seen in the figure 9 plot …” There is no Figure 9.

Response 7: Yes, a mistake has occurred here. In an earlier manuscript version, we had a figure 9 plot of the DL values versus their average blood concentrations values. We modified the sentence to point out to values in renumbered table 8: “An inspection of table 8 values indicates that, measured Cu DLs were more than an order of magnitude higher than the median human blood concentration.”

Comment 8: Lines 490-499. It is not entirely clear which specific lower peak area is being discussed here. Is this referring to the drop in signal at the last point on the right in the middle plot panels of Figure 6? If so, I would suggest a simpler explanation. Upon closer inspection of the middle plots in Figures 5–8, you can observe a similar drop at this specific point in almost all of them (6 out of 8 plots), including for the Fe and Zn Kα lines. Have you checked the deadtime of the detector during measurements at these points? Since the middle plots correspond to the phantom with the highest "yield" of fluorescent X-rays (due to the thin wall and full solution fill) and the highest concentration of the solution, the intensity of X-rays reaching the detector is highest here. It seems that, at this point, the detector starts to saturate, and the integral deadtime becomes too high, leading to undercounting of the X-ray quanta hitting the detector. If you were to continue measurements with solutions of even higher concentration, you would likely observe a continuous drop in the signal. This explanation seems more plausible than assuming a sudden and unusual malfunction of the analyzer. The chemical reaction is also unlikely to be the cause, as it would not change the concentration of atoms in the solution. The only exception would be if a reaction led to the formation of a precipitate in the solution, but in that case, you would have observed the precipitate. I recommend revisiting this point for further verification.

Response 8: We disagree with the reviewer on this point. Only in the figure 6 plot, is there an unusual and statistically significant drop in the number of counts corresponding to the Cu and Se Kalpha peak areas. The detector dead times for all experiments were very small (below 1%), far from the saturation point proposed by the reviewer. A sudden drop of the number of counts registered by the detector would have affected the Fe and Zn peaks as well, and that was not the case.

We remain open to an alternative explanation.  

Comment 9: Table 12. Please consider reformatting the top row of the table to improve clarity. This will help readers more easily interpret the data presented.

Response 9: Yes, we agree. The symbolic notations we employed were too long, we made them shorter to better fit in the table top row.

Comment 10: Lines 546-553. Please remove this section from the article, as it lacks clarity and is unnecessary. First, the limitation discussed refers to the equivalent dose, not the effective dose. Therefore, it should not be multiplied by the skin weighting factor but only by the radiation weighting factor (which equals 1 for X-rays). The effects on specific organs are already accounted for in these limitations (e.g., the whole-body dose limit is 1 mSv, compared to 50 mSv for the skin). Consequently, in your approach, you should compare 2.1 Gy * 1 = 2.1 Sv with the 50 mSv limit. However, this comparison is not necessary because the ICRP does not regulate exposure during medical procedures, and these exposures are often much higher than the limits set by the ICRP. Therefore, your comparison with other medical investigations in the first paragraph of this section is sufficient. The calculations in lines 546-553 are purely mathematical manipulations. You already have an upper estimate of the dose for your session (42-48 mGy), so you can compare it with other studies. For instance, you mentioned 8.45 mSv and 9.37 mSv for in vivo XRF tibia bone lead studies. This shows that your estimate is of a similar order of magnitude, though few times higher. However, since you are estimating the upper limit of the dose and not optimizing the scheme, it is possible to reduce this value in future iterations and should be addresses in future researches. Please be more precise when discussing radiation dose limits, as it an issue of public health.

Response 10: We agree that we did not compute the effective dose, albeit this value would be very small, and it would depend on specific shielding of nonexposed body parts.

In subsection 3.2, we did compare our computed equivalent dose to literature values.

We agree we should remove this paragraph since the 0.01 weighting factor is meant for effective dose computations, not equivalent dose. And we also agree on the purpose of the ICRP regulations.

Comment 11: In summary, I have mixed feelings about this manuscript. On one hand, it presents a thorough experimental study with a strong methodological approach, with only a few critical remarks. As such, it is well-suited for the Metrology journal. The detailed description of the experiment, peak fitting, and calibration processes will likely be of interest to researchers working with etalon-based quantitative XRF, particularly those dealing with solutions. On the other hand, I have significant doubts regarding the applicability of in vivo XRF for measuring elemental concentrations in human blood. Any serious medical examination typically involves blood collection and analysis at an early stage, making ex vivo XRF of blood a more practical and viable option. Ex vivo XRF would avoid issues such as X-ray scattering and absorption in the skin, the contribution of X-ray fluorescence from elements like Zn and Fe in the skin (as noted by the authors), and unnecessary additional radiation exposure to the patient. While the ~45 mGy dose in this study is not enormous, it is concentrated in a very small volume (a spot approximately 1.5 mm in size), which could lead to significant consequences. However, this is a question best addressed by radiobiology specialists, as I am not an expert in this field. In the case of XRF tibia bone studies, the in vivo approach is justified since ex vivo analysis is impossible. The same holds true for CT (µCT) studies. This approach could also be useful for skin investigations, as the authors propose. However, for blood, I do not believe in vivo XRF is a viable or promising approach.

Response 11: We envisioned a dedicated device that could measure blood Fe without blood collection. Zn and Fe measurements in skin are potentially added benefits. Clinically, there are many (perhaps hundreds??) tests performed on blood samples. Our effort was to eliminate one additional test.

XRF and TXRF measurements have already been performed on ex vivo blood samples as we referenced in our Introduction section. Our research aims at an in vivo method.

Our dose value was overestimated, as it assumed all photons were absorbed. The systematic effects due to a ‘concentrated’ dose the reviewer is alluding to are not a concern, but we agree that a better microdosimetry evaluation is required before clinical applications. M.W. Charles highlights that systematic effects (transient erythema, moist desquamation, and dermal necrosis) thresholds are ~ few grays to tens of grays [1]. Hence, reviewer’s concerns regarding dose to skin are not entirely justified. In our opinion, calibration method, skin elemental composition, and instrumental development are the more important challenges to clinical applications than the dose concerns.

[1]        M.W. Charles, The skin in radiological protection--recent advances and residual unresolved issues, Radiat. Prot. Dosimetry 109 (2004) 323–330. https://doi.org/10.1093/rpd/nch309.

Comment 12: Lines 633-635. “… and the equivalent dose to skin was below the annual limit for planned radiation exposures of the general public.” This is not applicable, as noted previously. Once again, when accounting for organ weighting factors, you are calculating the effective dose, not the equivalent dose.

Response 12: We agree with this comment. We replaced the original statement with the simpler statement “The equivalent dose to skin was estimated to be below 48 mGy for a 3-minute exposure.”

Comment 13: Lines 636-649: “Fe was the only element with detection limits significantly lower than the median Fe human blood concentration of ~480 mg/L indicating the possibility of clinical applications.”  This does not indicate the possibility of clinical application, particularly due to the issue of signal contamination from skin-deposited Fe. It suggests the potential for further development of instrumentation for medical applications rather than immediate clinical applicability.

Response 13: Yes, we agree, there is more work to be done before clinical applications. We revised this sentence as follows: “Fe was the only element with detection limits significantly lower than the median Fe human blood concentration of ~480 mg/L indicating the potential for further research towards medical applications”.

Comment 14: Lines 643-645: “Therefore, in vivo measurements of their concentrations in cutaneous blood will depend on the success of instrumental modifications tailored to mitigate external XRF signal contamination and x-ray scatter.” This is not entirely accurate. It is not just a matter of instrumental modification. At least for Zn, it also depends on the ability to separate signals from the blood and the skin. For copper, the experimental results do not reliably support the feasibility of such an application. One can only hope that ongoing developments and modifications in …

Response 14: Yes, we strongly agree with this comment. That is the reason why we included references related to skin measurements in our Introduction section. We rephrased the last paragraph in the Conclusions sections as follows:

“Cu, Zn, and Se detection limits were higher than their reported average human blood concentrations. In vivo measurements of their concentrations in cutaneous blood will be inherently linked to their skin concentrations. Instrumental modifications to mitigate external XRF signal contamination and X-ray scatter, as well as finding a suitable calibration method, are needed for measuring the concentrations of these elements in the skin and cutaneous blood.”

Reviewer 4 Report

Comments and Suggestions for Authors

The work is devoted to the creation and evaluation of the parameters of a compact X-ray fluorescence spectrometer for possible use as an analyzer of the elemental composition of blood, in particular, the determination of the content of individual atoms (iron, copper, zinc and selenium) by analyzing vessels in the surface layers of the skin. The authors developed a working model of the spectrometer and conducted studies of model systems, evaluating the spectral resolution, sensitivity to the elements in question and the resulting radiation dose. All these parameters were calculated with maximum accuracy and possible precision, which indicates a high experimental level of the authors.

Overall, the article leaves a good impression and may be of interest to specialists in the field of X-ray physics or metrology.

However, certain comments must be made.

The work under consideration states that "Clinical applications of ICP-MS, however, have certain disadvantages. In addition to the equipment acquisition cost (~$200,000), ICP-MS instruments require a supply of high purity (>99.999%) argon and/or helium gases for plasma production and adequate certified reference materials for accurate quantitative results" (lines 77-80). But at the same time, I would like the authors to compare the parameters of standard industrial devices with those they created. Not only with stationary fluorescent devices, but also with compact analyzers such as Thermo NITON XL5 Plus, usually used for express measurements. For example, in the form of a table. Including there the estimated cost of the device they created. Ideally, it would be possible to conduct a study of the model systems they created on standard spectrometers. This would significantly enhance the impression of the written work.

Technical note: Line 479: "As seen in the figure 9 plot", but figure 9 is not shown in the article.

Author Response

Comment 1:The work under consideration states that "Clinical applications of ICP-MS, however, have certain disadvantages. In addition to the equipment acquisition cost (~$200,000), ICP-MS instruments require a supply of high purity (>99.999%) argon and/or helium gases for plasma production and adequate certified reference materials for accurate quantitative results" (lines 77-80). But at the same time, I would like the authors to compare the parameters of standard industrial devices with those they created. Not only with stationary fluorescent devices, but also with compact analyzers such as Thermo NITON XL5 Plus, usually used for express measurements. For example, in the form of a table. Including there the estimated cost of the device they created. Ideally, it would be possible to conduct a study of the model systems they created on standard spectrometers. This would significantly enhance the impression of the written work.

Response 1: Although a price comparison would be good for readers not familiar with mass spectrometry or XRF analysis, it is a bit beyond the declared scope of this paper. In the revised version of our paper, we introduced a new recent reference that includes such analysis. (Adesina, K.E.; Burgos, C.J.; Grier, T.R.; Sayam, A.S.M.; Specht, A.J. Ways to Measure Metals: From ICP-MS to XRF. Curr. Environ. Health Rep. 2025, 12, 7, doi:10.1007/s40572-025-00473-y.)

We would like to thank the reviewer for his/her comments, suggestions, and corrections.

Our comments are below. 

Comment 2:  Line 479: "As seen in the figure 9 plot", but figure 9 is not shown in the article.

Response 2: Response 7: Yes, a mistake has occurred here. In an earlier manuscript version, we had a figure 9 plot of the DL values versus their average blood concentrations values. We modified the sentence to point to values in the renumbered table 8: “An inspection of table 8 values indicates that, measured Cu DLs were more than an order of magnitude higher than the median human blood concentration.”

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

I would like to thank the authors for addressing all of my comments from the first review. The only remaining point of uncertainty is the sudden drop in signal observed in the experiment with solution no. 5. However, since no better explanation has been found, I suggest retaining the current interpretation. Overall, I believe the manuscript is now suitable for acceptance in its present form. I wish the authors continued success in their future research.

Reviewer 4 Report

Comments and Suggestions for Authors

All my comments and questions were adequately answered.The corrected article can be published.