Dynamic Microscopic Optical Coherence Tomography as a New Diagnostic Tool for Otitis Media

Round 1

Reviewer 1 Report

Comments for author File: Comments.pdf

Author Response

We thank the editor and referees for their detailed and positive reviews and the opportunity to revise the paper and to improve it. We feel that responding to their concerns has significantly improved the paper. We very much appreciate your interest and hope this will find your approval.

A point-by-point response is provided below.

Reviewer 1

 

 

1. In the M&M section authors referenced that the dmOCT benchtop setup (Fig. 1) was only slightly more detailed by in [21,22]. Unfortunately, the dmOCT description reported in [22] is only slightly more detailed by naming the specific products used in the installation. At the same time, the speed of the described systems in the proposed article and 22 differs by 2 times, the width of the optical spectrum - by 10%, and the captions in figure 1 in the work proposed for review are more informative. I believe it would be rational to suggest that the authors expand the description of the dmOCT benchtop setup somewhat within the M&M section of this article, especially since the journal is more technical than diagnostic. And use the reference to work [22], if necessary, in a different context.

 

• We thank the reviewer for pointing out. We expand the section for the hardware description of the system as well as the description of the data processing. We also added new references.

(Redline, Line 105 – 153)

 

2. The same the ref.[19] co The dmOCT volumes were acquired and processed in my opinion, the description of this procedure in [16] is more detailed and relevant from the point of view of the indication as a source in the M&M section.

 

• Please see 1.

 

3. If possible, please, clarify the lateral resolution of the dmOCT image. If the lateral resolution in a focal plane is about 1.5 um, the Rayleigh length should be about 2-3 um for different wavelengths, so the image resolution at 500 um far from the focal plane should be about 19-25 um. The presence of very sharp focusing can also be judged by the inhomogeneity of brightness in Fig.3.c and 4.c. It is also mentioned in passing at the end of the second paragraph on page 5.

 

• For the central wavelength of 750 nm, the depth of field, defined of the Rayleigh length is around 13.5. Therefore indeed, the resolution decreases dramatically, as we have already mentioned in the manuscript. However, due to the high dynamic range, OCT differs significantly from other imaging techniques. Even signals far outside the Rayleigh length contribute meaningfully to the image. We have not yet measured the out-of-focus resolution, but in Figure 5f it is clear that we are able to visualize cellular structures over a range of 100 - 200 ¬µm.

 

 

4. Is the time of 11.52 minutes sufficient (taking into account the time for sample preparation) to consider that the intracellular processes in the removed tissues could be considered intact?

 

• Even though we do not have a quantitative marker for the state of the tissue, we think that we have enough evidence to assume that over a measurement period of 12 minutes no significant change occurs. First, there are no major differences between the different B-scans in a volume. On the other hand, we have shown in previous work (Kohlfaerber et. al.) longitudinal measurements over more than 30 min, which also demonstrate no signal change over time.

We have added this to discuss this further in the discussion. (Redline, Line 349-353)

 

 

5. It is unclear what the dmOCT image demonstrates? On the one hand, this may be the result of Brownian motion. In this case, the result will be determined by the presence of a liquid, its temperature, the size of suspended particles and the effective OCT resolution. On the other hand, it may be a manifestation of the vital activity of cells (electrochemical activity). What is the opinion of the authors on this issue?

 

• There is high evidence that the observed movement is due more to intracellular movement based on cell metabolism than to Brownian movement. We and our collaborators have found little correlation between temperature and movement in experiments. However, we expected this to be the case for Brownian motion, since more kinetic energy would be present in the system. In addition, we are working to identify the actual origin of the motion. However, this is not trivial, because on the one hand it is difficult to observe with other methods and on the other hand the blocking of metabolic processes does not always lead to a linear process in the movement.

 

 

6. Please clarify which tissue is presented as “chronic otitis media” on figure 2 and in its description in “results” section. The “chronic otitis media” term corresponds to diseases, not for any tissue in the middle ear region. If it is a piece of “normal” tissue neighboring to cholesteatoma, it should be pointed in both the text and figure caption.

 

• We apologize if this may be confusing. The tissue, which is presented as “chronic otitis media” in figure 2 a and c is specific taken out during surgery from a patient suffering from COM; clinically and intraoperatively confirmed, as well as later double checked with the histology report of our department of pathology. It is not taken out next to a cholesteatoma. All tissue samples are taken out from different patients representing different tissue types.

 

We have added this to the results. (Redline, Line 186 -192)

 

7. Judging by the nature of the image in figure 2 for cholesteatoma and the description in the "results" section, the signal source for dmOCT is the extracellular matrix, in this case it seems not entirely appropriate to use the term "intracellular movements" later in the same paragraph. Or the statement about "lack of intracellular movement in the cholesteatoma" should be made in the same paragraph, and not when describing the results of visualization of the polyp structure two paragraphs later.

 

• We thank the reviewer for pointing this out, so we can make this appropriate. We changed it accordingly in the results. (Redline, Line 200 -201)

 

.

8. The assumption of possible visualization of immune cells in the epithelium should be supported by more objective data than the reference to previous works ([19]), especially since the authors state that after the dmOCT study, the samples were given for detailed histological studies.

 

• We support this with our H&E staining in Figure 5g, where we can see single cells with a nucleus, such as neutrophils, lymphocytes, plasma cells and macrophages subepithelial in the perimatrix, which was confirmed and evaluated by 2 independent specialists from the Institute of Anatomy and the ENT Department. Moreover, from every patient suffering from a middle ear pathology, samples were sent for histopathological evaluation to the Department of Pathology according to our clinical standard procedure and to our protocol (22-041), which was approved by the ethic committee of the University of Lübeck (see in the methods). For this sample, we got the record, that in this tissue there are stacks of keratin and epithelial cells decomposed with immune and inflammatory cells corresponding to the picture of a medium to high-grade inflammatory cholesteatoma.

Furthermore, we added several references, showing an interaction of the immune system with cholesteatoma using histopathology and immunohistochemistry. Unfortunately, to date, there are no markers for OCT, which are able to specific label immune cells.

We added this to the discussion. (Redline, Line 321 – 333)

 

 

 

Author Response File: Author Response.docx

Reviewer 2 Report

The results of this work are of high interest. The described application of dmOCT for otitis investigation has wide potential in clinical and academic work.

It is clear that the design of this manuscript is a general demonstration of the dmOCT abilities. However, the authors stated that the samples from 12 patients were studied. But the paper covers only a few OCT images. It is essential, even using this small set of tissue samples, to discuss the statistics of  detecting otitis and cholesteatoma markers, comparing the OCT images with histological ones. Several quantitative metrics should be discussed. 

The second issue concerns the estimation of the metabolism in ex vivo samples. Maybe, the authors would discuss the impact of time after excision on the observed results, as well as the applicability of this marker for further in vivo investigations.

The final minor issue concerns the adjective 'non-contact' used for OCT. It is better to use 'non-invasive', since several OCT techniques, such as endoscopic, use contact OCT probes.     

Author Response

We thank the editor and referees for their detailed and positive reviews and the opportunity to revise the paper and to improve it. We feel that responding to their concerns has significantly improved the paper. We very much appreciate your interest and hope this will find your approval.

A point-by-point response is provided below.

 

Reviewer 2

The results of this work are of high interest. The described application of dmOCT for otitis investigation has wide potential in clinical and academic work.

1. It is clear that the design of this manuscript is a general demonstration of the dmOCT abilities. However, the authors stated that the samples from 12 patients were studied. But the paper covers only a few OCT images. It is essential, even using this small set of tissue samples, to discuss the statistics of  detecting otitis and cholesteatoma markers, comparing the OCT images with histological ones. Several quantitative metrics should be discussed. 

 

• We thank the reviewer for this point. This study is indeed intended to be a general demonstration of the dmOCT abilities. We performed our experiments according to our previous work by Kohlfaerber et al. 2022 and Münter et al., 2021. Our samples were confirmed with H&E stainings of the same location, analyzed and evaluated by 2 independent specialists from the Institute of Anatomy and the ENT Department.

Since this is a new diagnostic method or visualization technique, that has not yet been performed in patients with middle ear pathologies, we could not use an effect size from the literature to calculate a case number analysis. With the expected variance of cholesteatoma and chronic otitis media mesotympanalis tissues, 12-15 samples have been appointed to provide a good overview of how the different tissue structures are represented in dOCT in cooperation with the Institute for Biometry and Statistics. Concerning specific otitis media and cholesteatoma markers, further investigations are needed, with a much higher number of tissue samples; also for identifying and labeling specific immune cells and quantitative analysis needed to be performed and is under ongoing investigation.

2. The second issue concerns the estimation of the metabolism in ex vivo samples. Maybe, the authors would discuss the impact of time after excision on the observed results, as well as the applicability of this marker for further in vivo investigations.

• For us, the dynamic OCT technique is primarily used to increase the contrast in comparison to classical OCT. Although we are working on it, we currently do not assert any direct correlation between the observed colors and metabolic states or cellular connections. Regarding the comparison of sampling time and results, we have not conducted any specific studies. However, in one of our previous investigations, we demonstrated that over an extended duration of 37 minutes, there were no significant changes in color.

We have added this to discuss this further in the discussion. (Redline 349-353)

3. The final minor issue concerns the adjective 'non-contact' used for OCT. It is better to use 'non-invasive', since several OCT techniques, such as endoscopic, use contact OCT probes.   

• We thank the reviewer for this point. However, we believe that the term "non-contact" is justified in this particular context and frequently used within the OCT community. While OCT is occasionally used in a contact mode, it fundamentally differs from imaging methods like ultrasound or the practical implementation of confocal microscopy, where a contact medium is required. Considering the possibility of using endoscope, we do not believe that "non-invasive" is an appropriate term.

Reviewer 3 Report

This manuscript reports on an application of dynamic and micro-OCT (dmOCT) to imaging of tissue specimens of chronic otitis media and cholesteatoma. The dmOCT provides higher spatial resolution and sensitivity to microscale fluctuations compared with conventional OCT, and the authors exploited such features and well-presented convincing results that demonstrate potential viability of dmOCT as a diagnostic tool for such diseases. Overall, the manuscript is clearly written, and the results are impressive. As such, the reviewer would support its publication in Photonics. Below are some comments that hopefully can be addressed.

 -          The depth-resolved OCT signals were divided into three bands as fast fluctuations (4.9 – 25 Hz), medium motion (0.4 - 4.9 Hz) and slow motion (< 0.5Hz), which were then used to generate pseudo-colored dmOCT images. The division of the frequency bands, however, seems rather empirical. Could the authors provide some rationale on how they selected such frequency ranges for the three bands?

 -          While the dmOCT images are impressive, it is not clear how the information can help clinical diagnosis. It is suggested that the authors elaborate on the significance of dmOCT images on the diagnosis of relevant diseases.

Author Response

We thank the editor and referees for their detailed and positive reviews and the opportunity to revise the paper and to improve it. We feel that responding to their concerns has significantly improved the paper. We very much appreciate your interest and hope this will find your approval.

A point-by-point response is provided below.

 

Reviewer 3

This manuscript reports on an application of dynamic and micro-OCT (dmOCT) to imaging of tissue specimens of chronic otitis media and cholesteatoma. The dmOCT provides higher spatial resolution and sensitivity to microscale fluctuations compared with conventional OCT, and the authors exploited such features and well-presented convincing results that demonstrate potential viability of dmOCT as a diagnostic tool for such diseases. Overall, the manuscript is clearly written, and the results are impressive. As such, the reviewer would support its publication in Photonics. Below are some comments that hopefully can be addressed.

1. The depth-resolved OCT signals were divided into three bands as fast fluctuations (4.9 – 25 Hz), medium motion (0.4 - 4.9 Hz) and slow motion (< 0.5Hz), which were then used to generate pseudo-colored dmOCT images. The division of the frequency bands, however, seems rather empirical. Could the authors provide some rationale on how they selected such frequency ranges for the three bands?

 

 

• The assignment of the frequency bands to the color channels based on an early work using FF-OCT and matching well-known frequencies of intracellular motion. Changing the borders slightly doesn’t lead to significant different.

 

We included that into the methods. (Redline, Line 151 – 153)

 

2. While the dmOCT images are impressive, it is not clear how the information can help clinical diagnosis. It is suggested that the authors elaborate on the significance of dmOCT images on the diagnosis of relevant diseases.

 

• The aim is to enable the individualized determination of the inflammatory process in the middle ear during surgery, which allows to continuously assess the current inflammatory boundaries and to distinguish between inflammatory focus, cholesteatoma and normal mucosa, which is the key element for complete safe remediation in tortuous anatomical conditions of the middle ear. To further develop minimally invasive ear surgery.

By further implementing dynamic OCT in a fiber-scanning endoscope, an optical biopsy might be realized in the operating room, allowing the status of the pathological situation to be recorded.

 

Intraoperative individual evaluation of the inflammatory component followed by targeted selected resection will potentially also reduce recovery time and recurrence rates. For patients, this is an incredible gain in quality of life and socioeconomic value.

We revised the whole discussion section

Round 2

Reviewer 1 Report

The authors took into account my comments and answered my questions.