CC BY
13
MEASURING THE TYMPANIC MEMBRANE THICKNESS USING OPTICAL COHERENCE TOMOGRAPHY
A.A. Novozhilov1'2, P.A. Shilyagin2*, V.A. Novozhilova2, D.D. Smirnova1'2, A.A. Slepanov1, A.L. Dilenjan1'2, M.B. Klimycheva1, G.G. Gelikonov2, V.M. Gelikonov2, A.V. Shakhov1
1 Privolzhsky Regional Medical Center, 20a M. Voronova St., Nizhny Novgorod' 603950' Russia;
2 Institute of Applied Physics, Russian Academy of Sciences, 46 Ulyanova St., Nizhny Novgorod, 603950, Russia;
3 Privolzhsky Research Medical University, University Clinic, 190 Rodionova St., Nizhny Novgorod, 603950, Russia.
* Corresponding author: paulo-s@mail.ru
Abstract. Introduction: the average thickness of the tympanic membrane (TM) normally fluctuates around 100 ^m. Most of the measurements of the TM thickness presented in previous works were performed on cadaveric material. The anatomical parameters of the TM can be measured using optical coherence tomography (OCT). Materials and methods: we used spectral-domain OCT with a non-contact probe to study the structures of the middle ear. The results of OCT examination of 58 tympanic membranes of patients without middle ear pathology were analyzed. Quantitative analysis of OCT images was performed using the ImageJ program. The thickness of the tympanic membrane was calculated manually at 5 randomly selected points. To convert optical values into metric ones, we used an average refractive index of 1.45. Results: the TM thickness in patients without ear pathology was 138 ^m (±29). Conclusions: OCT is an effective method of non-invasive examination of the TM and can be successfully used to measure the TM thickness. Improving the algorithms for processing OCT images in order to automatically calculate the TM thickness and make TM topographic maps remains a crucial task and its solution is expected to significantly improve the diagnostic properties of the method.
Keywords: optical coherence tomography; tympanic membrane thickness.
List of Abbreviations
OCT - optical coherence tomography OME - otitis media with effusion EAC - external auditory canal TM - tympanic membrane CT - computer tomography MRI - magnetic resonance imaging
Introduction
Studying normal anatomical and topographic parameters of the human body structures and organs is one of the most important tasks of modern medicine. Determining the reference range is necessary to differentiate pathological conditions.
According to the results obtained by D.J. Lim, the thickness of the pars flaccida (Fig. 1) can range from 30 |m to 230 |m, while the pars tensa thickness varies from 30 |m to 90 |m (Lim, 1970). According to Ruah at al., the TM thickness depends on age and ranges from 80 |m in the elderly to 2,400 |m in children in the pars flaccida (Ruah et al., 1991), while fluctuations in the TM thickness in the pars tensa are less pronounced, from 30 |m in the anterior and
posterior quadrants in adults to 1,500 |im in children (Kuypers et al., 2006).
However, despite the high diagnostic potential, normal and pathological values of the human tympanic membrane (TM) thickness have been studied rather poorly. Most of the measurements of the TM thickness presented in studies were performed on cadaveric material. Due to irreversible cadaveric changes (for the elements of the inner ear described in (Cho et al., 2021), for the cornea of the eye described in (Napoli et al., 2016)), these data cannot be used as a reference when assessing the TM thickness in a living organism. The same problem refers to the measurements made on surgical material. To date, the most common method for studying biological tissues is histological examination. However, this method is invasive which means that there is a significant time interval between sample collection and analysis, in which case the measured values may also vary and differ from those in vivo (De Greef et al., 2016; Ruah et al., 1991).
The in vivo examination of the structures of the middle ear is performed using tomographic
Fig. 1. Otoscopic image of a normal tympanic membrane. Pars tensa: 1 - Anterior superior quadrant, 2 - Posterior superior quadrant, 3 - Anterior inferior quadrant, 4 - Posterior inferior quadrant, 5 - Pars flaccida
diagnostic methods (CT and MRI), which are indispensable in otorhinolaryngology, but require expensive equipment, highly qualified personnel, specially equipped rooms and in CT case expose patients to radiation (Boboshko et al., 2003; Pont & Mazon, 2017; Swartz & Harnsberger, 1990). In addition, although the resolution of CT and MRI reaches the submillimeter level (up to a maximum of 125 |m), this is not enough to effectively measure the TM thickness (Kerkfeld & Meyer, 2018; Linden, 2012).
Optical coherence tomography (OCT) is an in vivo non-invasive method for examining biological tissues (Boppart, 2003). A distinctive feature of OCT is a high resolution, from 5 to 15 |im, and a small, but sufficient for examining the structures of the middle ear, probing depth - from 1 to 3 mm (Adhi & Duker, 2013; Monroy et al., 2017; Novozhilov et al., 2020; Rubinstein et al., 2009). The OCT method is based on probing tissues with light in the near-infrared and analyzing back-reflected waves to measure the depth in which the reflection occurred. In general, OCT is not a measuring tool,
since the physical principles the technology is based on allow to measure with high accuracy only the magnitude of the wave delay, which is related to the real geometric dimensions of the medium through a refractive index that varies depending on the biological tissues. An additional hindering factor to effective use of OCT as a measuring tool is soft tissue deformation when using contact probes (Agrba & Kirillin, 2016). However, this problem has been solved by non-contact probing (Novozhilov et al., 2020a; Shilyagin et al., 2018). The geometric thickness can be calculated in this case if a refractive index of a particular biological tissue is known. The value of the refractive index in biological tissues generally depends on the presence of water and for most soft tissues ranges from 1.335 to 1.560 (Khan et al., 2021), while individual structures can have a significantly narrower range of refractive index values. For example, the refractive index of collagen fibers varies from 1.44 to 1.47 (Khan et al., 2021; Van der Jeught et al., 2013).
The first measurements of the TM thickness were performed on cadaveric material, but the introduction of OCT probes combined with otoscopes and otoendoscopes made it possible to obtain both 2D images (Meller et al., 2014; Shilyagin et al., 2018) as well as in vivo 3D images (MacDougall et al., 2016; Park et al., 2017; Van der Jeught et al., 2013). This allowed in 2015 to present the measurements of the TM thickness in acute and chronic otitis media, localized changes, tympanosclerosis (Guder et al., 2015; Hubler et al., 2015), although the total numbers of patients were considerably small.
In addition, small groups of patients were examined in the studies (Hubler et al., 2015; Monroy et al., 2015) and (Pande et al., 2016) published in 2015 and 2016, respectively. An important aspect of these works is the determination of the average refractive index of the TM. In (Pande et al., 2016) the index was 1.45, while in (Hubler et al., 2015) the index was 1.44, which coincided with the value determined for the TM ex vivo in (Van der Jeught et al., 2013). The listed works, however, are focused, to a greater extent, on demonstrating the possibilities of OCT rather than on identifying statistically substantiated regularities.
The purpose of this study is to measure the in vivo thickness of the normal TM.
Materials and Methods
The research was approved by the local ethical committee of Privolzhsky Research Medical University (Protocol No.7 July 03, 2017). The research was made in the ENT department of Clinic No.3, Privolzhsky Regional Medical Center (Nizhny Novgorod, Russia).
We examined 58 patients without middle ear pathology. The examined group of patients included 24 men and 34 women aged 18 to 76 years. Examination of patients was performed in the otorhinolaryngological office in the outpatient clinic. All patients were examined for complaints, anamnesis and general somatic status. All patients underwent a routine otorhino-laryngological examination as well as otomi-croscopy, tone threshold audiometry and tympanometry. All the subjects signed written informed consent for the study.
After examination and diagnosis, all patients underwent an OCT examination of the middle ear with the use of a diagnostic microscope without anesthesia in the sitting position.
For the study, data were selected from the patients who had not been diagnosed with an inflammatory process, OME, etc., and who had no history of chronic otitis media in any form.
We used an approved for clinical use compact spectral optical coherence tomograph with a non-contact probe developed at the Institute of Applied Physics of the Russian Academy of Sciences (Nizhny Novgorod) (Shilyagin et al., 2021).
The OCT image of the TM was recorded in real time on a computer screen. The TM thickness was measured using OCT in the pars tensa, at the same distance from the annulus fibrosus and where the malleus is attached to the TM, in the anterior-inferior and posterior-inferior quadrants, in the line-of-sight zone. Quantitative analysis of the acquired OCT images was performed using the ImageJ open-source software. To do this, the OCT image in the selected section was converted into a separate file in a graphic format with a linear brightness coding.
The thickness of the tympanic membrane was calculated manually, at five randomly selected points, at equal distances from each other (Fig. 2).
With this method of measurement, the values of the TM thickness are measured in arbitrary units - pixels. The conversion of the measured values of the TM thickness d from arbitrary units to metric units was carried out using the following formula: i
D=--S-d,
n
where n is the refractive index of the medium, 5 is the metric value of the image resolution element, determined by the following ratio:
Ô =
Dmax
where Dmax and dmax are values of the maximum imaging depth of the OCT system in metric and arbitrary units, respectively. For the OCT system used in the study, these values were Dmax = 3,200 |m, dmax = 256. The value of the refractive index (n) was 1.45.
d
The measurements were recorded and analyzed using the MS Excel program. Statistical analysis was carried out using the Statgraphics Centurion, v.9 software package. The data are presented as M ± m, where M is the mean value, m is the standard deviation.
The interpolation of experimental data by Gaussian curve was made using Mathcad's built-in fitting function.
Results
In the examined group of patients, in all cases, normal otoscopy status were observed, all pa-
tients had normal audiometric parameters: tympanometry type A, no air-bone interval. A typical OCT image of the TM is shown in Fig. 3.
The absolute values of the geometric thickness of the TM in the examined patients calculated according to the described method were 138 (±35) |m. The biggest obtained value was registered at level 223 um, the smallest -90 um. The median registered value was 133 um (±13um), which is slightly different from mean value.
Fig. 2. Method for measuring the thickness of the tympanic membrane. A - external auditory canal, B - tympanic cavity, C - TM. The lines indicate the places where the thickness was measured
Fig. 3. OCT image of the normal TM. A - external auditory canal, B - tympanic cavity, C - TM
Discussion
The obtained measurements of the thickness of the normal TM generally correspond to the results obtained by other researchers. At the same time, it should be noted that, in general, the TM thickness values obtained by OCT (in vivo) are higher than post-mortem ones both in this investigation and other OCT-based measurements.
The distribution of thickness in registered data was calculated with a step of 10 um (Fig. 5). It looks close to normal (Gaussian) distribution with center at 125 um and FWHM 24 um.
These once again indicate the need for large-scale intravital studies of the value of the TM thickness.
It seems necessary to pay attention also to the range of thickness values recorded in each individual case. In most cases (83%) the standard deviation calculated for 5 measurements does not exceed 20%, but the highest deviation was registered to be 44%. This is caused by the inhomogeneity of TM profile and indicates the need for clarify the area of interest of these measurements. The best way to take the TM thickness inhomogeneity would be the use of area mapping techniques similar to ones used in retinal and corneal OCT imaging (Napoli et al., 2016) and reported for TM in (Kuypers et al., 2006) and (Pande et al., 2016; Van der Jeught et al., 2013) for cadaveric material.
TM thickness, um
200
150
100
50
Iii
I
iiii
our results Lim, 1970 Ruah et al., Kuypers et Van der Monroy et Guder et al., Pande et al., (58) (NA) 1991 al., 2006 Jeught et al., al., 2015 2015 2016
(22) (6) 2013 (13) (10) (4)
(1)
Fig. 4. TM thickness measured in our study and by other researchers. Parentheses indicate the size of the sample of patients for whom the given data were obtained
12 10
Thickness histogram
♦ ♦ ♦
■ - -♦- ----♦-
80 100 120 140 160 180 200 220 240
Thickness, um
♦ measured data
Gaussian interpolation
Fig. 5. Measured TM thickness distribution
0
8
6
4
2
0
Conclusions
OCT is an effective method of non-invasive examination of the TM and can be successfully used in the diagnosis and study of the nature of structural changes resulting from pathological conditions.
Improving the algorithms for processing OCT images in order to automatically calculate the TM thickness and make TM topographic maps remains a crucial task and its
solution is expected to significantly improve the diagnostic properties of the method.
Acknowledgements
This study was partly supported by the state task from the Institute of Applied Physics, Russian Academy of Sciences (Project No. 0030-2021-0013) in terms of material support for the research.
References
ADHI M. & DUKER J.S. (2013): Optical coherence tomography--current and future applications. Curr Opin Ophthalmol 24(3), 213-221.
AGRBA P.D. & KIRILLIN M.Y. (2016): Effect of temperature regime and compression in OCT imaging of skin in vivo. Photonics & Lasers in Medicine 5(2), 161-168.
BOBOSHKO M., BRYZGALOVA S.V. & LOPOTKO A.I. (2003): The condition of the auditory tube in patients with chronic purulent otitis media as shown by computed tomography. Vestn Otorinolaringol 4, 16-19.
BOPPART S.A. (2003): Optical coherence tomography: technology and applications for neuroimaging. Psy-chophysiology 40(4), 529-541.
DE GREEF D., GOYENS J., PINTELON I., BOGERS J.P., VAN ROMPAEY V., HAMANS E., VAN DE HEYNING P. & DIRCKX J.J.J. (2016): On the connection between the tympanic membrane and the malleus. Hear Res 340, 50-59.
GUDER E., LANKENAU E., FLEISCHHAUER F., SCHULZ-HILDEBRANDT H., HUTTMANN G., PAU H.W. & JUST T. (2015): Microanatomy of the tympanic membrane in chronic myringitis obtained with optical coherence tomography. Eur Arch Otorhinolaryngol 272(11), 3217-3223.
HUBLER Z., SHEMONSKI N.D., SHELTON R.L., MONROY G.L., NOLAN R.M. & BOPPART S.A. (2015): Real-time automated thickness measurement of the in vivo human tympanic membrane using optical coherence tomography. Quantitative imaging in medicine and surgery 5(1), 69-77.
KERKFELD V. & MEYER U. (2018): Higher Resolution in Cone Beam Computed Tomography Is Accompanied by Improved Bone Detection in Peri-implant Bone Despite Metal Artifact Presence. Int J Oral MaxillofacImplants 33(6), 1331-1338.
KHAN R., GUL B., KHAN S., NISAR H. & AHMAD I. (2021): Refractive index of biological tissues: Review, measurement techniques, and applications. Photodiagnosis and Photodynamic Therapy 33, 102192.
KUYPERS L.C., DECRAEMER W.F. & DIRCKX J.J. (2006): Thickness distribution of fresh and preserved human eardrums measured with confocal microscopy. Otol Neurotol 27(2), 256-264.
LIM D.J. (1970): Human tympanic membrane. An ultrastructural observation. Acta Otolaryngol 70(3), 176186.
LINDEN D.E.J. (2012): The Challenges and Promise of Neuroimaging in Psychiatry. Neuron 73(1), 8-22.
MACDOUGALL D., FARRELL J., BROWN J., BANCE M. & ADAMSON R. (2016): Long-range, wide-field swept-source optical coherence tomography with GPU accelerated digital lock-in Doppler vibrog-raphy for real-time, in vivo middle ear diagnostics. Biomed Opt Express 7(11), 4621-4635.
MELLER A., SHAKHOVA M., RILKIN Y., NOVOZHILOV A., KIRILLIN M. & SHAKHOV A. (2014): Optical coherence tomography in diagnosing inflammatory diseases of ENT. Photonics & Lasers in Medicine 3(4), 323-330.
MONROY G.L., PANDE P., SHELTON R.L., NOLAN R.M., SPILLMAN JR. D R., PORTER R.G., NOVAK M.A. & BOPPART S.A. (2017): Non-invasive optical assessment of viscosity of middle ear effusions in otitis media. J. Biophotonics 10(3), 394-403.
MONROY G.L., SHELTON R.L., NOLAN R.M., NGUYEN C.T., NOVAK M.A., HILL M.C., MCCORMICK D.T. & BOPPART S.A. (2015): Noninvasive depth-resolved optical measurements of the tympanic membrane and middle ear for differentiating otitis media. The Laryngoscope 125(8), E276-E282.
NAPOLI P.E., NIOI M., D'ALOJA E. & FOSSARELLO M. (2016): Post-Mortem Corneal Thickness Measurements with a Portable Optical Coherence Tomography System: a Reliability Study. Scientific Reports 6(1), 30428.
NOVOZHILOV A.A., SHILYAGIN P.A., SHAKHOV A.V. & GELIKONOV V.M. (2020): Overview of modern methods for the diagnosis of exudative otitis media. Vestnik Oto-Rino-Laringologii 85(3), 6874.
PANDE P., SHELTON R.L., MONROY G.L., NOLAN R.M. & BOPPART S.A. (2016): A Mosaicking Approach for In Vivo Thickness Mapping of the Human Tympanic Membrane Using Low Coherence Interferometry. J Assoc Res Otolaryngol 17(5), 403-416.
PARK K., CHO N.H., JANG J.H., LEE S.H., KIM P., JEON M., BOPPART S.A., KIM J. & JUNG W. (2017): In vivo 3D imaging of the human tympanic membrane using a wide-field diagonal-scanning optical coherence tomography probe. Appl Opt 56(9), D115-D119.
PONT E. & MAZON M. (2017): Indications and radiological findings of acute otitis media and its complications. Acta Otorrinolaringol Esp 68(1), 29-37.
RUAH C.B., SCHACHERN P.A., ZELTERMAN D., PAPARELLA M M. & YOON T. H. (1991): Age-related morphologic changes in the human tympanic membrane. A light and electron microscopic study. Arch Otolaryngol Head Neck Surg 117(6), 627-634.
RUBINSTEIN M., SCHALCH P., DI SILVIO M., BETANCOURT M.A. & WONG B.J. (2009): Optical coherence tomography applications in otolaryngology. Acta Otorrinolaringol Esp 60(5), 357-363.
SHILYAGIN P.A., NOVOZHILOV A.A., ABUBAKIROV T.E., DILENYAN A.L., SHAKHOV A.V., MOISEEV A.A., TERPELOV D.A., KSENOFONTOV S.Y., MATKIVSKY V.A., GELIKONOV V.M. & GELIKONOV G.V. (2021): Optical coherence tomograph for non-invasive examination of the human middle ear. Quantum Electronics 51(1), 38-42.
SHILYAGIN P.A., NOVOZHILOV A.A., ABUBAKIROV T.E., GELIKONOVA V.G., TERPELOV D.A., MATKIVSKY V.A., GELIKONOV G.V., SHAKHOV A.V. & GELIKONOV V.M. (2018): Time domain optical coherence tomography is a useful tool for diagnosing otitis media with effusion. Laser Physics Letters 15(9), 096201-096205.
SWARTZ J.D. & HARNSBERGER H.R. (1990): The temporal bone: magnetic resonance imaging. Top Magn Reson Imaging 2(4), 1 -16.
VAN DER JEUGHT S., DIRCKX J.J., AERTS J R., BRADU A., PODOLEANU A G. & BUYTAERT J.A. (2013): Full-field thickness distribution of human tympanic membrane obtained with optical coherence tomography. J Assoc Res Otolaryngol 14(4), 483-494.
