Tayfun Akgül, Tuba Yilmaz, Seda Keskin

Mikrodalga Dielektrik Spektroskopi ile Sert ve Yumuşak Doku Karakterizasyonunun Ön İncelenmesi

Mikrodalga frekanslarında sert ve yumuşak dokular arasındaki dielektrik özellik farkı yumuşak dokularin sert dokulardan ayrılması için kullanılanılabilme potansiyeline sahiptir. Biyolojik dokuların mikrodalga frekanslarda dielektrik özellikleri geleneksel olarak açık uçlu koaksiyel prob tekniği ile ölçülür. Bununla birlikle, doku heterojenitesi, kullanıcı hataları, matematiksel yaklaşım ve kalibrasyon bozulması nedeniyle kullanılan teknik yüksek hata oranlarına sahiptir. Farklı değerlere sahip veri gruplarına makine öğrenimi algoritması uygulandığında verinin yüksek doğrulukta sınıflandırılabileceği bilinmektedir. Bu nedenle, tekniğe özgü hatalardan en az etkilenebilecek bir sınıflandırma parametresinin seçilmesi, doku kategorizasyonunun doğruluğunu artırmak için kritik öneme sahiptir. Emprik olarak, mikrodalga frekanslarındaki dielektrik özellikler güç yasasına uyar. Bu olguya göre daha önce araştırılmamış bir parametre, dielektrik özelliklerden elde edilebilecek güç parametresidir. Bu kapsamda güç parametresinin farklı dokuları, özellikle sert ve yumuşak dokuları, ayırabilme potansiyeli bu çalışmada literatürde verilmiş olan veri gruplarından yola çıkarak araşırılmıştır. Ayrıca güç parametresinin etkinliğini araştırmak amaçlı sağlıklı ve kanserli karaciğer dokularına ait dielektrik özellik ölçümlerine ait güç parametreleri kullanılarak makına öğrenme algoritmalarıyla sınıflandırma yapılmıştır. Uygulanan teknik sonucu %82 doğruluk elde edilmiştir. Bu kapsamda güç parametresinin doku sınıflandırılmasında dielektrik özelliklere ek olarak farklı bilgi barındıran bir özellik olarak kullanılabilmesi öngörülmektedir. Alternatif olarak, bazı durumlarda dielektrik özellikler yeterli bilgi sağlamaz, bir örnek sert ve yumuşak dokuların ayrılmasıdır, bu koşullar altında güç parametresi sınıflandırma amacıyla kullanılabilir. Bu yaklaşım yüksek maliyetli görüntüleme veya mutasyon tarama testlerine alternatif bir hızlı tanı yöntemi olarak kullanılabilir.

Preliminary Investigation of Hard and Soft Tissue Characterization with Microwave Dielectric Spectroscopy

The dielectric property discrepancy between hard and soft tissues at microwave frequencies can potentially be utilized for the seperation of these tissues from one another. Microwave dielectric properties of biological tissues are traditionally measured with the open-ended coaxial probe technique. However, the technıque suffers from high error rates due to tissue heterogeneity, user errors, mathematical approach and calibration degredation. It is known that datasets with dıfferent values can be classified with high accuracy when a machine learnin algorithm is applied. Therefore, choosing a classification parameter that can be least affected by inherent errors is critical for increasing the accuracy of tissue categorization. Emprically, dielectric properties at microwave frequencies abides by the power law. Based on this fact, one unexplored parameter is the power parameter which can be derived from the dielectric properties. To this end, this work presents investigations on the potential use of the power parameter to separate different tissues, spesifıcally hard and soft tissues, based on the datasets available in the literature. Additionally, in order to investigate the effectiveness of the power parameter, classification was performed with machine learning algorithms using the power parameters obtaıned from dielectric property measurements of healthy and malignant liver tissues. Through the application of the technique 82% accuracy was obtined. Towards this goal, it is predicted that the power parameter can be used as a feature containing different information in addition to dielectric properties in tissue classification. Alternatively, in some cases dielectric properties do not provide enough information, one example is the seperation of hard and soft tissues, under such conditions the power parameter can be employed for classification purposes. This approach can possibly be used as an alternative rapid diagnostic method to highcost imaging or mutation screening tests.

PDF

___

[1] S. Gabriel, R. W. Lau, and C. Gabriel, “The dielectric properties of biological tissues: III. Parametric models for the dielectric spectrum of tissues,” Physics in Medicine and Biology, vol. 41, no. 11, pp. 2271–2293, Nov. 1996.
[2] T. Yilmaz, M.A. Kılıç, M. Erdoğan, M. Çayören, D. Tunaoğlu, İ. Kurtoğlu, Y. Yaslan, H. Çayören, A. E. Arıkan, S. Teksöz and G. Cancan, “Machine learning aided diagnosis of hepatic malignancies through in vivo dielectric measurements with microwaves,” Physics in Medicine and Biology, vol. 61, no. 13, pp. 5089–5102, Jun. 2016.
[3] H. Sahin, S. Abdullazade, and M. Sanci, “Mature cystic teratoma of the ovary: a cutting edge overview on imaging features,” Insights into Imaging, vol. 8, no. 2, pp. 227–241, Jan. 2017.
[4] F. Kikkawa et al., “Diagnosis of squamous cell carcinoma arising from mature cystic teratoma of the ovary,” Cancer, vol. 82, no. 11, pp. 2249–2255, Jun. 1998.
[5] B. Cakmak, M. Nacar, N. Aliyev, D. Koseoglu, and Z. Ozsoy, “Mature cystic teratomas: Relationship between histopathological contents and clinical features,” Nigerian Journal of Clinical Practice, vol. 18, no. 2, p. 236, 2015.
[6] A. Hackethal, D. Brueggmann, M. K. Bohlmann, F. E. Franke, H.-R. Tinneberg, and K. Münstedt, “Squamous- cell carcinoma in mature cystic teratoma of the ovary: systematic review and analysis of published data,” The Lancet Oncology, vol. 9, no. 12, pp. 1173–1180, Dec. 2008.
[7] T. Yilmaz, “Multiclass Classification of Hepatic Anomalies with Dielectric Properties: From Phantom Materials to Rat Hepatic Tissues,” Sensors, 2020, vol. 20, pp. 530.
[8] F. S. Kaplan, M. Le Merrer, D.L. Glaser, R. J. Pignolo, R. E. Goldsby, J. A. Kitterman, J. Groppe and E. M. Shore, “Fibrodysplasia ossificans progressiva,” Best Practice & Research Clinical Rheumatology, vol. 22, no. 1, pp. 191–205, Mar. 2008.
[9] M. Lazebnik et al., “A large-scale study of the ultrawideband microwave dielectric properties of normal, benign and malignant breast tissues obtained from cancer surgeries,” Physics in medicine and biology, vol. 52, no. 20, pp. 6093–115, 2007.
[10] C. Gabriel, S. Gabriel, and E. Corthout, “The dielectric properties of biological tissues: I. Literature survey,” Physics in medicine and biology, vol. 41, no. 11, pp. 2231–49, 1996.
[11] A. La Gioia, E. Porter and I. Merunka, “Open-Ended Coaxial Probe Technique for Dielectric Measurement of Biological Tissues: Challenges and Common Practices,” Diagnostics (Basel, Switzerland), vol. 8, no. 2, p. 40, 2018.
[12] T. W. Athey, M. A. Stuchly, and S. S. Stuchly, “Measurement of Radio Frequency Permittivity of Biological Tissues with an Open-Ended Coaxial Line: Part I,” IEEE Transactions on Microwave Theory and Techniques, vol. 30, no. 1, pp. 82–86, Jan. 1982.
[13] M. A. Stuchly, T. W. Athey, G. M. Samaras, and G. E. Taylor, “Measurement of Radio Frequency Permittivity of Biological Tissues with an Open-Ended Coaxial Line: Part II- Experimental Results,” IEEE Transactions on Microwave Theory and Techniques, vol. 30, no. 1, pp. 87–92, Jan. 1982.
[14] W. Anderson, “Microwave biopsy probe- Anderson, Wendell,” Freepatentsonline.com, Apr. 2006.
[15] P. J. Debye, “Polar molecules, “Chemical Catalog Company, Incorporated, 1929.
[16] K. S. Cole and R. H. Cole, “Dispersion and Absorption in Dielectrics I. Alternating Current Characteristics,” The Journal of Chemical Physics, vol. 9, no. 4, pp. 341–351, Apr. 1941.
[17] K. S. Cole and R. H. Cole, “Dispersion and Absorption in Dielectrics II. Direct Current Characteristics,” The Journal of Chemical Physics, vol. 10, no. 2, pp. 98–105, Feb. 1942.
[18] A. K. Jonscher, “Dielectric relaxation in solids,” Journal of Physics D: Applied Physics, vol. 32, no. 14, pp. R57– R70, Jan. 1999, doi: 10.1088/0022-3727/32/14/201.
[19] D. Andreuccetti, “Dielectric Properties of Body Tissues: Home page,” Ifac.cnr.it, 2018. [Online].
[20] S. Gabriel, R. W. Lau, and C. Gabriel, “The dielectric properties of biological tissues: II. Measurements in the frequency range 10 Hz to 20 GHz,” Physics in Medicine and Biology, vol. 41, no. 11, pp. 2251–2269, Nov. 1996.
[21] Yı̇lmaz, T. "Microwave Spectroscopy Based Classification of Rat Hepatic Tissues: On the Significance of Dataset”. Balkan Journal of Electrical and Computer Engineering 8 (2020): 307-313
[22] Yilmaz, T. Multiclass Classification of Hepatic Anomalies with Dielectric Properties: From Phantom Materials to Rat Hepatic Tissues. Sensors 2020, 20, 530.