Deteksi Denyut Jantung Embrio Ikan Oryzias Celebencis Menggunakan Dekomposisi Kuantil Abu-Abu

Inna Ekawati; Annisa Firasanti

doi:10.15575/telka.v9n1.62-73

Deteksi Denyut Jantung Embrio Ikan Oryzias Celebencis Menggunakan Dekomposisi Kuantil Abu-Abu

Inna Ekawati, Annisa Firasanti

Sari

Studi ini mengembangkan pendekatan untuk mengekstrak ciri warna sebagai referensi utama untuk mendeteksi area objek yang bergerak dalam video. Video dalam proyek ini memuat rekaman denyut jantung ikan Oryzias Celebencis yang masih dalam rupa embrio. Metode yang diusulkan pertama-tama melakukan penapisan frame berwarna dalam ruang RGB memakai filter median dan bilateral. Lalu, frame yang telah ditapis menjalani proses konversi ke ruang warna abu-abu. Pada titik ini, frame keabuan akan diurai memakai pendekatan kuantil, ke segmen-segemen warna pembentuknya. Segemen ini dideteksi keberadaan pergerakan objek didalamnya sekaligus disertai dengan perhitungan denyutannya yang timbul berdasarkan pelebaran/penyusutan region objek saat proses pemompaan darah oleh jantung. Metode yang diajukan mampu mengekstrak objek bergerak dalam video dengan IoU sebesar 85% luasan area objek yang bergerak. Penelitian ini berhasil mendeteksi jantung dengan MSE sebesar 0,5; lebih kecil jika dibandingkan dengan rata-rata intensitas objek bergerak yang deteksi oleh software imageJ yang mendapatkan nilai MSE sebesar 396,625. Metode yang diusulkan mampu mengurangi pembentukan puncak palsu dalam grafik puncak/lembah denyutan jantung.

This study develops an approach to extract colour characteristics as the primary reference for detecting areas of moving objects in video. The video in this project contains recordings of the heartbeat of the Oryzias Celebencis fish, which is still in the form of an embryo. The proposed method first performs filtering of coloured frames in RGB space using median and bilateral filters. Then, the filtered frame converts to a grey colour space. The grey frame will be decomposed using a quantile approach into its constituent colour segments. This segment detects the presence of the object's movement and the calculation of its pulsation that arises based on the pulsation of the object's region during the process of blood pumping by the heart. Our method can extract moving objects in video with an IoU of 85% of the moving object area. We have detected a heart with an MSE of 0.5, more minor than the average intensity of moving objects detected by the ImageJ software, which obtained an MSE value of 396,625. Our proposed method can reduce the formation of false peaks in the heart rate peak/valley graph.

Kata Kunci

Filter; median; bilateral; imageJ; segmentasi; puncak

Teks Lengkap:

PDF

Dilihat:

Sari 316 kali

PDF 135 kali

Referensi

D. K. Sari, I. Andriani, K. Yaqin, and A. M. Satya, “The Use of Endemic Sulawesi Medaka Fish (Oryzias celebensis) as an animal model candidate,” Proc. 20th FAVA Congr. 15th KIVNAS PDHI, pp. 564–565, 2018.

E. De Luca et al., “ZebraBeat : A flexible platform for the analysis of the cardiac rate in zebrafish embryos,” Sci. Rep., vol. 4, pp. 1–13, 2014.

K. A. Kurnia, F. Saputra, M. J. M. Roldan, and A. L. Castillo, “One-Stop Method to Measure Multiple Cardiac Performance Endpoints in Daphnia and Zebrafish by Kymograph,” no. October, 2020.

F. Santoso et al., “An overview of methods for cardiac rhythm detection in zebrafish,” Biomedicines, vol. 8, no. 9, 2020.

L. Echeazarra, M. P. Hortigón-Vinagre, O. Casis, and M. Gallego, “Adult and Developing Zebrafish as Suitable Models for Cardiac Electrophysiology and Pathology in Research and Industry,” Front. Physiol., vol. 11, no. January, pp. 1–13, 2021.

N. Hafner, J. C. Drazen, and V. M. Lubecke, “Fish heart rate monitoring by body-contact Doppler radar,” IEEE Sens. J., vol. 13, no. 1, pp. 408–414, 2013.

Y. Shen et al., “Estimating Heart Rate and Detecting Feeding Events of Fish Using an Implantable Biologger,” pp. 37–48, 2020.

B. P. Sampurna, G. Audira, S. Juniardi, Y. H. Lai, and C. Der Hsiao, “A simple ImageJ-based method to measure cardiac rhythm in zebrafish embryos,” Inventions, vol. 3, no. 2, pp. 1–11, 2018.

X. Li and B. Pu, “Strong robustness heart rate estimation using discrete fourier transform and personality heart rate characteristic,” Chinese Control Conf. CCC, pp. 9030–9033, 2017.

J. Gierten et al., “Automated high-throughput heartbeat quantification in medaka and zebrafish embryos under physiological conditions,” Sci. Rep., vol. 10, no. 1, pp. 1–12, 2020.

E. Puybareau, H. Talbot, and L. Marc, “Automated heart rate estimation in fish embryo,” pp. 379–384, 2015.

R. Nepstad, E. Davies, D. Altin, T. Nordtug, and B. H. Hansen, “Automatic determination of heart rates from microscopy videos of early life stages of fish,” J. Toxicol. Environ. Heal. - Part A Curr. Issues, vol. 80, no. 16–18, pp. 932–940, 2017.

E. Moya-Albor, J. Brieva, H. Ponce, O. Rivas-Scott, and C. Gomez-Pena, “Heart Rate Estimation using Hermite Transform Video Magnification and Deep Learning∗,” Proc. Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. EMBS, vol. 2018-July, pp. 2595–2598, 2018.

C. P. Kang, H. C. Tu, T. F. Fu, J. M. Wu, P. H. Chu, and D. T. H. Chang, “An automatic method to calculate heart rate from zebrafish larval cardiac videos,” BMC Bioinformatics, vol. 19, no. 1, pp. 1–10, 2018.

A. M. Naderi et al., “Deep learning-based framework for cardiac function assessment in embryonic zebrafish from heart beating videos,” Comput. Biol. Med., vol. 135, 2021.

A. Farhan et al., “An opencv-based approach for automated cardiac rhythm measurement in zebrafish from video datasets,” Biomolecules, vol. 11, no. 10, pp. 1–21, 2021.

U. Sara, M. Akter, and M. S. Uddin, “Image Quality Assessment through FSIM, SSIM, MSE and PSNR—A Comparative Study,” J. Comput. Commun., vol. 07, no. 03, pp. 8–18, 2019.

L. Gou, S. Wu, J. Yang, H. Yu, and X. Li, “Gaussian guided IoU: A better metric for balanced learning on object detection,” IET Comput. Vis., vol. 16, no. 6, pp. 556–566, 2022

DOI: https://doi.org/10.15575/telka.v9n1.62-73