Panduan Penulisan
Evaluasi Efektivitas Model U-Net untuk Segmentasi Citra Renal Scintigraphy pada Penilaian Fungsi Ginjal
Sari
Renal scintigraphy is a nuclear medicine procedure commonly used to quantitatively assess kidney function and monitor various clinical conditions. Image analysis requires segmentation of the kidney’s region of interest (ROI), which is typically performed manually by experienced operators. This manual approach is time-consuming and prone to inter-observer variability. This study develops and evaluates a Convolutional Neural Network (CNN) U-Net model to perform automated ROI segmentation of the kidneys in Tc-99m DTPA–based renal scintigraphy images. The image dataset underwent preprocessing, normalization, and data augmentation, and was then split into training, validation, and testing sets. Model performance was evaluated using the Dice Coefficient on both validation and testing datasets. The results showed an average Dice Coefficient of 0.900 on the validation set and 0.889 on thetesting set. Frame-by-frame analysis demonstrated stable model performance across all acquisition phases, with Dice Coefficient values ≥ 0.87. These findings demonstrate that the U-Net model can accurately and consistently segment kidney ROIs, and has the potential to be integrated into clinical decision-support systems to enhance the efficiency and consistency of renal scintigraphy interpretation.
Keywords: U-Net, Medical Image Segmentation, Renal Scintigraphy, Nuclear Medicine, Deep Learning.
ABSTRAK
Renal scintigraphy merupakan prosedur kedokteran nuklir yang umum digunakan untuk menilai fungsi ginjal secara kuantitatif dan memantau berbagai kondisi klinis. Proses analisis citra memerlukan segmentasi region of interest (ROI) ginjal, yang umumnya dilakukan secara manual oleh operator berpengalaman. Metode manual ini memakan waktu dan rentan terhadap variabilitas antar- pengamat. Penelitian ini mengembangkan dan mengevaluasi model Convolutional Neural Network (CNN) U-Net untuk melakukan segmentasi otomatis ROI ginjal pada citra renal scintigraphy berbasis radiofarmaka Tc-99m DTPA. Dataset citra yang digunakan melalui proses pra-pemrosesan, normalisasi, dan data augmentation, kemudian dibagi menjadi data latih, validasi, dan uji. Evaluasi kinerja model menggunakan metrik Dice Coefficient pada dataset validasi dan uji. Hasil menunjukkan nilai rata-rata Dice Coefficient sebesar 0,900 pada data validasi dan 0,889 pada data uji. Analisis per frame menunjukkan stabilitas performa model di seluruh faseperekaman, dengan Dice Coefficient ≥0,87. Temuan ini membuktikan bahwa model U-Net mampu melakukan segmentasi ROI ginjal secara akurat dan konsisten, serta berpotensi diintegrasikan dalam sistem pendukung keputusan klinis untuk meningkatkan efisiensi dan konsistensi interpretasi citra renal scintigraphy.
Kata Kunci: U-Net, Segmentasi Citra Medis, Renal Scintigraphy, Kedokteran Nuklir, Deep Learning.
Teks Lengkap:
Download ArtikelReferensi
Bi, L., Kim, J., Kumar, A., Feng, D., & Fulham, M. (2020). Automatic liver lesion detection using cascaded deep residual networks. IEEE Journal of Biomedical and Health Informatics, 24(4), 1027–1036.
Bikbov, B., Purcell, C. A., Levey, A. S., Smith, M., Abdoli, A., Abebe,M., ... & Murray, C. J. L. (2020).
Global, regional, and national burden of chronic kidney disease, 1990–2017: a systematic analysis. The Lancet, 395(10225), 709–733. DOI: https://doi.org/10.1016/S0140- 6736(20)30045-3
Bomanji, J., Gacinovic, S., & De Palma, D. (2014). Nuclear medicine in clinical diagnosis and treatment. CRC Press.
Chen, L. C., Papandreou, G., Kokkinos, I., Murphy, K., & Yuille, A. L. (2018). Deeplab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs. IEEE Transactions on Pattern Analysis and Machine Intelligence, 40(4), 834–848.
Çiçek, Ö., Abdulkadir, A., Lienkamp, S. S., Brox, T., & Ronneberger, O. (2016). 3D U-Net: Learning dense volumetric segmentation from sparse annotation. International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), 424–432.https://doi.org/10.1007/978-3- 319-46723-8_49
Erdi, Y. E., Mawlawi, O., Larson, S. M., & Imbriaco, M. (2017). Segmentation of lung lesions in PET scans: A comparison of manual, semiautomatic, and automatic methods. European Journal of Nuclear Medicine and Molecular Imaging, 44(1), 121–130.https://doi.org/10.1007/s00259-016-3480-5
Falk, T., Mai, D., Bensch, R., Çiçek, Ö., Abdulkadir, A., Marrakchi, Y.,... & Ronneberger, O. (2019). U- Net: deep learning for cell counting, detection, and morphometry. Nature Methods, 16(1), 67–70. DOI:https://doi.org/10.1038/s41592-018-0261-2
Gopalakrishnan, V., Somasundaram, K., & Subashini, T. S. (2019). Kidney segmentation in CT images using hybrid deep learning model. Journal of Medical Systems, 43(5), 1–12. https://doi.org/10.1007/s10916-019-1248-7
Gordon, I., Piepsz, A., & Sixt, R. (2014). Guidelines for standard and diuretic renogram in children. European Journal of Nuclear Medicine and Molecular Imaging, 38(6), 1175–1188.
Ha, S., Park, B. S., Han, S., et al. (2024). Deep learning-based measurement of split glomerular filtration rate with 99mTc-DTPA renal scan. European Journal of Nuclear Medicine and Molecular Imaging.https://doi.org/10.1007/s00259-024-06755-0
Hill, N. R., Fatoba, S. T., Oke, J. L.,Hirst, J. A., O’Callaghan, C. A., Lasserson, D. S., & Hobbs, F. D. (2016). Global prevalence of chronic kidney disease–A systematic review and meta- analysis. PLoS ONE, 11(7), e0158765.
Isensee, F., Petersen, J., Klein, A., Zimmerer, D., Jaeger, P. F., Kohl, S., ... & Maier-Hein, K. H. (2018). nnU-Net: Self-adapting framework for U-Net-based medical image segmentation. arXiv preprint arXiv:1809.10486.
Jha, V., Garcia-Garcia, G., Iseki, K., Li, Z., Naicker, S., Plattner, B.,... & Yang, C. W. (2013). Chronic
kidney disease: global dimension and perspectives. The Lancet, 382(9888), 260–272.
Kumar, D., Wong, A., & Clausi, D. A. (2018). Lung nodule classification using deep features in CT images. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 1340– 1348.
Levey, A. S., & Coresh, J. (2012). Chronic kidney disease. The Lancet, 379(9811), 165–180.
Litjens, G., Kooi, T., Bejnordi, B. E., et al. (2017). A survey on deep learning in medical image analysis. Medical Image Analysis, 42, 60–88. https://doi.org/10.1016/j.medi a.2017.07.005
Nurfadhillah D, Santoso G, Fatimah, Murti Wibowo G, Darmini, Nuryatno. Effectiveness of Automatic Detection of Osteoarthritis using Convolutional Neural Network (CNN) Method with DenseNet201 on Digital Images of Knee Joint Radiography. Isnanto R, Hadiyanto, Warsito B, editors. E3S Web Conf [Internet]. 2023 Nov 17;448:02052. Available from: https://www.e3s- conferences.org/10.1051/e3sco nf/202344802052
Nelder, J. A., & Wedderburn, R. W. (2018). Generalized linear models. Journal of the Royal Statistical Society: Series A (General), 135(3), 370–384.
Oktay, O., Schlemper, J., Folgoc, L. L., Lee, M., Heinrich, M.,Misawa, K., ... & Rueckert, D. (2018). Attention U-Net: Learning where to look for the pancreas. arXiv preprint arXiv:1804.03999.
Phung, S. L., Bouzerdoum, A., & Chai, D. (2005). Skin segmentation using color pixel classification: Analysis and comparison. IEEE Transactions on Pattern Analysis and Machine Intelligence, 27(1), 148–154.
Piepsz, A., & Ham, H. (2010). Pediatric applications of renal nuclear medicine. Seminars in Nuclear Medicine, 40(1), 16–35.
Rajpurkar, P., Irvin, J., Zhu, K., Yang, B., Mehta, H., Duan, T., ... & Ng,A. Y. (2017). CheXNet:
Radiologist-level pneumonia detection on chest X-rays with deep learning. arXiv preprint arXiv:1711.05225.
Ronneberger, O., Fischer, P., & Brox,T. (2015). U-Net: Convolutional networks for biomedical image segmentation. International Conference on Medical Image Computing and Computer- Assisted Intervention (MICCAI), 234–241.https://doi.org/10.1007/978-3- 319-24574-4_28
Sejin Ha, Park, B. S., Han, S., et al. (2024). Deep learning-based measurement of split glomerular filtration rate with 99mTc-DTPA renal scan. European Journal of Nuclear Medicine and Molecular Imaging.DOI:https://doi.org/10.1007/s00259-024-06755-0
Shulkin, B. L. (2014). Pediatric nuclear medicine. Springer Science & Business Media.
Taylor, A., Brandon, D., & Blaufox,M. D. (2018). Procedure standard for diuretic renal scintigraphy in adults: 2018 revision. Journal of Nuclear Medicine Technology, 46(1), 58–65.https://doi.org/10.2967/jnmt.1 17.204420
Zhang, Z., Wu, C., Coleman, S., & Kerr, D. (2018). Clinical applications of artificial intelligence in kidney disease. Nature Reviews Nephrology, 15(1), 21–33.
Zhou, Z., Siddiquee, M. M. R., Tajbakhsh, N., & Liang, J. (2019). UNet++: A nested U-Net architecture for medical image segmentation. Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support, 3–11. https://doi.org/10.1007/978-3- 030-13969-8_1
DOI: https://doi.org/10.33024/mahesa.v6i4.22074
Refbacks
- Saat ini tidak ada refbacks.
Publisher: Universitas Malahayati Lampung
Semua artikel dapat digunakan dibawah lisensi Creative Commons Attribution-ShareAlike 4.0 International License
