ANALISIS KOMPONEN UTAMA - PENYESUAIAN HISTOGRAM ADAPTIF TERBATAS DAN ATT-UNET UNTUK SEGMENTASI RAMBUT

Okky Darmawan Kostidjan; Dwi  Sunaryono; Yudhi  Purwananto

doi:10.46764/teknimedia.v7i1.361

Okky Darmawan Kostidjan Institut Teknologi Sepuluh Nopember, Surabaya
Dwi Sunaryono Teknik Informatika, Fakultas Teknik Elektro dan Informatika Cerdas, Institut Teknologi Sepuluh Nopember, Surabaya
Yudhi Purwananto Teknik Informatika, Fakultas Teknik Elektro dan Informatika Cerdas, Institut Teknologi Sepuluh Nopember, Surabaya

DOI: https://doi.org/10.46764/teknimedia.v7i1.361

Kata Kunci: Gambar dermoskopi, CLAHE, PCA, Segmentasi

Abstrak

Dalam bidang analisis gambar medis, artefak seperti rambut dermal menjadi tantangan utama baik dalam interpretasi visual maupun pemrosesan gambar otomatis selama pemeriksaan dermoskopik. Rambut yang menutupi area lesi dapat menyamarkan batas lesi, mengurangi kualitas ekstraksi fitur, dan menyebabkan kesalahan segmentasi dan klasifikasi. Studi terbaru menunjukkan bahwa rambut dermal tetap menjadi salah satu artefak paling persisten yang memengaruhi analisis otomatis, bahkan dalam model segmentasi canggih. Artefak ini juga menurunkan kinerja sistem berbasis kecerdasan buatan yang bergantung pada informasi visual. Studi ini menerapkan Analisis Komponen Utama (PCA) sebagai metode pengabuan untuk mengurangi beban komputasi sambil mempertahankan fitur gambar esensial, serta Equalization Histogram Adaptif Terbatas Kontras (CLAHE) untuk meningkatkan kontras lokal dan memperjelas struktur rambut yang tipis atau berkontras rendah. Kombinasi PCA dan CLAHE berfungsi sebagai tahap prapemrosesan untuk meningkatkan kualitas gambar masukan pada model segmentasi berbasis deep learning. Evaluasi dilakukan menggunakan arsitektur AttU-Net dengan metrik Dice Similarity Coefficient (DSC) dan Jaccard Index (JAC). Prasunting PCA–CLAHE yang diusulkan mencapai nilai DSC dan JAC sebesar 75,24% dan 60,04%, masing-masing, melebihi model tanpa prasanting. Hasil ini menunjukkan bahwa PCA–CLAHE secara efektif meningkatkan kualitas gambar dan akurasi segmentasi sambil mempertahankan efisiensi komputasi.

Referensi

[1] Z. Wang et al., “Influence of Hair Presence on Dermoscopic Iimage Analysis by AI in Skin Lesion Diagnosis,” Comput. Biol. Med., vol. 183, p. 109335, Dec. 2024, doi: 10.1016/j.compbiomed.2024.109335.
[2] L. Jütte, H. Patel, and B. Roth, “Advancing Dermoscopy Through a Synthetic Hair Benchmark Dataset and Deep Learning-Based Hair Removal,” J. Biomed. Opt., vol. 29, no. 11, Nov. 2024, doi: 10.1117/1.JBO.29.11.116003.
[3] N. Lama et al., “ChimeraNet: U-Net for Hair Detection in Dermoscopic Skin Lesion Images,” J. Digit. Imaging, vol. 36, no. 2, pp. 526–535, Nov. 2022, doi: 10.1007/s10278-022-00740-6.
[4] F. Pérez-García, R. Sparks, and S. Ourselin, “TorchIO: A Python Library for Efficient Loading, Preprocessing, Augmentation and Patch-Based Sampling of Medical Images in Deep Learning,” Comput. Methods Programs Biomed., vol. 208, p. 106236, Sep. 2021, doi: 10.1016/j.cmpb.2021.106236.
[5] S. Joseph and O. O. Olugbara, “Preprocessing Effects on Performance of Skin Lesion Saliency Segmentation,” Diagnostics, vol. 12, no. 2, p. 344, Jan. 2022, doi: 10.3390/diagnostics12020344.
[6] O. A. M. F. Alnaggar et al., “Efficient Artificial Intelligence Approaches for Medical Image Processing in Healthcare: Comprehensive Review, Taxonomy, and Analysis,” Artif. Intell. Rev., vol. 57, no. 8, p. 221, Jul. 2024, doi: 10.1007/s10462-024-10814-2.
[7] A. S. Ashour et al., “Cascaded Hough Transform-Based Hair Mask Generation and Harmonic Inpainting for Automated Hair Removal from Dermoscopy Images,” Diagnostics, vol. 12, no. 12, p. 3040, Dec. 2022, doi: 10.3390/diagnostics12123040.
[8] M. Doğan and İ. A. Özkan, “Automated Hair Segmentation in Dermoscopy Images with U-Net Based Approaches,” in 2024 13th Mediterranean Conference on Embedded Computing (MECO), IEEE, Jun. 2024, pp. 1–4. doi: 10.1109/MECO62516.2024.10577890.
[9] R. K. Sidhu, J. Sachdeva, and D. Katoch, “Segmentation of Retinal Blood Vessels by a Novel Hybrid Technique- Principal Component Analysis (PCA) and Contrast Limited Adaptive Histogram Equalization (CLAHE),” Microvasc. Res., vol. 148, p. 104477, Jul. 2023, doi: 10.1016/j.mvr.2023.104477.
[10] O. Ronneberger, P. Fischer, and T. Brox, “U-Net: Convolutional Networks for Biomedical Image Segmentation,” 2015, pp. 234–241. doi: 10.1007/978-3-319-24574-4_28.
[11] F. I. Diakogiannis, F. Waldner, P. Caccetta, and C. Wu, “ResUNet-a: A Deep Learning Framework for Semantic Segmentation of Remotely Sensed Data,” ISPRS Journal of Photogrammetry and Remote Sensing, vol. 162, pp. 94–114, Apr. 2020, doi: 10.1016/j.isprsjprs.2020.01.013.
[12] Ozan Oktay et al., “Attention U-Net: Learning Where to Look for the Pancreas,” arXiv preprint arXiv:1804.03999, 2018.
[13] Y. Xiong, K. Yu, Y. Lan, Z. Lei, and D. Fan, “Hair Cluster Detection Model Based on Dermoscopic Images,” Front. Phys., vol. 12, Feb. 2024, doi: 10.3389/fphy.2024.1364372.
[14] J. Chen et al., “TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation,” 2021, doi: 10.48550/arXiv.2102.04306.
[15] D. Arous, S. Schrunner, I. Hanson, N. Frederike Jeppesen Edin, and E. Malinen, “Principal Component-Based Image Segmentation: a New Approach to Outline in vitro Cell Colonies,” Comput. Methods Biomech. Biomed. Eng. Imaging Vis., vol. 11, no. 1, pp. 18–30, Jan. 2023, doi: 10.1080/21681163.2022.2035822.
[16] Reza Karimzadeh, Emad Fatemizadeh, and Hossein Arabi, “A Novel Shape-Based Loss Function for Machine Learning-Based Seminal Organ Segmentation in Medical Imaging [pre-print],” 2022, doi: 10.48550/arXiv.2203.03336.
[17] N. P. Singh and R. Srivastava, “Retinal Blood Vessels Segmentation by Using Gumbel Probability Distribution Function Based Matched Flter,” Comput. Methods Programs Biomed., vol. 129, pp. 40–50, Jun. 2016, doi: 10.1016/j.cmpb.2016.03.001.
[18] K. Shunmuga Priya and V. Selvi, “Enhanced Skin Disease Image Analysis Using Hybrid CLAHE-Median Filter and Salient K-Means Cluster,” 2024, pp. 459–471. doi: 10.1007/978-981-97-1488-9_34.
[19] S. Chakraverti, P. Agarwal, H. S. Pattanayak, S. P. S. Chauhan, A. K. Chakraverti, and M. Kumar, “De-Noising the Image Using DBST-LCM-CLAHE: A Deep Learning Approach,” Multimed. Tools Appl., vol. 83, no. 4, pp. 11017–11042, Jan. 2024, doi: 10.1007/s11042-023-16016-2.
[20] S. Saifullah and R. Dreżewski, “Advanced Medical Image Segmentation Enhancement: A Particle-Swarm-Optimization-Based Histogram Equalization Approach,” Applied Sciences, vol. 14, no. 2, p. 923, Jan. 2024, doi: 10.3390/app14020923.
[21] Y. Yoshimi et al., “Image Preprocessing with Contrast-Limited Adaptive Histogram Equalization Improves the Segmentation Performance of Deep Learning for the Articular Disk of the Temporomandibular Joint on Magnetic Resonance Images,” Oral Surg. Oral Med. Oral Pathol. Oral Radiol., vol. 138, no. 1, pp. 128–141, Jul. 2024, doi: 10.1016/j.oooo.2023.01.016.
[22] X. Liu and T. D. C. Nguyen, “Medical Images Enhancement by Integrating CLAHE with Wavelet Transform and Non-Local Means Denoising,” Academic Journal of Computing & Information Science, vol. 7, no. 1, 2024, doi: 10.25236/AJCIS.2024.070108.
[23] S. M. Pizer, R. E. Johnston, J. P. Ericksen, B. C. Yankaskas, and K. E. Muller, “Contrast-Limited Adaptive Histogram Equalization: Speed and Effectiveness,” in [1990] Proceedings of the First Conference on Visualization in Biomedical Computing, IEEE Comput. Soc. Press, 1990, pp. 337–345. doi: 10.1109/VBC.1990.109340.
[24] F. I. Diakogiannis, F. Waldner, P. Caccetta, and C. Wu, “ResUNet-a: A deep learning framework for semantic segmentation of remotely sensed data,” ISPRS Journal of Photogrammetry and Remote Sensing, vol. 162, pp. 94–114, Apr. 2020, doi: 10.1016/j.isprsjprs.2020.01.013.
[25] H. Zhao, J. Shi, X. Qi, X. Wang, and J. Jia, “Pyramid Scene Parsing Network,” in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, Jul. 2017, pp. 6230–6239. doi: 10.1109/CVPR.2017.660.
[26] O. Oktay et al., “Attention U-Net: Learning Where to Look for the Pancreas,” Apr. 2018.