[ABSTRAK
Intensitas keabuan yang sangat dekat memungkinkan terjadinya kesalahan dalammenginterpretasikan citra hasil Computed Radiography (CR). Maka diperlukanalgoritma yang dapat mempermudah tim medis mendiagnosa kondisi pasienkhususnya bagian paru. Penelitian ini menggunakan tingkat keabuan /intensitascitra sebagai dasar clustering dan segmentasi Region of Interest (ROI ) yang akandilakukan dengan sistem komputerisasi. Sehingga hasil pembacaan lebih akuratdibanding secara manual. Data sampel berupa 100 citra hasil CR pasien parudewasa Rumah Sakit Pusat Pertamina yaitu 50 citra norma sebagai citra acuan dan50 citra uji (normal dan abnormal). Pada clustering diuji coba dengan jumlahcluster (k) bervariasi yaitu 3, 4, .., 10. Citra hasil clustering yang terbaikditunjukkan pada k = 8 karena dapat memvisualisasikan batas warna dengan lebihjelas dibanding dengan k yang lain. Pada segmentasi ROI, citra paru dibagimenjadi 33 daerah sesuai posisi anatomi paru yang terdiri dari 6 daerah apex, 11daerah hilum dan 16 daerah peripheral. Selanjutnya, masing-masing daerahpembagian diukur intensitasnya. Intensitas citra acuan dijadikan dasar untukmenentukan abnormalitas citra uji, intensitas citra uji yang lebih tinggi dariintensitas citra normal dikategorikan sebagai citra abnormal. Akurasi sistem padapenelitian ini adalah 66%.
ABSTRACT
Gray intensity is very close to allow for errors in interpreting the ComputedRadiography (CR) image. It would require an algorithm that can facilitate medicalteam to diagnose the patient's condition especially the lungs. Clustering k-meansclustering and segmentation Region of Interest (ROI) will be done by acomputerized system based on the image gray level / intensity. 100 CR imageused as the sample data from Rumah Sakit Pusat Pertamina, 50 image asreferences images and 50 images as tested image. On clustering tested by thenumber of clusters (k) varies the 3, 4, .., 10. The clustering of the best imageresults are shown in k = 8 because it can visualize the color boundaries moreclearly than the other k. At ROI segmentation, lung image is divided into 33regions corresponding anatomical position lung consist of 6 regional apex, hilumarea 11 and 16 peripheral areas. Furthermore, each regional division of themeasured intensity. The intensity of the reference image used as the basis fordetermining abnormality test images, test image intensity higher than normalimage intensity categorized as abnormal image. The system accuracy in this studywas 66%., Gray intensity is very close to allow for errors in interpreting the ComputedRadiography (CR) image. It would require an algorithm that can facilitate medicalteam to diagnose the patient's condition especially the lungs. Clustering k-meansclustering and segmentation Region of Interest (ROI) will be done by acomputerized system based on the image gray level / intensity. 100 CR imageused as the sample data from Rumah Sakit Pusat Pertamina, 50 image asreferences images and 50 images as tested image. On clustering tested by thenumber of clusters (k) varies the 3, 4, .., 10. The clustering of the best imageresults are shown in k = 8 because it can visualize the color boundaries moreclearly than the other k. At ROI segmentation, lung image is divided into 33regions corresponding anatomical position lung consist of 6 regional apex, hilumarea 11 and 16 peripheral areas. Furthermore, each regional division of themeasured intensity. The intensity of the reference image used as the basis fordetermining abnormality test images, test image intensity higher than normalimage intensity categorized as abnormal image. The system accuracy in this studywas 66%.] |
T43838-Atina.pdf :: Unduh