Penelitian ini mengembangkan Computer Aided Diagnosis CAD radiografi paru anak dengan menggunakan metode segmentasi Deformable Models untuk membantu mendeteksi abnormalitas. Metode Deformable Models mencari abnormalitas berdasarkan nilai piksel citra. Metode Deformable Models dikerjakan dengan dua variasi yaitu median filter Deformable Models dan wiener filter Deformable Models. Nilai piksel paru-paru abnormal dengan segmentasi wiener filter Deformable models adalah 186-255 dan median filter Deformable Models adalah 191-255. Metode wiener filter Deformable models menghasilkan nilai ROC lebih tinggi dibandingkan metode median filter dengan nilai akurasi 78,5, sensitivitas 74,5, spesifitas 80, presisi 90,0 dan overall error 21,0.

---

This study developed a correlation test Computer Aided Diagnosis CAD radiographic of children pulmonary using segmentation Deformable Models method for detecting Abnormalities. Deformable models method searched abnormalities by value of the image pixel. Deformable models method used two variations, namely median filter Deformable Models and wiener filter Deformable Models. Abnormal result lung pixel values with segmentation Wiener filter Deformable models is 186 255 and median filter Deformable Models is 190 255. Wiener filter Deformable models method have ROC result relatively higher than median filter Deformable models with value of accuracy 78,5, sensitivity 74,5, specificity 80.0, precision 90,0 and overall error of 21,0.