UI - Tesis Membership :: Back

UI - Tesis Membership :: Back

Computer aided diagnosis pada mammografi menggunakan metode segmentasi watershed = Computer aided diagnosis mammography using segmentation watershed / Samsu Herawan

Samsu Herawan; Prawito Prajitno, supervisor; Djarwani Soeharso Soejoko, examiner; Supriyanto Ardjo Pawiro, examiner; Adhi Harmoko Saputro, examiner; I Putu Susila, examiner ([Publisher not identified] , 2016)

 Abstract

ABSTRAK
Pembacaan mammografi merupakan aktifitas yang memerlukan pengetahuan dan kemampuan yang handal. Keberhasilan pengobatan kanker payudara tergantung pada deteksi dini dan diagnosis kelainan payudara. Mamografi adalah pemeriksaan terbaik yang tersedia untuk mendeteksi tanda-tanda awal kanker payudara seperti massa, kalsifikasi, asimetri bilateral dan distorsi arsitektur. Karena keterbatasan pengamat manusia, komputer memiliki peran utama dalam mendeteksi tanda-tanda awal kanker. Metode watershed diharapkan dapat memberikan informasi berbagai fitur yang menentukan kelainan dan fakta bahwa mereka sering tidak bisa dibedakan dari jaringan sekitarnya.
computer aided diagnosis mammography diharapkan dapat membantu dalam pembacaan ketidak normalan pada payudara . Segmentasi watershed dengan pemilihan filter yang tepat dapat menghasilkan citra yang bisa membantu dalam melakukan diagnosa. Untuk proses diagnosis diperlukan nilai spesifisitas dan sensitivitas yang tinggi. Hasil evaluasi pada metode watershed dan batas ambang untuk nilai sensitivitas dan spesifisitas memiliki perbedaan 45% dan 12%. evaluasi ROC kombinasi sobel watershed memiliki nilai akurasi 83% dan kombinasi prewitt watershed memiliki nilai akurasi 85%

ABSTRACT
The reading of mammography is an activity that requires knowledge and a powerful ability. Successful treatment of breast cancer depends on early detection and diagnosis of breast abnormalities. Mammography is the best available inspection to detect early signs of breast cancer such as mass, calcification, bilateral asymmetry and architectural distortion. Due to the limitations of the human observer, the computer has a major role in detecting early signs of cancer. Watershed method is expected to provide information on various features that define the disorder and the fact that they often can not be distinguished from the surrounding tissue. mammography computer-aided diagnosis is expected to assist in the reading of abnormalities in the breast. Watershed segmentation with the selection of the right filter can produce images that could help to make diagnosis. For the diagnostic process is required specificity and high sensitivity. The results of the evaluation at watershed method and the threshold for sensitivity and specificity have a difference of 45% and 12%. ROC evaluation Sobel combination watershed has a value of 83% accuracy and combination prewitt watershed has a value of 85 % accuracy

 Digital Files: 1

Shelf
 T46686-Samsu Herawan.pdf :: Download

LOGIN required

 Metadata

Collection Type : UI - Tesis Membership
Call Number : T46686
Main entry-Personal name :
Additional entry-Personal name :
Additional entry-Corporate name :
Study Program :
Subject :
Publishing : [Place of publication not identified]: [Publisher not identified], 2016
Cataloguing Source LibUI ind rda
Content Type text
Media Type unmediated ; computer
Carrier Type volume ; online resource
Physical Description xv, 97 pages : illustration ; 28 cm + appendix
Concise Text
Holding Institution Universitas Indonesia
Location Perpustakaan UI, Lantai 3
  • Availability
  • Review
  • Cover
Call Number Barcode Number Availability
T46686 15-18-769799493 TERSEDIA
Review:
No review available for this collection: 20433231
Cover