(For USM Staff/Student Only)

EngLib USM > Ω School of Electrical & Electronic Engineering >

‫Image segmentation using ant and bee algorithms

‫Image segmentation using ant and bee algorithms / Khor Wai Peng
Kanser payudara telah menjadi salah satu daripada punca utama kematian akibat kanser. Oleh itu, mengenal pasti sel-sel tumor dalam imej-imej Pengimejan Resonans Magnet (atau lebih dikenali sebagai MRI) sangat membantu proses rawatan pesakit pada peringkat awal kanser. Justeru itu, ia berupaya mengurangkan kadar kematian akibat kelewatan mendapatkan rawatan. Dalam hal ini, mengenal pasti dengan tepat sel sel tumor dalam imej-imej MRI payudara menjadi amat penting. Ini termasuk membezakan antara tisu-tisu kulit dan tisu-tisu payudara dan juga mensegmentasi sel sel tumor daripada tisu-tisu payudara. Haiwan seperti semut dan lebah mampu mengenalpasti sumber makanan terbaik untuk ikhtiar hidup. Cara mereka telah menjadi bahan kajian dan berkembang dan dikenali sebagai algoritma berinspirasikan haiwan. Penggunaan algoritma berasaskan pengoptimuman koloni semut dan lebah tiruan dapat membantu dalam mengenalpasti sel-sel tumor daripada tisu-tisu payudara. Dengan menjadikan sel-sel tumor sebagai sumber makanan terbaik, kedua-dua algoritma haiwan telah diaplikasikan untuk mencari sel-sel tumor dengan bantuan operasi-operasi pra pemprosesan. Operasi-operasi pra-pemprosesan seperti menggunakan penuras median untuk mengurangkan hingar, pengambangan untuk mengurangkan luas pencarian kawasan-kawasan penting, operasi-operasi morfologi untuk menutup lubang serta mengeluarkan titik-titik dan juga pelabelan kawasan-kawasan penting dapat menyediakan kawasan pencarian untuk pengaplikasian algoritma semut dan lebah. Kedua-dua algoritma mampu mengenalpasti secara tepat sel-sel tumor daripada imej imej MRI payudara namun algoritma lebah mengambil masa pemprosesan yang lebih singkat. _______________________________________________________________________________________________________ With breast cancer being one of the main cause of cancer death, identifying tumor cells in breast Magnetic Resonance Imaging (MRI) images is very helpful for patients to undergo treatment at an early stage of cancer and thus reduce the rate of fatalities due to last minute treatment. In this respect, it is very important to identify accurately the tumor cells in the MRI images of breasts. This includes differentiating the skin tissues from breast tissues as well as segmenting the tumor cells from the breast tissues. Creatures such as ants and bees have been able to identify best food sources to survive. Their methods have been used as subjects of studies and evolved into animal inspired algorithms. Applying the Ant Colony Optimization (ACO) and Artificial Bee Colony (ABC) Optimization algorithms is helpful in identifying tumor cells from among the breast tissues. By treating the tumor cells as the best food source, both the ant and bee algorithms are applied to find the tumor cells with the help of some preprocessing operations. Preprocessing operations such as using median filter to reduce noise, thresholding to reduce area of search for region of interest, morphological operations to close holes and remove dots and also labelling of regions of interest prepares the area of search for the application of the ant and bee algorithms. Both algorithms are able to identify the tumor cells from the breast MRI images, with a difference in speed whereby the bee algorithm takes lesser time.
Contributor(s):
Khor Wai Peng - Author
Primary Item Type:
Final Year Project
Identifiers:
Accession Number : 875004776
Barcode : 00003096396
Language:
English
Subject Keywords:
breast cancer; tumor; the MRI images
First presented to the public:
6/1/2012
Original Publication Date:
3/12/2018
Previously Published By:
Universiti Sains Malaysia
Place Of Publication:
School of Electrical & Electronic Engineering
Citation:
Extents:
Number of Pages - 83
License Grantor / Date Granted:
  / ( View License )
Date Deposited
2018-03-12 11:53:33.87
Date Last Updated
2019-01-07 11:24:32.9118
Submitter:
Mohd Jasnizam Mohd Salleh

All Versions

Thumbnail Name Version Created Date
‫Image segmentation using ant and bee algorithms1 2018-03-12 11:53:33.87