Teoh, Chai Ling (2024) Modeling and Quantitative Measurement of Nuclear Pleomorphism in Breast Histopathology Images Using a Novel Harmonic Mean Method. Masters thesis, Tunku Abdul Rahman University of Management and Technology.
Text
2. Teoh Chai Ling (MES).pdf Restricted to Registered users only Download (3MB) |
Abstract
To date, breast cancer remains a global challenge. Recent reports found that the incidence rates of breast cancer are soaring in developed countries whereas high mortality rate is reported in undeveloped countries, mainly due to late diagnostics. In 2020, a total of 684,996 deaths (15.5% of all cancer cases amongst women) were reported, proclaimed as the leading mortality (cancer) amongst women across the globe. To determine the breast cancer grade, the Nottingham Histopathology Grading (NHG) system is used. Nuclear pleomorphism is one of the core features underpinning the system, contributing to the overall grade of breast cancer. Several recent works demonstrate promising capability in the detection of nuclear pleomorphism in breast cancer. These methods however partly or entirely different from the description stipulated by the NHG system, barricading the implementation in real-world applications. In addition, the in-depth knowledge of nuclear pleomorphism is very limited, and investigations on the morphological features of the pleomorphic nucleus are still new and have yet to be comprehended thoroughly. In this study, the main focus lies on the quantification of the pleomorphic nucleus/ nuclear pleomorphism in breast cancer according to the NHG system. The breast histopathology slides were obtained from the Pathology Department, Hospital Tuanku Fauziah, Kangar, Perlis, Malaysia. A total of 48 breast histopathological slides from 48 patients (one slide per patient), with anonymous identities, were collected. Briefly, the study quantifies nuclear pleomorphism across scores 1 to 3 and validates the proposed measurement method against ground truth data. Initially, the descriptions of nuclear pleomorphism were first deconstructed into three core elements, namely size, shape, and appearance. These elements are subsequently mathematically modeled into equations, termed
Item Type: | Thesis / Dissertation (Masters) |
---|---|
Subjects: | Technology > Technology (General) Technology > Engineering (General) |
Faculties: | Faculty of Engineering and Technology > Master of Engineering Science |
Depositing User: | Library Staff |
Date Deposited: | 31 Dec 2024 04:20 |
Last Modified: | 31 Dec 2024 04:20 |
URI: | https://eprints.tarc.edu.my/id/eprint/31409 |