Deteksi Thalasemia Pada Citra Sel Darah Merah Menggunakan Mikroskop Digital Berbasis Support Vector Machine

Arisyi, Adlan (2025) Deteksi Thalasemia Pada Citra Sel Darah Merah Menggunakan Mikroskop Digital Berbasis Support Vector Machine. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5023211069-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (5MB) | Request a copy

Abstract

Talasemia adalah kelainan genetik yang menyebabkan kerusakan pada hemoglobin sebagai protein yang berperan dalam mengangkut oksigen dalam sel darah merah. Kondisi ini terjadi akibat cacat pada gen yang mengatur produksi rantai globin alfa atau beta, sehingga memicu munculnya Talasemia alfa maupun beta. Metode diagnosis saat ini, seperti Complete Blood Count (CBC) dan High-Performance Liquid Chromatography (HPLC), memerlukan fasilitas laboratorium yang canggih, yang tidak selalu tersedia secara merata di berbagai wilayah. Untuk mengatasi keterbatasan tersebut, penelitian ini mengusulkan sistem deteksi Talasemia berbasis citra digital, yang mengintegrasikan mikroskop digital portabel dengan algoritma klasifikasi berbasis machine learning. Sistem perangkat keras mencakup mikrokomputer Raspberry Pi, sensor kamera, sensor estimasi perbesaran, serta sistem optik berupa lensa okuler dan objektif dengan total perbesaran hingga 1000x. Inti dari metode yang diusulkan terletak pada perangkat lunak, yang mencakup tahapan preprocessing citra, segmentasi sel darah merah, ekstraksi fitur morfologi, tekstur, dan warna, serta klasifikasi menggunakan algoritma Support Vector Machine (SVM). Untuk meningkatkan akurasi, dilakukan seleksi fitur menggunakan Pearson Correlation Coefficient (PCC), yang menghasilkan 29 fitur paling relevan dari 64 fitur awal. Sistem ini berhasil mengklasifikasikan sel darah merah abnormal dengan akurasi sebesar 95,45%. Penelitian ini menunjukkan bahwa integrasi mikroskop digital dan machine learning berpotensi menjadi solusi deteksi dini Talasemia yang terjangkau dan mudah diakses, terutama untuk daerah dengan keterbatasan infrastruktur kesehatan.
=================================================================================================================================
Thalassemia is a genetic disorder that causes damage to hemoglobin, the protein responsible for transporting oxygen in red blood cells. This condition is caused by abnormalities in the genes that regulate the production of alpha or beta globin chains, resulting in alpha or beta thalassemia. Current diagnostic methods like Complete Blood Count (CBC) and High Performance Liquid Chromatography (HPLC) need well-equipped labs, which are not always available in all areas. To deal with these limitations, this study suggests an image-based detection system for thalassemia, which combines a portable digital microscope with machine learning-based classification algorithms. The hardware system includes a microcomputer Raspberry Pi, camera sensor, magnification estimation sensor, and optical setup consisting of an eyepiece lens and objective lens with a total magnification of 1000x. The key to the proposed method lies in the software workflow, which includes image preprocessing, red blood cell segmentation, feature extraction (morphological, texture, and color-based), and classification using a Support Vector Machine (SVM). To improve the accuracy, feature selection was carried out using Pearson Correlation Coefficient (PCC), filtering 29 highly relevant features from a larger feature set. The system successfully detected abnormal red blood cells with a classification accuracy of 95.45%. This study was conducted to explore the integration of digital microscopy and machine learning to provide an affordable and accessible solution for early thalassemia screening. The proposed method offers a promising alternative, particularly for areas with limited resources and insufficient comprehensive diagnostic infrastructure.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Mikroskop Digital, Raspberry Pi, Sel Darah Merah, Support Vector Machine (SVM), Talasemia, Digital Microscope, Raspberry Pi, Red Blood Cells, Support Vector Machine (SVM), Thalassemia
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. R Medicine > RC Internal medicine > RC78 Diagnosis, Radioscopic--Examinations, questions, etc. R Medicine > RZ Other systems of medicine T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.546 Computer algorithms T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7871.674 Detectors. Sensors
Divisions:	Faculty of Electrical Technology > Biomedical Engineering > 11410-(S1) Undergraduate Thesis
Depositing User:	Adlan Arisyi
Date Deposited:	04 Aug 2025 01:04
Last Modified:	04 Aug 2025 01:04
URI:	http://repository.its.ac.id/id/eprint/125665

Actions (login required)

View Item