Thamilarasan, Aunbuaresan (2025) Development of single cell analysis algorithm using AI. Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)
![]() |
Text (Full Text)
Development of single cell analysis algorithm using AI.pdf - Submitted Version Download (2MB) |
Abstract
In effort to push the boundaries of biomedical research an AI-based algorithm for single-cell analysis was developed. The objective was to increase the precision. It also aimed to enhance the efficiency of identifying and categorizing individual cells within complex biological samples. This project was undertaken to address significant questions in cellular biology and medicine. These included elucidating disease mechanisms at the cellular level and enhancing diagnostic precision. The methodology employed involved training a deep learning model on a comprehensive dataset of labeled single-cell images. Convolutional neural networks (CNNs) were utilized to extract features. They were also used to classify cells based on their morphology, marker expression and functional characteristic. The algorithm was crafted to be adaptable to various types of single-cell data. It encompasses flow cytometry and microscopy images. The findings from this study indicated that the AI algorithm markedly surpassed traditional manual analysis methods in terms of both speed and reliability. It demonstrated high accuracy in cell classification. It was capable of processing extensive datasets expeditiously. The importance of this research lies in its potential to transform single-cell studies. By automating the analysis process more expansive studies can be conducted with larger datasets. Leading to more profound understanding of cellular heterogeneity and its implications for health and disease. In conclusion, AI-driven single-cell analysis algorithm represents a significant advancement for the scientific community. It not only achieves objective of improving analysis accuracy and efficiency but also broadens the scope of study in cellular research. The findings emphasize the algorithm’s ability to manage intricate biological data which could have substantial applications in clinical diagnostics and therapeutic development.
Item Type: | Final Year Project (Project Report) |
---|---|
Uncontrolled Keywords: | Single cell analysis, Dimensionality reduction, Clustering, Auto-encoder, Cell annotation |
Subjects: | T Technology > T Technology (General) T Technology > TK Electrical engineering. Electronics Nuclear engineering |
Divisions: | Library > Final Year Project > FTKEK |
Depositing User: | Norfaradilla Idayu Ab. Ghafar |
Date Deposited: | 08 Oct 2025 02:21 |
Last Modified: | 08 Oct 2025 02:21 |
URI: | http://digitalcollection.utem.edu.my/id/eprint/36575 |
Actions (login required)
![]() |
View Item |