Data Anaylsis and AI Projects
• Development of an AI Model with Radiomic-Patomic Integration for Glioblastoma Multiforme Diagnosis
This study aims to develop a non-invasive decision support system based on magnetic resonance imaging for Glioblastoma Multiforme (GBM) diagnosis and treatment processes. A multimodal dataset was created using radiomic features extracted from DICOM and NIfTI format MR images with the CaPTk library and patomic morphometric data obtained from NDPI format histopathological slides with HistomicsTK. The developed model can predict 12 critical histopathological parameters from MR images and shows potential to significantly reduce biopsy requirements.
• Quantitative Profiling of Glioblastoma Multiforme Heterogeneity: Radiomic-Patomic Clustering Analysis
This complementary study focuses on the quantitative characterization of tumor heterogeneity in Glioblastoma Multiforme using advanced clustering techniques. Principal Component Analysis (PCA) and hierarchical clustering were applied to identify distinct tumor subregions based on radiomic features. Four distinct clusters were identified with different characteristics including gradient variations, Haralick contrast patterns, intensity distributions, and necrotic regions. ANOVA analysis revealed significant differences in intensity parameters, establishing quantitative biomarkers for tumor heterogeneity assessment.
• Survival Prediction in COVID-19 Patients: Insights from Biochemical and Radiological Analysis
In this project, statistical analyses were conducted using various biochemical and radiological parameters to predict survival outcomes in COVID-19 patients. Differences between the COVID-19 and control groups were analyzed using the Wilcoxon rank-sum test and independent two-sample t-test. Mann-Whitney U tests were applied, and a logistic regression model was developed for survival prediction. The model's performance was assessed using ROC curve analysis, and post-hoc power analysis was conducted for significant results. This study aims to identify key biomarkers for predicting survival in COVID-19 patients.
• Exploring Genotype-Phenotype Associations: A Comprehensive Analysis
This project employs the Apriori algorithm to uncover significant associations between genotypes and phenotypes. By analyzing genetic data alongside phenotypic characteristics, the study aims to identify potential markers for specific traits. The results provide insights into genetic influences on phenotypes, contributing to the understanding of genotype-phenotype relationships. This analysis has implications for fields such as personalized medicine and genetic research, offering a data-driven approach to exploring genetic correlations.
• Exploring Genetic Diversity in Human Populations: The 1000 Genomes Project
This project focuses on the 1000 Genomes Project, which aimed to create the most detailed map of human genetic variation. By sequencing the genomes of over 2,500 individuals from diverse populations, this study provides insights into human genetic diversity and its implications for health and disease. The project employs bioinformatics tools to analyze genetic data, highlighting variations associated with various traits and conditions. The findings contribute to a deeper understanding of human genetics and the role of genetic variation in health, offering valuable resources for researchers in genomics and personalized medicine.
• Investigating the Role of the Prolactin Gene in Health and Disease
This project focuses on analyzing the prolactin gene and its associated variations, which play a crucial role in various physiological processes, including lactation and reproductive health. By examining genetic data related to the prolactin gene, the project aims to uncover its potential associations with health conditions such as infertility, hormonal imbalances, and certain cancers. Utilizing bioinformatics tools and statistical methods, this study provides insights into how genetic variations can influence prolactin levels and overall health outcomes. The findings are intended to contribute to the understanding of the prolactin gene's implications in clinical settings and its potential as a biomarker for various health conditions.
• Decoding Athletic Performance: Genetic Variations and Their Impact
In this project, information from a MySQL database is utilized to analyze the relationship between genetic variations and physical performance scores. By merging relevant tables in the database, genotype data related to specific genes and user performance metrics is obtained. This analysis helps in understanding the impact of genetic factors on individuals' physical abilities. Genotype data is encoded for statistical modeling, and Ordinary Least Squares (OLS) regression is employed to examine the effects of specific genotypes on motor performance scores. The project provides insights into genetic factors influencing athletic abilities, enabling the development of personalized training and health strategies.