Stanford BioProject by iGEM
This work uses serum and plasma samples for predicting biomarkers for lung cancer. We are currently working on using protein data for identifying potential biomarkers. Metabolomic biomarkers play a vital role in the early identification and prediction of cancer. We present several steps divided into two phases for determining the most important metabolomic biomarkers in the blood for lung cancer prediction using Plasma and Serum samples. We used the Shapiro–Wilk Test, Bartlett’s Test, Levene’s Test, Student’s t-Test, and Kruskal–Wallis Test in the first phase to determine the potential biomarkers. Recursive Feature Elimination with Random Forest was used to identify the final most dominant metabolomic biomarker at the second phase. Lastly, we ended with Ridge Classifier and XGBoost Classifier to assess the consistency of our approaches. Despite the declining number of metabolites up to a greater level, our prediction accuracy was 100% and 90.91% for Plasma and Serum samples, respective
About Me
I am Arnav Sonavane, Electronics and Computer Science student at the University of Mumbai, trying for pushing the boundaries of technology and its applications in healthcare and artificial intelligence. My academic journey has provided me with a strong foundation in core CSE topics and mathematics. Currently, I'm juggling multiple roles that showcasing my diverse interests and skills. As a Research Intern at Harvard University, I'm working on optimizing LLM inference for VIDUR, my work being for optimization of inferencing. Simultaneously, I'm contributing to the medical imaging field as a Machine Learning Intern at JuliaHealth, where I'm developing GPU-accelerated containers to enhance healthcare technology. My passion for bridging technology and healthcare is exemplified by my role as a Research Intern at the National Institute of Health (USA). Here, I've successfully identified biomarkers for Lung Cancer with high accuracy, working with both serum and plasma samples. This experience builds upon my previous internship at UCLA, where I tackled Click Through Rate Models and synthetic data generation, honing my skills in handling large datasets and privacy-preserving techniques. Throughout my journey, I've had the opportunity to work on impactful projects. One such project is "Synthia," a synthetic data generation tool I developed using Phi-3 and NVIDIA AI Workbench. Experienced multiple hackathon wins, including the Live-AI Global Harvard-Duke Hackathon and Stanford Biohacks. These achievements reflect my ability to apply my knowledge in competitive, fast-paced environments and collaborate effectively with diverse teams. I'm also proud to have contributed to the academic community through publications. I've authored a chapter on "Integration of IoT and Quantum Computing in Manufacturing" published by IGI Global, and first-authored papers on "HepaScope: Densely Connected UNet for CT Volume-Based Liver and Tumor Segmentation" and " Enhancing Privacy in Digital Marketing: A Data Clean Room Framework for Synthetic Data Generation and CTR Prediction" currently under review. If you're interested in discussing potential collaborations, research opportunities, or simply sharing insights about the exciting developments in our field, please don't hesitate to reach out.
$ 15 USD/hr
NEW FREELANCER!
