Ankitha Thyagarajan

Hands on training covering operating principles, troubleshooting, maintenance of biomedical equipment.

Training in both marketing and human resources divisions

Training on operating principles, application, maintenance of biomedical equipment, hands on training at the PCB and calibration lab

Training on nano scaffold fabrication techniques and general microbiology

Thesis : Estimating Historical Migration Rates and Population Sizes from Whole Genome Data Using Scottish Wildcats This is a research project focused on inferring demographic history from genomic data. The study used dadi (Diffusion Approximation for Demographic Inference) to analyze genetic data from two populations, Scottish wildcats and domestic cats, in order to estimate key demographic parameters such as historical migration rates , divergence time and effective population sizes. The analysis was based on joint site frequency spectrum (JSFS) data generated from whole genome sequences. Three demographic models were implemented and compared : the Isolation with Migration model, Bottleneck model, and Fractional Split with Isolation with Migration model. HPC BluePebble was utilized to generate chromosome-specific text files required for the analysis. This study demonstrates the application of population genomics and computational tools to study species divergence and gene flow.

Final Year Project: Automated Skin Disease Detection Using Deep Learning This project focuses on the classification of nine different skin diseases using deep learning techniques. The diseases included are Actinic Keratosis, Basal Cell Carcinoma, Dermatofibroma, Melanoma, Nevus, Pigmented Benign Keratosis, Seborrheic Keratosis, Squamous Cell Carcinoma, and Vascular Lesion. The system was developed using Python and trained on dermatoscopic skin images collected from Kaggle. Image preprocessing techniques such as resizing, watermark removal and segmentation were applied to enhance model accuracy. A web-based application was also developed using Python, allowing users to upload an image and obtain the predicted disease class along with the segmented lesion image on the results page. This project highlights the application of deep learning in medical image analysis and supports early diagnosis of skin diseases.