Chest X-Ray Image Classification using Transfer Learning with Convolutional Neural Networks The objective of the thesis was to evaluate the performance of a transfer learning system based on Convolutional Neural Networks on medical images. Firstly, a comprehensive systematic literature study conducted to present the most recent state of transfer learning systems on medical images. Secondly, on a pneumonia chest x-ray image dataset, Convolutional Neural Network and transfer learning models were implemented. Finally, the techniques were tested on a medical image dataset, and the results evaluated the effectiveness of the techniques used. The VGG16 model achieved 92.46% test accuracy and an AUC score of 90.55% in classifying the chest X-ray images as Normal or Pneumonia on the chest X-ray image dataset.

Healthcare Predictive Analytics • Utilized R, Knime, and PMML for predictive analytics in the healthcare sector. • Designed advanced analytics-based healthcare products using machine learning and predictive analytics, enabling healthcare organizations to optimize staffing and reduce patient readmission rates by predicting inflow and length of stay.

Supply Chain Management Predictive Analytics • Spearheaded predictive analytics in supply chain management, focusing on procurement data analysis to derive actionable insights for informed decision-making. • Developed predictive models to assess vendor performance key performance indicators (KPIs) such as demand vs. supply, delivery performance (including lead time and accuracy), return and quality rejection rates, and procurement cycle time analysis.