1) Deeptek.AI (https://www.deeptek.ai/)
- Validated the Convolutional Neural Network (CNN) model’s ability to identify Pulmonary Embolisms (PE) from CT scans, increasing diagnostic accuracy by 15% compared to previous methods.
- Conducted statistical analyses on the classification and regression metrics of Deeptek’s CNN, ensuring a 98% classification accuracy and a 92% regression accuracy, validating the model’s reliability and performance.
- https://www.deeptek.ai/blog/the-indispensable-expert-in-the-loop-in-ai-based
- https://rohankall.com/wp-content/uploads/2023/01/Statistical-Analysis-of-AI-Models.pdf
- Found optimal routes for 4 waste collectors to visit each of 34 centers from the central store of a pathology firm, AG Diagnostics, achieving a 35% transport cost reduction for the firm.
- Collected and analyzed location data from 34 centers, and developed a Mixed Integer Linear Program in Julia to optimize routing efficiency and minimize total transport costs.
- https://cambridge-research.org/student-spotlights/rohans-project-commercialised-by-ag-diagnostics-a-leading-pathology-lab-in-india/
- https://rohankall.com/wp-content/uploads/2022/10/Minimizing-costs-of-Covid-19-biomedical-waste-collection-using-Mixed-Integer-Linear-Programming-approach_RohanKalluraya_01.pdf