Working on optical tracking across different sports, addressing various subproblems including camera calibration, player detection, team classification, reidentification, jersey recognition, and pose estimation.

Building an end to end solution to automate the collection of optical tracking data across multiple sports:

• R&D to build ML models, including camera calibration, player detection, team recognition, player reidentification and pose detection
• Development of the infrastructure required to train, deploy and monitor models in production, using tools like Kubeflow, Torchserve, FastAPI, Kubernetes, Docker, Prometheus and Grafana

Conducted research aimed at defining radiomics signatures that can become imaging biomarkers of response to targeted-therapies and immunotherapy.

Conducted a study of multishot acquisitions in diffusion MRI under rigid motion. A joint algorithm that both reconstructs the image and estimates the phase maps was developed.

Conducted a study about noise filtering using the random matrix theory and the Marchenko-Pastur distribution. A new method for noise estimation was developed exploiting similarity between non-local patches within the image.

Conducted a study about the appearance of spatially-varying noise in multi-echo MRI acquisitions due to accelerated GRAPPA reconstructions. A new algorithm for noise characterization was proposed.

Received an FPI scholarship to pursue my PhD thesis titled "Artifact Reduction In Magnetic Resonance Imaging: Noise Modelling In 2D/3D GRAPPA Accelerated Acquisitions And Motion–Induced Ghosting Correction In Multishot Diffusion MRI".

Completed a 5-year degree in Telecommunications Engineering, where we studied the transmission of information across different levels in the stack: electromagnetic physics, electronics, telematics protocols, signal processing and programming.