nvahmadi / monai Goto Github PK

I am Seyed-Ahmad Ahmadi (@nvahmadi), Senior Solution Architect at NVIDIA, for healthcare and life sciences accounts in the EMEA region.

I am interested in applied machine & deep learning in medicine, especially its translation into clinical environments. More generally, I am interested in using and combining various data types in the healthcare domain and beyond:

• Tabular data: mixed-type structured data in the wild (incl noisy and missing features). Preferred tools: NVIDIA's accelerated data science library RAPIDS, which offers GPU-accelerated versions of:
  • cuDF, a fast version of the most popular features of pandas, sharing the same API
  • cuML, a fast version of the most popular models and methods in scikit-learn, sharing the same API
  • cuGraph, a fast version of the most popular models and methods in networkX, sharing the same API
• Grid data: in 1D/2D/3D, mostly in the domain of medical image analysis. Preferred tool: MONAI.
• Text/language data: NLP using LLMs and RAG agents for retrieval from knowledge bases. Preferred tool: NeMo Framework
• Graph data: graph analytics and graph neural nets (GNNs)/Geometric Deep Learning. Preferred tool: PyG, via our PyG containers on NGC.

My main interest is to use deep learning for multimodal fusion of above data types, on healthcare data (multi-omics fustion), and other domains.

If you are interested in my publications, please have a look at my Google Scholar profile.

If you are interested in connecting with me, please visit my LinkedIn profile.