PharmaX

PharmaX, one stop solution for all your development and commercialization/access requirements.

Comment

Inspiration

I have been working in the pharmaceutical consulting space for about more than 7 years and observed that with AI we can do a lot of things in much efficient manner. So, I started working on this project with the team and now we are on the way to support pharmaceutical companies in bringing their innovative therapies quicker to the market and cost-efficiently.

What it does

It is a platform where one can do market landscape study for any indication, competitive intelligence and analogue assessment, build pricing strategy for the upcoming drug, build reimbursement and access strategy, then can also perform systematic literature review, KOL (key opinion leader) identification and mapping along with personalized outreach to KOLs for the scientific dissemination of upcoming therapeutic.

How we built it

We were doing it manually for more than 6-7 years now and had all the knowledge on how we can do something with AI. Once we applied GenAI and ML algorithms on sample dataset it worked well, but we identified a lot of challenges which were to be solved. So, we worked on resolving them and now we are about to come up with the final product.

Challenges we ran into

Key challenges we ran into were related to data procurement, safety, and security. For which we partnered with a few data providers, ensured that we are ethical and then with ML algorithms and GenAI challenges were more related to accuracy and crispness. Which we overcame with fine-tuning, and self-learning approaches.

Accomplishments that we're proud of

Major accomplishment was to build an extensive knowledge graph with more than 25 lakh documents. Which actually helped us in achieving 20% of the outcomes from this platform.

What we learned

Learnings are still going on, and will be a part of journey. But we learnt that AI can do amazing tasks but we have to be patient, and thoughtful along with the dedication. In terms of technical understanding we learnt a lot on how different databases can interact with each other to address the challenges across the value chain of drug from ideation to commercialization.

What's next for PharmaX

Next for PharmaX is to make this accessible by all the pharmaceutical companies across the globe with regular feedbacks to understand how we may further improve it. Also, along with this we want to streamline 40% of the consulting work being outsourced to a lot of consulting firms with this platform.

Built With

  • azure
  • azureai
  • clinicaltrialregisteries
  • genai
  • knowledgegraph
  • machine-learning
  • microsoft
  • patentauthorities
  • pubmedcentral
Share this project:

Updates