Topic

AI LCA Tool for Healthcare Industry

Tushar Sharma
Tushar Sharma • 8 May 2024

Hi Everyone, 

We are looking for feedback for our AI LCA tool. It can be used by anyone to start measuring emissions of various medical products and services. The tool is built on an in-house research of 1000s of manufacturing process and energy inputs associated with the medical supply chain. I would really appreciate any feedback you might have for us:

App Link: https://app.ecomedic.io/ai
App Demo: https://youtu.be/iotdbcRSux8 

There is a feedback button on left side panel of the app. It uses Exiobase for emission calculations. Ecoinvent integration coming soon. 

Many thanks in advance. 

Best,
Tushar

Comments (2)

Matthew Sawyer
Matthew Sawyer

Thanks Tushar. Ive had a play with the tool. It seems to give pretty random results tho. The references dont appear to relate to the questions asked.

If I follow the video and click 'Create Table', it says
'Error fetching data tables
Invalid `prisma.dataTable.findMany()` invocation: The column `(not available)` does not exist in the current database.'

The text seems ok for a GCSE science answer, but I dont think I could use it in a report/project etc currently.

Tushar Sharma
Tushar Sharma

Hi Matthew,

Thanks so much for trying the app. Sorry for the issue. It is all fixed now. Regarding the response from AI, we have built the app around the LCA work flow. E.g. When asking for energy inputs for a material and process, it will provide data from published scientific sources. The plan is to get published data for all manufacturing processes and energy inputs related to medical products.

Any generic questions such as what is the carbon footprint of Atrial Fibrillation will generate generic responses from the web. Here the links may or may not be correct as it doesn't use our database. In the future, we could attach it to published sources of such questions if that is what people really want.

We aim to help conduct LCA for medical products within an hour. It will always be more accurate and comparable than say a published data from a year ago. Thanks!


Please log in or sign up to comment.