View Feed
group-icon
Coffee Room
Discuss anything here - everything that you wish to discuss with fellow engineers.
12920 Members
Join this group to post and comment.
mamatha paithara
mamatha paithara • Feb 10, 2019

How to evaluate every single device for a disease diagnosis,monitoring and intervention?

Artificial intelligence approach for a disease evaluation

Kaustubh Katdare
Kaustubh Katdare • Feb 11, 2019

Some more description of what exactly are you trying to do would help. I'm sure there are large number of devices being used in medical domain for diagnosis, monitoring and intervention. 

Let's take an example - the BP monitor. What kind of evaluation do you wish to do on it?

PS: Tagging @Ramani Aswath sir for inputs. 

There are many applicable international standards prescribed by agencies like US FDA, CE and such that describe what tests the device must pass including controlled clinical trials where applicable.

The problem is that with the advance of technology, a whole lot of devices are being developed for which such standards are not yet available. Even these have to be certified by the agencies for use in humans. 

They are treated on a case to case basis.

The problems are ethical as well as legal.

From an ethical point, if a device is used for diagnosis and even intervention, it may function well in a lab set up but give wrong output in real life. This can lead to harm for the patient. Legally, a healthcare facility or insurance or a clinician can get dragged into a litigation.

For example if a wearable device for Cardiac Diagnosis with built in expert system is deployed in a remote location, unless it has built in fully validated algorithms for artifacts correction, the results may be quite off the mark and lead to a hazard for the patient, who is prescribed a treatment based on these results by an absent specialist, who has to rely on the data.

Now hacking of devices has become a major problem and a lot of protection has to be included.

Share this content on your social channels -