Hi largeraj - thanks and yes a couple of posts did seem to...

Hi largeraj - thanks and yes a couple of posts did seem to energise a bit of discussion.

The charade comment. Well no-one seems to know what the analysis of cough sounds adds to the prediction over and above what you get from patient reported symptoms. If it doesn’t add anything it would be a charade wouldn’t it?

The FDA meeting. All companies have a pretty stock standard formulaic paragraph to report news about meetings with the FDA. Something about forward progress, FDA very positive, clearer path to commercialisation, greatly encouraging  etc. These words are not really material to anything.

The azbo post (Blokey). In my opinion it is based on a lack of understanding about the reference diagnosis in a diagnostic accuracy study. The reference diagnosis here is not the result of a democratic vote between clinicians who are applying unstructured / non-standardised clinical judgment to a diagnosis.

Take pneumonia - the first primary outcome. The accuracy of the app was compared to WHO primary endpoint pneumonia (for children receiving chest x-ray) + clinical diagnosis for those children not receiving chest x-ray.

Step one in this trial would be to have trained the adjudicators to consistently apply the WHO interpretative standards for child x-rays. You would be hoping to achieve levels of adjudicator agreement in line with a credible benchmark – lets say the 86% achieved in a study reported here:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6059459/

If you are not achieving good agreement find the black sheep and provide better training. It is very silly to start a diagnostic accuracy trial with poor adjudicator agreement. This is because it calls into question the credibility of your reference.

Improving the quality of the reference diagnosis was in fact the whole point of RAP changing to centralised standardised adjudication between the first and second US trials.

Think about it conceptually. No-one cares two hoots whether the accuracy of a diagnostic test is better than the diagnostic agreement between two doctors in Wogga Wogga; or two doctors in your trial.  

It is a very subtle point … but one that has just cost FTT shareholders $60m. Their wound technology went up against an active comparator – standard of care. You might think standard of care means the standard of care that is delivered in practice.

But it’s not - standard of care in a clinical trial will nearly always now be best evidence based practice; even though in real life few people actually (consistently) receive this. If it were otherwise all trialists would do is scour the country for sites providing poor care to put their new treatment up against.

Translated to diagnostic accuracy … RAP has to get app agreement to a credible “best practice” diagnosis. The accuracy level it needs to achieve is where the FDA is likely to view a favourable risk / benefit scenario. And in this trial this was set at 75%.

As far as I can make out the whole basis for this adjudicator agreement argument is Tony told someone somewhere that approval only requires you to beat or be as good as adjudicator agreement. It is then repeated on HC, seems to make common sense when everything else is very complicated (how do you define clinical significance in diagnostic accuracy testing) and so it becomes a “truth”.

Perhaps ask Tony to put this argument in an ASX announcement. Ask him to provide an example of one single medical device that has ever been approved on this basis by the FDA.  Ask him to tell you who has told him of this.

Better still ask him to make the protocol public – so that investors can understand the adjudication process better. Is the protocol commercial in confidence? Only in RAP’s head. The first AI FDA approved diagnostic medical device was from IDx – the protocol (a proven winner) is here:

https://static-content.springer.com...0-6/MediaObjects/41746_2018_40_MOESM1_ESM.pdf

So basically looks to me like what RAP has here is a set of Tinker Bell clinical results.

RAP hopes that if enough people can be found at tech shows to believe that the US trial results are good results then they really will be good results. RAP are completely correct in this - RAP’s light only shines while retail investors believe in her good results; because no-one else does.

But anyway there is nothing worse than repetitive negative posts on HC … so its back to the salt mine for me.  Perhaps look at it this way though. I have given you guys something else to look forward to. You had the FDA approval to look forward to and now you will also the pleasure of dragging up these posts of mine for everyone's great entertainment when the approval comes in.