Hmmmm, back to the black box approach.They highlight in the ASX...

Hmmmm, back to the black box approach.

They highlight in the ASX announcement today that "significantly more" true positives were identified using EchoSolv assessment as against using human assessment. So how significant is significant? In the study at St Vincents they reported on in their presentation on 23 May they indicated that 218 true positives were identified by the specialists but 376 true positives were picked up when EchoSolv was used. That is 72% more using EchoSolv which I suggest is a very significant improvement.

So how significant was the improvement seen in the American study? Turns out it was a 15% improvement. Hmmmm. True, things go better with EchoSolv but that improvement in the US study is quite insignificant compared to the result from the St Vincents study. Is a 15% improvement enough to convince hospitals and specialists in the US to sign up to EchoSolv? And how can there be such a disparity between the two studies? Maybe yank cardiologists are simply better at their jobs than the Aussie specialists and have less need for AI back-up? I would have liked to have seen some acknowledgement of the disparity and an attempt to explain why that would have happened. Also no mention of any false positives using EchoSolv (aka specificity), surely they are not suggesting that EchoSolv never returns a positive result which in fact should be negative.

Also, would be good to hear either the Chief Commercial Officer or Prof Strange to give a debrief of their trip to the States.

Anyway, looking forward to what the US FDA says about EchoSolv in August.