NHS - best in the world

[sasguru]

Incidentally the particular pharma in question has luckily stockpiled drugs in the UK to treat cancer patients. Let's hope they've stockpiled enough.

[Big Blue Plymouth]

Ruined by assguru

[sasguru]

Ruined by assguru

?

[vetran]

I'm building deep learning models for medical diagnosis using various modalities of medical imaging. You?
While effective, we've probably hit the limit of what they can do. You still have to run biopsies and all other tests on flagged and unflagged patients who have symptoms.
Early screening using AI is being used, but the gains so far are marginal, because the bottleneck is in the testing. Screening just provides an indication of who to test and its not foolproof by any means.

So you don't believe the advances already made in the last decade by the most respected Companies, Universities & hospitals in the world on Screening/Testing & treating cancer is not going to mark a massive opportunity to change the way cancer is treated?

Sasguru versus Google or MIT anyone want a piece of the action?

Screening if as suggested is becoming more effective earlier means that testing where indicated can be done earlier and this will reduce mortality and probably treatment costs.

If screening becomes nearly 100% effective with few false negatives then its an easy case to increase spend on biopsies (though with robotic/keyhole surgery I strongly suspect the costs have gone down).

As far as testing most biopsies are examining via microscope the extracted material and counting the dodgy cells or even in blood as it flows. It seems that too is being automated.

Application of automated image analysis reduces the workload of manual screening of sentinel lymph node biopsies in breast cancer. - PubMed - NCBI
Ultrafast automated image cytometry for cancer detection. - PubMed - NCBI


My point is that its coming we should embrace it and push ourselves up the league while building IP that the rich world will want to buy.


As always the sooner you detect & fix a disease the less pain & cost is incurred.

[Big Blue Plymouth]

?

I was talking about my bad back and now you've taken us off on one of your inneresting tangents.

[sasguru]

So you don't believe the advances already made in the last decade by the most respected Companies, Universities & hospitals in the world on Screening/Testing & treating cancer is not going to mark a massive opportunity to change the way cancer is treated?

Sasguru versus Google or MIT anyone want a piece of the action?

Screening if as suggested is becoming more effective earlier means that testing where indicated can be done earlier and this will reduce mortality and probably treatment costs.

If screening becomes nearly 100% effective with few false negatives then its an easy case to increase spend on biopsies (though with robotic/keyhole surgery I strongly suspect the costs have gone down).

As far as testing most biopsies are examining via microscope the extracted material and counting the dodgy cells or even in blood as it flows. It seems that too is being automated.

Application of automated image analysis reduces the workload of manual screening of sentinel lymph node biopsies in breast cancer. - PubMed - NCBI
Ultrafast automated image cytometry for cancer detection. - PubMed - NCBI


My point is that its coming we should embrace it and push ourselves up the league while building IP that the rich world will want to buy.


As always the sooner you detect & fix a disease the less pain & cost is incurred.

Jeez, I know you're not the sharpest tool in the box, but stop digging and think for a second.
Of course AI and tech has improved screening and other downstream activities, but if the NHS doesn't use it effectively because its a cash-strapped, Soviet, statist type of organisation, that tech might as well not exist.

We all know "the sooner you detect & fix a disease the less pain & cost is incurred". The gains from AI in "possible diagnosis" are marginal. 5-10% here or there in specificity or sensitivity is not the issue. You're going to have run the gamut of tests on everybody with a red flag symptom regardless of whether "computer says no". The real problem is downstream.

[sasguru]

I was talking about my bad back and now you've taken us off on one of your inneresting tangents.

AFAIK this thread was about the NHS and in particular why cancer outcomes are relatively poor.
Sorry we didn't go down Big Blue's bad back tangent

[vetran]

It could help, if the rest of the process caught up with it.
Ai says you *might* have lung cancer.
Unlike other rich countries, NHS maybe can't make an appt within 60 days due to lack of staff to conduct biopsies etc.
This is what's happening at the mo.

How do I know? I'm working for a non-Uk pharma in the oncology therapy area.

So three separate problems.

1. AI will tell you earlier with greater certainty and with probably less time elapsed & cost that you are very likely to have lung cancer with fewer false positives so fewer biopsies are needed.

2. There are not enough staff to perform biopsies which are starting to be replaced with less invasive tests and better surgical techniques mean it should be possible with less cost & time

Lung Biopsy | Johns Hopkins Medicine

3. Testing the result of the biopsy is currently done by sending the tissue to a lab and highly trained specialist counting and taking pictures of abnormal cells but of course AI can recognise pictures and count them faster and more consistently than any human.

two of the three could really benefit from AI and save money for the middle one.

[Zigenare]

So three separate problems.

1. AI will tell you earlier with greater certainty and with probably less time elapsed & cost that you are very likely to have lung cancer with fewer false positives so fewer biopsies are needed.

2. There are not enough staff to perform biopsies which are starting to be replaced with less invasive tests and better surgical techniques mean it should be possible with less cost & time

Lung Biopsy | Johns Hopkins Medicine

3. Testing the result of the biopsy is currently done by sending the tissue to a lab and highly trained specialist counting and taking pictures of abnormal cells but of course AI can recognise pictures and count them faster and more consistently than any human.

two of the three could really benefit from AI and save money for the middle one.

You're wrong! You Cretin!

Zig - in "sasguru" mode!

[vetran]

Jeez, I know you're not the sharpest tool in the box, but stop digging and think for a second.
Of course AI and tech has improved screening and other downstream activities, but if the NHS doesn't use it effectively because its a cash-strapped, Soviet, statist type of organisation, that tech might as well not exist.

We all know "the sooner you detect & fix a disease the less pain & cost is incurred". The gains from AI in "possible diagnosis" are marginal. 5-10% here or there in specificity or sensitivity is not the issue. You're going to have run the gamut of tests on everybody with a red flag symptom regardless of whether "computer says no". The real problem is downstream.

You are rolling all the problems into one rather breaking each of them down and analysing the possible improvements and you accuse me of not being sharp.

5-10% improvement with reduced or the same costs would be considered massive by any normal organisation. Key point here is that detection is earlier and with AI (and it seems with oncologists) the more information you give it the better it gets. If the cost falls you can automate more cheaper and do more tests. As detection gets better and you start doing more scans earlier then false negatives start to disappear you do less tests.