Blood test that finds 50 types of cancer is accurate enough to be rolled out • Hacker News Board

Re: Blood test that finds 50 types of cancer is accurate enough to be rolled out

#331

Post by scarecrowboat » Fri, Jun 25, 2021, 8:07 PM UTC

moralestapia wrote: post #293

nevinera wrote: post #159

0.5% false positive rate is actually quite high. I don't have the numbers, but I'd be surprised if more than 1% of the people that would receive such a screening actually already have cancer , which would mean that it produces about as many false positives as false negatives. Picture a population of a million people, all receiving the test. 1% of them have cancer (unknown to them) - half of those people get a negativ…

I feel it's a net win anyway, if the procedure is not invasive and cheap, you could take one every so often (yearly?) and if something shows up you just double check with a better method or w/e. Better than nothing, which is the current alternative.

What if the followup procedure is invasive and expensive? It's possible that this procedure could be net negative in aggregate depending on the % of false positives and # of resulting needless followup procedures. Not to mention the emotional stress.

Re: Blood test that finds 50 types of cancer is accurate enough to be rolled out

#332

Post by kilotaras » Fri, Jun 25, 2021, 8:10 PM UTC

rackjack wrote: post #190

fallingknife wrote: post #186

Earlier quoted context omitted.

> but doctors can't always predict this and aren't perfect decision making machines So mass screen healthy people, collect the data, run models, and get better at it. People like you would rather not try, and this is the same reaction I get from doctors. Applying the same tech that we use to improve ad targeting to disease prediction is a no brainer to anyone whose cushy job doesn't depend on the current medieval sta…

Same vibe as "I would just simply..." What do you think the entire medical, pharmaceutical, bioinformatic, etc. industries have been doing for decades? Are you really so arrogant to think that other people are too lazy and stupid to think of your brilliant idea and that you are simply more intelligent than everybody else? Do you really think people haven't been trying?

> Are you really so arrogant to think that other people are too lazy and stupid to think of your brilliant idea and that you are simply more intelligent than everybody else?

Here's a report of a person whose wife had seasonal affective disorder. He

- deduced that a powerful enough lightbox should be able to cure SAD - didn't find any examples of such treatments in literature - spent $600 to build "LUMINATOR" - "And as of early 2017, with two winters come and gone, Brienne seems to no longer have crippling SAD—though it took a lot of light bulbs, including light bulbs in her bedroom that had to be timed to go on at 7:30am before she woke up, to sustain the apparent cure."

https://www.lesswrong.com/posts/zsG9yKcriht2doRhM/inadequacy...

Re: Blood test that finds 50 types of cancer is accurate enough to be rolled out

#333

Post by hebrox » Fri, Jun 25, 2021, 8:11 PM UTC

rackjack wrote: post #238

function_seven wrote: post #216

Earlier quoted context omitted.

If they aren't mass collecting this data in the first place (for the reasons outlined above), then how can they be doing this? Or, if they are doing this from quality samples of the population, then we have data to show that, for example, "Yes, Mr. Function Seven, your scan shows elevated levels of Widget-5a enzymes. This is often a precursor to Gadget Cancer, but 30% of the population shows this elevation, while onl…

They are mass collecting the data. They have more data than they know what to do with. Bioinformatics abuses data science at levels comparable to Google and Facebook. They are getting quality samples from the population. But it's not a simple as "high enzyme = maybe good chance of cancer". How are you going to get those enzymes from their blood in a way that can be applied to the general population? Before that, how…

Would you say that they could collect way more data?

Re: Blood test that finds 50 types of cancer is accurate enough to be rolled out

#334

Post by captainmuon » Fri, Jun 25, 2021, 8:12 PM UTC

TaupeRanger wrote: post #96

reacharavindh wrote: post #3

This is one of those medical revolutions that I am waiting dearly for. Facilities that are not hospitals(to avoid the risk of occupying medical devices that sick people need) built to _regulary_ check up otherwise healthy people for preventive care. Heck, I have so many alerts defined on my monitoring setup for servers to watch for signals of failure before they get too big. But, my own body is not observed until som…

Actually you don't want this, and you are describing a nightmare scenario that everyone who studies health policy understands all too well. Mass screening of healthy people will result in extreme iatrogenics and unnecessary psychological damage and stress (which leads to physical effects as well), not to mention overwhelming the medical system. The fact is that many things that could be detected will never result in…

If you leave the treatment criteria the same, then yes, you are absolutely right. But you have to adjust those, too.

Just a made-up example: Let's say someone has an advanced test for cancer run, and it comes back positive. You know from studies that treating people the same way as before with the advanced test leads to worse outcomes because of unneccessary treatment etc.. So what you could do is you do the simpler test afterwards. If it comes back negative, you don't treat -> you are in the same situation as before, only you can be more vigilant in future and see if the cancer grows. If the simpler test also comes back positive, you do the treatment.

It's all about getting data (statistical and otherwise) on what the best treatment is, and acting on that.

Re: Blood test that finds 50 types of cancer is accurate enough to be rolled out

#335

Post by vidarh » Fri, Jun 25, 2021, 8:14 PM UTC

Robotbeat wrote: post #309

jghn wrote: post #274

Earlier quoted context omitted.

To go further, a large issue/distraction in biomedical research has been Big Tech types coming in assuming that the roadblock all along has been lack of smart computational people in the process. I will not name names, but have seen so many instances of personal tech heroes coming in & claiming the underlying problem was that some of the brightest computational I've ever encountered simply didn't know how to computer…

While it’s absolutely true that naive non-experts can end up adding a lot more heat than light, I’ve also seen non-expert people come into stagnant domains and absolutely completely transform and improve upon the state of the art. I am skeptical of the “more data is bad” meme of screening hesitancy. It cannot be scientifically true in the strictest sense, and to the degree it’s an accurate assertion, it really seems…

It's not that more data is bad.

It's that 1) sometimes collecting the data itself is harmful at scale. E.g. mammography can cause breast cancer, or cause it to spread. 2) the patients actions as a result of the data can, and does at sufficient scale, cause further harm.

Re: Blood test that finds 50 types of cancer is accurate enough to be rolled out

#336

Post by starkd » Fri, Jun 25, 2021, 8:16 PM UTC

phunehehe0 wrote: post #202

starkd wrote: post #194

"It correctly identified when cancer was present in 51.5% of cases, across all stages of the disease, and wrongly detected cancer in only 0.5% of cases." Doesn't this seem kind of low? Just a bit better than a coin flip. Of course, it rises to 65% and 87% for certain circumstances and the false positive rate is low, but it seems like a lot of cancers could fail to be detected and give a false assurance patients are c…

This is much better than a coin flip. A coin flip would give about 50% true positive rate (which is about the same as the test), but it would also give about 50% false positive rate. To put it in perspective, imagine 10 people have cancer in a population of 1000 (a rate of 10% which is too high compared to what I think the real number should be). The test would fail to identify 5 of the people with cancer, and it wou…

I see. I guess its more important to keep false positives low than increase detection accuracy.

Thank you very much for this explanation.

Re: Blood test that finds 50 types of cancer is accurate enough to be rolled out

#337

Post by shard » Fri, Jun 25, 2021, 8:17 PM UTC

handmodel wrote: post #265

TaupeRanger wrote: post #96

Earlier quoted context omitted.

Actually you don't want this, and you are describing a nightmare scenario that everyone who studies health policy understands all too well. Mass screening of healthy people will result in extreme iatrogenics and unnecessary psychological damage and stress (which leads to physical effects as well), not to mention overwhelming the medical system. The fact is that many things that could be detected will never result in…

I'm not sure. I weight myself everyday. Some people tell me not to do that but its honestly pretty good at keeping me in check and because I weigh myself everyday I know that sometimes I just get super bloated and put on four pounds of water and it will be gone in a couple days. When you get more data you learn to adjust how you interpret it and make better decisions. Likewise, if getting a false result 50% of the ti…

Yes.

I meticulously tracked every calorie I ate for several months. Now I know how much of different types of food is how many calories, how many calories I eat per day, and most importantly what my different hunger levels mean in terms of the number of calories ingested. Combined with my weight tracking, which tells me my set weight / weight variance over a day or several days / food intake to gain or lose a certain amount of weight, I can easily control portions to hit my weight goals without thinking about it.

I would imagine more health data would allow me to optimize in this fashion as well. If I can correlate health markers to my lifestyle often and directly, I can make better and more informed choices. Advocating against easy access to health data because some people can misuse it is the same kind of nanny state thinking that says encryption shouldn't be available to the masses because criminal enterprises can use it to hide their activities.

Re: Blood test that finds 50 types of cancer is accurate enough to be rolled out

#338

Post by vmception » Fri, Jun 25, 2021, 8:17 PM UTC

nevinera wrote: post #159

0.5% false positive rate is actually quite high. I don't have the numbers, but I'd be surprised if more than 1% of the people that would receive such a screening actually already have cancer , which would mean that it produces about as many false positives as false negatives. Picture a population of a million people, all receiving the test. 1% of them have cancer (unknown to them) - half of those people get a negativ…

So still get a second opinion just as you would have anyway.

It’s still getting caught

Re: Blood test that finds 50 types of cancer is accurate enough to be rolled out

#339

Post by vidarh » Fri, Jun 25, 2021, 8:18 PM UTC

throwawaynotmd wrote: post #315

Fomite wrote: post #304

Earlier quoted context omitted.

"More information can never be harmful." In my particular field of biomedical research, there has recently been a push for "Diagnostic Stewardship" because more information has demonstrably been harmful to patient well-being.

It’s not enough to say it’s harmful. How is it harmful? A blood test, for example, cannot be inherently harmful. (Okay, no more harmful than drawing blood). So it must be about what people then decide to do based on that test. That’s what my post is about.

Do enough of them, and even basic blood tests cause harm. Infections happen. Rupturing veins happens.

If your tests are for conditions that are rare enough, and where early detection does little enough to improve outcomes, even a tiny risk like that becomes a problem.

Re: Blood test that finds 50 types of cancer is accurate enough to be rolled out

#340

Post by shard » Fri, Jun 25, 2021, 8:21 PM UTC

rackjack wrote: post #184

plank_time wrote: post #147

Earlier quoted context omitted.

We don’t give colonoscopies to 20 years olds because it’s an expensive, invasive test with a low rate of cancer. We don’t give mammograms to under 40 year olds these days because the same and it’s notoriously inaccurate. We give blood glucose level tests and cholesterol tests to everyone every year at the physical because they are cheap and easy to administer. Even though the effects of both may take decades to have…

> It correctly identified when cancer was present in 51.5% of cases, across all stages of the disease, and wrongly detected cancer in only 0.5% of cases. If all 300 million Americans get blood tested each year, that is 1 and a half MILLION people who are falsely told they have CANCER. One and a half MILLION people whose life gets turned upside down, WRONGLY, due to a blood test -- at least, until they take a second o…

Are you comparing that against the MILLIONS who get a correct diagnosis? I expect that the quality-adjusted life year total over all tested is higher with the test than without. Wouldn't that be worth it?

I think this leans more towards the "the needs of the many outweigh the needs of the few" and less towards "tyranny of the majority".