Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

With a nudge from AI, ketamine emerges as a potential rare disease treatment

statnews.com

Post reply on HN
Print view View original

51–60 of 140 posts

Re: With a nudge from AI, ketamine emerges as a potential rare disease treatment

#51

Post by PragmaticPulp »

chillwaves wrote: post #46
PragmaticPulp wrote: post #42

Earlier quoted context omitted.

> Are we going to see similar harms from these drugs a few years after they gain more popularity? I think we’re already starting to see this on a smaller scale. One of my friends is a therapist who does a lot of social work predominantly for low-income people coming from difficult situations. She encounters a lot of people who have become heavily dependent on marijuana under a mistaken impression that smoking multipl…

Did you ever compare the side effects of marijuana to the popular antidepressant, antianxiety and ADHD medications being given to children? > There is a growing mistrust of pharmaceutical companies and mainstream psychiatry that mirrors the rise of things like anti-vaxxers This statement seems to ignore the crisis brought by narcotic drugs. People should be skeptical of pharmaceutical companies. There is a long histo…

There’s a lot of “whataboutism” when it comes to marijuana, but it doesn’t negate the fact that heavy daily marijuana usage is a net negative for mental health and produces worse long-term depression scores than abstaining.

And yes, she generally refers people to psychiatric treatment under the guidance of doctors who can monitor and adjust medications over time. Outcomes on mainstream depression and anxiety medications are far more positive and the people do much, much better after getting proper treatment with proper medications.

There is no comparison. Marijuana is not an effective depression treatment even though it may temporarily dull feelings and emotions. Long-term studied outcomes are poor.

Re: With a nudge from AI, ketamine emerges as a potential rare disease treatment

#52

Post by jfk13 »

ctoth wrote: post #45
jfk13 wrote: post #37

Earlier quoted context omitted.

I'm not familiar with that "awesome herb", so I wondered what other sources might have to say. While the numbers are small (unsurprisingly, if it's not nearly as widely known and used as some other substances), it may not be quite as innocuous as you seem to imply: > "A 2019 paper analyzing data from the National Poison Data System found that between 2011-2017 there were 11 deaths associated with kratom exposure. Nin…

This is the most reefer madness comment I have ever seen on Hacker News. Seven possible deaths with just Kratom and this justifies banning it? How many deaths and ruined lives will those enforcing a Kratom ban be responsible for? Perhaps we should consult the history of bans of other drugs to get some idea. Anyone claiming that their reason for adding yet another substance to The Forever War on Drugs is harm reductio…

If you're going to label scientific papers you don't like as a "wall of FUD" [edit: I see you've decided to remove this remark], I'm not sure there's much we can discuss further.

I didn't say it should be banned. Or that it shouldn't. I don't know enough to form an opinion on the question.

I just wondered how balanced and objective the post from irthomasthomas was, did a quick search for other points of view, and felt it was worth offering some additional comments for consideration.

Re: With a nudge from AI, ketamine emerges as a potential rare disease treatment

#53

Post by technothrasher »

ctoth wrote: post #45
jfk13 wrote: post #37

Earlier quoted context omitted.

I'm not familiar with that "awesome herb", so I wondered what other sources might have to say. While the numbers are small (unsurprisingly, if it's not nearly as widely known and used as some other substances), it may not be quite as innocuous as you seem to imply: > "A 2019 paper analyzing data from the National Poison Data System found that between 2011-2017 there were 11 deaths associated with kratom exposure. Nin…

This is the most reefer madness comment I have ever seen on Hacker News. Seven possible deaths with just Kratom and this justifies banning it? How many deaths and ruined lives will those enforcing a Kratom ban be responsible for? Perhaps we should consult the history of bans of other drugs to get some idea. Anyone claiming that their reason for adding yet another substance to The Forever War on Drugs is harm reductio…

The person you were replying to said nothing about banning it. They simply said that a little poking around showed the previous poster's description of it as completely harmless was possibly contested. Your knee jerk reaction that asking questions about a drug is the same as demanding a ban on it reveals a pretty strong bias on your end.

Re: With a nudge from AI, ketamine emerges as a potential rare disease treatment

#54

Post by rscho »

civilized wrote: post #22
pizza wrote: post #4

> Proof-of-concept for reasoning over medical knowledge graphs, using miniKanren + heuristics + indexing. https://github.com/webyrd/mediKanren

GitHub says it's mostly written in Scheme and Racket! Would love to hear the story behind that... is this just one researcher's idiosyncratic choice, or are these languages providing some kind of material advantage in this application? (I'm aware that Lisps have a long history in AI but I understood that to be kind of in the past at this point...)

Scheme is the ideal language for something like MiniKanren, because of its 'almost-purely-functional' nature and its powerful macros. MiniKanren is all lambdas at its core with clever macros sprinkled on top.

And also, Dan Friedman, Matt Might and Will Byrd are VERY clever people. So much for people complaning nothing useful ever comes out of Racket!

Re: With a nudge from AI, ketamine emerges as a potential rare disease treatment

#55

Post by nicoburns »

lumost wrote: post #39
pitaj wrote: post #33

Earlier quoted context omitted.

The US has been cracking down on opioid prescriptions to the point that a lot of people can't get opioids for when they really need them for chronic pain and even after surgeries! In the meantime, opioid overdoses have continued to climb. Turns out that forcing people into the black market to deal with terrible pain is not a good policy.

No one has had a good approach for opioid addiction since the first cases of morphine addiction a century ago. Banning opiate use results in deaths, restricting prescriptions moves addicts to the black market, the status quo results in more addicts.

Portugal seems to have found a decent solution: addicts can get easy access to legitimate supply, but non-addicts can't. Seems to have dramatically improved the situation there.

Re: With a nudge from AI, ketamine emerges as a potential rare disease treatment

#56

Post by rscho »

loceng wrote: post #2

"The tool, called mediKanren, scanned millions of biomedical abstracts hunting for relationships between existing compounds and the gene involved in the disease." Is this AI or someone put the effort into compiling a database and added automatic search function?

This is logic programming. So AI in the symbolic sense. And much better suited to medical 'AI' in the current state of medical information retrieval capability than most of the statistical crap that gets all the hype nowadays.

Re: With a nudge from AI, ketamine emerges as a potential rare disease treatment

#57

Post by rscho »

doit4thebitties wrote: post #10
loceng wrote: post #2

"The tool, called mediKanren, scanned millions of biomedical abstracts hunting for relationships between existing compounds and the gene involved in the disease." Is this AI or someone put the effort into compiling a database and added automatic search function?

I worked at a biomedical informatics shop that did this very thing by AI/ML 10 years ago. Essentially: automated metanalysis.

I don't think that's automated meta analysis. It's meta research, which is not the same thing. Take a look at the MediKanren papers. This is logic programming and it's not based on statistics, but first order logic.

In fact, it's a kind of expert system. But in the 80's, expert systems didn't have access to that:

https://skr3.nlm.nih.gov/SemMedDB/

Re: With a nudge from AI, ketamine emerges as a potential rare disease treatment

#58

Post by PragmaticPulp »

scythe wrote: post #38
tomato-sauce wrote: post #12

It feels like in the last few years I have been constantly hearing about new research showing “fun” drugs as cures to a wide array of medical conditions. There’s marijuana for chronic pain, mushrooms for depression, mdma for ptsd, and now ketamine. They are often talked about as if they have little or no harmful side effects. I’m glad we are moving away from the incredibly destructive drug policies that have been in…

>marijuana for chronic pain, mushrooms for depression, mdma for ptsd Is it that much of a surprise that psychoactive drugs treat psychogenic symptoms? In the particular case of pain, the two major available drug classes -- opioids and benzodiazepines -- mimic the activity of addictive recreational drugs opium and alcohol. THC is used per se instead of derivatives because of its price and safety record (contrast opium…

> Is it that much of a surprise that psychoactive drugs treat psychogenic symptoms?

The surprising part is the growing narrative that these psychoactive drugs are basically miracle cures without downsides.

Drugs like LSD are being explored as adjuncts to intense therapy spanning many sessions, but the pop-science portrayal of these drugs ignores that intense therapy and instead imagines that tripping on mushrooms or LSD is a cure for psychiatric illness. It also ignores the fact that bad trips are a very real possibility and worsening of psychiatric illness is not uncommon among illicit users of these drugs. There are plentiful reports of psychedelics causing weeks or months of dysphoria or even precipitating long-lasting episodes of major depression, and it’s not hard to find them either.

If we want to get anywhere with these substances, we need to quit exaggerating their positive effects and downplaying their negatives. That’s a setup for failure when they’re further studied and the reality can’t match the unreasonably loft pop-culture presentation of these drugs as miracle cures that act alone without any downsides.

It is, as the grandparent comment said, reminiscent of the early days of opioids when we were bombarded with stories about how they were miracle cures without downsides. The truth is that they’re helpful in controlled circumstances but can be harmful when overdosed or prescribed without supervision, which doesn’t sound that different then the situation with drugs like ketamine.

Re: With a nudge from AI, ketamine emerges as a potential rare disease treatment

#59

Post by rscho »

nickthemagicman wrote: post #13
tomato-sauce wrote: post #12

It feels like in the last few years I have been constantly hearing about new research showing “fun” drugs as cures to a wide array of medical conditions. There’s marijuana for chronic pain, mushrooms for depression, mdma for ptsd, and now ketamine. They are often talked about as if they have little or no harmful side effects. I’m glad we are moving away from the incredibly destructive drug policies that have been in…

What there a really are no or very few side effects?

There are. Hypertension, arrhythmia, confusion, delirium... Mostly at anesthetic or at least pain killer dosage, though. Ketamine can in fact be a dangerous drug when you don't know what you're doing.

Re: With a nudge from AI, ketamine emerges as a potential rare disease treatment

#60

Post by JPLeRouzic »

Some 15 years ago someone (sorry I forgot who) in Waterloo university published a paper about how to use a search engine to discover hidden relations.

It was something like this:

1. Search for the key word "A"

2. Find new keywords (b,c,..) that are associated with "A" (I did this with lingpipe [0]

3. Search each the new keywords (b,c, ...)

4. Find for each of them their associated keywords (finding (s, t, ...)

5. Repeat 3.

6. repeat 4. finding (z, x, ...)

7. If in (z, x, ...) one of them is associated with "A" then you discovered an hidden relation between "A" and one of the members of (s, t, ...) (hidden if it did not appear in the initial search).

[0] http://www.alias-i.com/lingpipe/

Post reply on HN