First images of cerebrospinal fluid in the brain during sleep (2019) | Hacker News Reader

First images of cerebrospinal fluid in the brain during sleep (2019)

bu.edu

347 pointsrobg111 comments

Very cool. My daughter has a brain shunt due to complications of spina bifida and so we’re thinking of cerebrospinal fluid on a regular basis. As a layman I guess there’s a particular cerebrospinal fluid pressure you want, and that being too high or too low causes issues? Apparently the shunt just drains into her belly so I also found that interesting. Too high and she’ll have terrible headaches and an “alien” head (sorry, don’t mean offense by that but it’s what I’m trying to avoid for my child) and I don’t know what would happen with too little, maybe someone could tell me. Probably not good I’m assuming.

I frequent some FB groups for parents of children with SB and am shocked at how large some people let their kids heads get without consulting with or complaining to a doctor.

what exactly do you mean by alien head? not offended, just curious

a physically large head, shaped similarly to a stereotypical alien head

I believe they mean enlarging of the fontanelle as is typical in those with hydrocephalus.

You can google "hydrocephalus", though I should warn the images can be a bit disturbing

At least in the US, early intervention allows doctors to recommend shunts so extreme cases of hydrocephalus can be avoided and head sizes normalize a bit as kids grow. The photos are usually from med textbooks and for teaching future doctors and usually demonstrating the extreme forms.

Let’s not say alien head. My biggest worry as a young dad was my child being called names because of his head circumference. He too was born with HC.

> Let’s not say alien head.

To be clear, I never said that and agree the term shouldn't be used at all.

I can’t post the specific child’s picture as it wasn’t posted publicly and I wouldn’t want to disrespect them, but here’s some examples of less severe hydrocephalus you’ll see often in spina bifida parenting groups [0] [1]. Often people don’t comment on ones like that as everyone wants to get along but if it’s left untreated it can eventually progress to be much more severe and cause debilitating headaches and mental retardation. Interestingly, in googling this apparently low pressure cerebrospinal fluid headaches are also a thing [2], so keeping just the right amount of cerebrospinal fluid draining is a must.

[0] https://www.hydroassoc.org/people-view/wylie-6-months/

[1] https://www.travelblog.org/Photos/253290

[2] https://www.hopkinsmedicine.org/health/conditions-and-diseas...

Too little CSF leads to the brain crushing itself under gravity. The brain is almost entirely lipids which contributes to its buoyancy in CSF. Take away too much CSF, nothing for it to float in. Too much CSF? The skull is a fixed volume (in adults) so the CSF compresses the brain tissue. In kids, the sutures or junctions between bones of the skull are still flexible. This allows for childbirth and normal rapid brain growth. It’s easy to tell if there too much fluid around the brain in a newborn because their front soft spot (Fontanelle) will bulge or become firm. A bulging fontanelle also occurs with other pathologies like meningitis.

Had a spinal leak after a spinal tap. So too little fluid. It’s a very unique pain.

Severe pain at top of head unless laying flat.

I spent an entire week laying flat on a couch with a happy cat sitting on my chest.

Initial treatment was an IV of caffeine. Then they they had to patch the hole.

>Initial treatment was an IV of caffeine.

For you, or for the cat?

HA!...thank you for that! :)

I've heard not having enough CSF is supposed to be one of the most painful headaches there are. How bad was it?

Fine if I was flat. A step below severe kidney stone otherwise. So not quite suicidal, but close.

It was only tolerable Because I could lay flat and the pain would quickly resolve. If I laid flat for three or four hours I could stand up for about five minutes before the symptoms hit again.

Not sure how strong the leak was. I suppose it could have been worse.

Interesting. Fascinating how you can go a lifetime without noticing CSF but one thing goes wrong and suddenly you can't even stand up. Good to hear you got better

I'm going through this right now. 3 trips to the ER to get a diagnosis, since it just happened spontaneously. Two months so far laying flat 20+ hours a day. I haven't been able to see a neurologist yet about it (I have an appointment scheduled soon now). For me, the headache is usually nearer the bottom of my head, but it tends to move around with different posture.

This random stranger hopes you get to see the neurologist and hopefully get it resolved soon!

Had the same also after a tap, went away after 5 months without a patch. Only things that helped was laying flat and coffein (less so than laying flat). A hard part of it was thinking it might never go away. Doctors told me to wait 6 months before trying a patch and they said it only works sometimes.

Ouch. Mine offered the next day. I waited a week before going back and getting one.

If there is too little CSF (e.g. a CSF leak), the brain 'slumps' but it doesn't "crush itself under gravity".

Too much CSF can lead to hydrocephalus, which is enlargement of the ventricles. I wouldn't describe it as CSF "compressing" the brain tissue.

I think it drains into the peritoneum but I'm not completely sure. There's a resistance on the shunt to maintain pressure. Too high will cause hydrocephalus with all sorts of complications. Too low will cause a massive headache short term, long term I'm not sure. I'm an anesthesiologist and frequently work with neurosurgeons.

They tunnel it in the subq until the abdomen then they make a tiny incision in the peritoneum and depending on the neurosurgeon may have general surgery take a look laparoscopically to make sure there’s CSF flow then use the grasper to try to toss it in the right pericolic gutter over the liver.

How do they avoid bacteria from going up a shunt to the brain?

I thought that was an interesting question as a layman. I found this, which I thought was an interesting read. [0]

[0]: https://www.sciencedirect.com/topics/medicine-and-dentistry/...

Yeah sterile technique has to be meticulous in shunt cases. There’s evidence that the number of people scrubbed into the case influences the shunt infection rate so students like me occasionally would not be permitted to scrub.

There’s generally no bacteria outside of the food tube though it certainly could happen the way you describe if there happened to be.

The technical founder of this company is developing wireless monitors for this cerebrospinal fluid pressure:

https://www.rhaeos.com/

We got to hear him speak pre-pandemic, definitely cool tech.

I know someone who could benefit significantly from this. How would he go about acquiring one?

Too little CSF leads to slit ventricle syndrome with a scary clinical image but not life threatening. Direct consequence is bilateral subdural hygromas.

I'm not a doctor and don't at all want to pretend to presume I can understand for sure what would or wouldn't be helpful. And I'm not assuming you need help--but I just thought something I know might be helpful somehow for your daughter and you.

Certain type of yoga breathing exercises (pranayama), as a result of involving contracting muscles in the body while breathing, help to pump CS fluid through its "ventricles", as well as to move lymph around. I have no idea how a shunt works, or would interact with these exercises, but as a result of the shunt intervention I'm worried these exercises may be dangerous to your daughter. At the same time, I consider they may also be helpful, by perhaps helping to regulate pressures, or assist training the body to improve its ability to do so, or encourage circulation of the CSF and lymph. Stuff like this might be something useful to explore for you. Have a good day!

This is curious. During my research into neutral networks we depended on a neuroscience result that found waves of trophic factor (i.e. the stuff used to build up neurons [i.e. learn]) travel in cascading waves from the core of the brain to the periphery. By adding a sense of location and a learning term contingent on the factor we saw learning (xor) prohibited by Minsky's result (not really prohibited because we changed the model).

The take away was that our models of ANNs benefit from considering more that just neural signaling (i.e. solution contents and fluid dynamics). Is anyone aware of modern work expanding on that front?

Was this research conducted prior to 2005?

I find the whole xor result to be pretty irrelevant, which I think is representative of the field writ large.

> Is anyone aware of modern work expanding on that front?

I am curious about this as well, as well as the general state of biological inspiration in the field. My impression was that biological analogy had largely been pretty much completely moved away from, except as someplace to take inspiration from in terms of what optimization can achieve.

Yes, 2004 but now I'm curious why you ask. ?

just because 90s-early 00s is the time period where I perceive this style of research to have been at its peak, so I was curious if my assumptions were correct

Itwas moved away from, but in recent years it's been picked up again, with growing intensity. Just look at deepmind for example, which has been publishing with neuroscientist wet labs. Two papers come to mind - one showing the "prefrontal cortex as meta reinforcement learning", and another about distributional temporal difference learning wherein specific neurons were examined.

I will also be interested to see how this develops our understanding of the relationship between inflammation and psychiatric diseases, e.g. depression

That’s pretty cool. What is interesting about your studies is you have the opportunity to apply modified fluid dynamics principles to a neural environment. You might also try an approach of allowing the network to have a refresh cycle where you add byproducts that accumulate as a result of operation and must be cleansed during a separate cycle. In this other cycle you could apply other NN operations such as GAN ect to allow the network to “dream” and create/remove various pathways. This is probably beyond the scope of your work but it seems like pretty cool opportunities for experimentation since there are probably only a few thousand people on the planet that fully understand all of these concepts.

Where AI and neural networks still have so much to learn from actual intelligence. Seems like the technical and computational problem set for this century, where a greater understanding of brains will branch out more broadly.

My own evolving hypothesis is that pressure(s) in the body is a very important "setting," as well as the effects of natural Faraday cages that occur throughout the body/brain and at different layers.

To note, I'm really curious how a no-to-low inflammatory vs. high inflammatory diet would impact this visual; likewise seeing differences between short-term and long-term, and things like before and after if switching between/cleaning up a diet or going from low inflammation to high inflammation over 1-5-10 years.

What was Minsky's result?

This [0] is probably a reasonable overview though I didn't read it thoroughly, it was on the right track.

Basically that Hopfield networks couldn't learn XOR.

The significance is that it moved us away from biologically plausible networks and precipitated back propagation which was the dominant learning paradigm in neural networks and mostly remains so.

[0] https://alan.do/minskys-and-or-theorem-a-single-perceptron-s...

The problem with XOR is that you need a line with dimension N, where N is the number of distinct inputs to the XOR, to separate true from false in the truth space.

What I demonstrated was up to 6 dimensional XOR functions being learned in Hopfield networks using Hebbian learning with an additional learning term. I ran a "wave of learning" over the network modifying that terms. In a sense it is a layer free network because no explicit later exists but in another sense the modulation provided a transient layers since the learning term caused different parts of the network to learn at different times out of sync with one another.

Could you share a link to your work? I'd love to read more about it.

It's rather dated but my dissertation is saved for posterity [0] and includes references and old Octave code.

[0] https://drive.google.com/file/d/1R4mS_4vzf5mycQez76O_mmpQDpT...

Thank you!

Not what you asked -

I (no expert in neuroscience) had come across a neuroscience article in around 2004, which I could not find again, and on which I never came across anything else either. This said that myelination happens differently in different parts of the brain, culminating everywhere by adulthood. I had found that very interesting. If someone knows more about this, I would like to hear. :-)

Re falling asleep during MRI - I was a subject in a study where for about an hour you had to lie in a machine and repeatedly indicate which of two pictures depicted a larger object. They said subjects falling asleep was a big problem! I find sleeping very difficult and even I was feeling extremely drowsy after 45 mins of lying down and establishing "elephants are bigger than mice. Houses are bigger than shoes. Trees are bigger than flowers. Cars are bigger than dogs etc etc etc"

I get an MRI once a year and I frequently fall asleep because nothing puts me to sleep like mechanical white noise. It's kind of a problem because I try to move or jerk awake, which ruins the MRI. My MRIs take forever. I feel terrible for the attendants.

I used to get a funny look when instead of a choice of music I opted for silence in the headphones during an MRI. Then I think I started to say it dismissively enough with conviction and got nonreactions.

I never understood opting for music. MRIs are so loud; the last thing I want on top of it is competing noise. How is that not sheer cacophony?

MRIs are stunningly loud indeed. The music would distract me from the noise patterns which other than being interesting to me don't keep me from napping in what to me is a cozy tunnel (whereas the music might).

I find loud screamo music the most soothing and compliments background machinery well.

I'm not sure what's the problem. I have done MRI twice in the last year, I always told I don't want music and I don't see how someone would find it weird. I don't want to damage my ears - the music was super loud, and if they turned the volume down I would probably hear the noise anyway - and what if the music sucks? And I have to lay still listening to it, now that would be torture. The white noise is quite relaxing actually - and also funny, sometimes it feels like a word repeating over and over.

How is your aircraft sleeping?

Amazing provided people shut up. I'm also 5'3" so I'm comfortable in airline seats.

I have to get an MRI roughly once a year to monitor a benign cyst and ensure it doesn't start to grow. My first experiences with MRI were okay. Certainly not pleasant but not an issue. I can't even imagine actually falling asleep inside an MRI though.

Last year I actually went the other direction. I was switched to a somewhat older MRI machine in a different building, apparently somehow "due to COVID." I thought the tunnel in the usual MRI machine was tight but this older model was even smaller. Also, due to COVID they insisted that I have on an N95 mask inside the machine, underneath the 'Magneto-like' helmet used to hold head position. I was able to complete the lengthy session but these changes were enough to trigger what I assume was some claustrophobia and it was not fun. It required maximum mental effort just to make it to the end.

I just got the call from my doc to schedule this year's scan and apparently the same requirements are still in place although they are "hoping things get back to normal soon." I mentioned how challenging last year's scan was and he suggested we just defer it until Summer. I felt conflicted about this as I really don't want to be postponing diagnostics but damn, last year was pretty awful, so I went for the reschedule.

My first MRI was in a tiny tube model. And I’m a big dude.

Never knew I was claustrophobic. I wouldn’t talk to anyone for a full day.

Used anxiety meds after that. Doesn’t bother me anymore, but haven’t been in one that small since

Hi all!

If you go to scihub and type in the name of the article you can download a pdf

https://sci-hub.41610.org/amp

Scihubs address moves, so you can type into google ‘where is scihub’ if the link doesn’t work

The name of the original article is: Coupled electrophysiological, hemodynamic, and cerebrospinal fluid oscillations in human sleep

What is the ideal sleeping position to maximize brain wash? Is it back, side, slightly inclined? Fluids theory would surely dictate how we can better align our body to encourage this. Intuition says our neck torsion must be instrumental to a good flow.

If I'm lucky, constant tossing and turning.

A few years ago, before Covid era (BC :), I did a tour of medieval castles in Europe and one observation that stuck in my memory was that the beds used by the royalty were constructed with an inclined position. I’m really curious how prevalent this was and if it was the preferred luxury style of the rich. In parts of Asia hard pillows (reeds/woven) were preferred over down.

Head up or head down?

Head up, like leaned back sitting.

In some cultures, it was traditional to use the skull of your enemy as a pillow.

I assume you would want to rotate.

I will sleep in a washing machine from now on

It would really just take sleeping on your back with head to the left, and head to the right. This approach works for sinus drainage as well.

So many people cannot/do not sleep on their back, though.

Centrifugal bed from IKEA.

Of all the worldly anxieties dominating modern existence I cannot imagine why you would want to add another one

Probably because THIS is finally the magic bullet, that not knowing it, was holding me back for years.

You don't know what effect this has. Why do you want to maximise it?

Only if they didn't read the article, which explains "Earlier studies have suggested that CSF flow and slow wave activity both help flush toxic, memory-impairing proteins from the brain." and more.

Studies may have suggested something, does not mean they have proven that same thing. Also, "toxic memory impairing proteins" is some kind of hand-wavy journalistic clickbait and not actual science.

> proven

Studies don't "prove" things.

"toxic memory impairing proteins" is literally what they are. They impair memory and when they accumulate, they are toxic. There's nothing hand-wavy about it. These words have meaning.

Hello!

Exactly! However, there has been also a lot of research that has built upon this 2019 paper -- here is a nice review article that touches on what we now know about changes in CSF and how that is coupled with neurodegenerative diseases.

https://doi.org/10.1016/j.smrv.2021.101572

Wake up in the middle of the night, shake your head violently and turn to the other side. That probably will do it.

Think of sleep as a powerwash for the sticky trash-covered movie theater floor that is your brain. When the big screen goes dark, break out the hoses and scrub away all that accumulated crud.

Sensory input goes dark, but the brain is quite active creating simulated spaces and threat response environments.

I had some crazy dream once when some unexpected home visitors in black suits suddenly started shooting at me and I did an all-out-effort to escape them, running/jumping like crazy. While I ran as fast as I could (it really felt that way), I heard some dampened noise as if somebody was screaming. Soon I somehow escaped and was suddenly waking up. That was when I realized it was me screaming while dreaming about escaping the attack. My heartbeat was over 180 as well. I am wondering what was my brain preparing me for.

Publications from the lab: https://www.lewisneurolab.org/publications

Funny that the lab provides an "open-access link" to this 2019 paper which turns out to be the opposite.

This is very cool. Should be tagged 2019 though.

Would be interesting to see if individual differences in this process is correlated with psychiatric disorders at all.

Cool video, but it would be much cooler if there was some idea of the timescale (i.e. how often do these waves occur?). I doubt that it's shown real-time; perhaps 10x or 100x?

And, I've been searching for the actual study/paper, but with no success. 'Twould be good for the BU press blurb to link to it...

Found it, I think: https://www.science.org/doi/abs/10.1126/science.aax5440

But, it isn't open-access...

Pre-print: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7309589/

Final (need $$): https://www.science.org/doi/10.1126/science.aax5440

The pre-print say the pulses occur with a frequency of ~0.05 Hz (once every 20s). So at a guess the video is sped up 20x.

Why can’t we garbage collect while conscious? (Serious question. If you simulated this washing behaviour in a conscious brain, what would go wrong and is that addressable?)

Hello!

I think we have to parse this apart a bit. There have been some recent studies that show CSF is related to movement of metabolites in the brain, however these findings are very new.

So if you theoretically wanted to induce more CSF flow while awake, I think this would be possible, but you would have to use a pretty obnoxious stimulus or use a deep breathing exercise - e.g meditation

But because we don’t know the complete reason of why we have changes in CSF flow when people are asleep, maybe there are other mechanisms going on that are acting simultaneously with sleep — e.g large fluctuations in blood flow may be necessary to induce CSF flow

Maybe the same reason you cannot VACUUM FULL a Postgres database while used in production?

This really scares the shit out of me, because I have chronic insomnia, and I am aware of the effect of sleep on not just how my brain feels - but also my vision and eye-lids (being SUPER heavy even when my mind is racing....)

I know how important sleep is to brain function, clearly...

And it causes ADDITIONAL anxiety, snowballing into more insomnia...

Fucking sucks.

What helped me the most against my insomnia is simply accepting it. I just stopped fighting to try to sleep. It didn‘t help immediately, but at least the time that I wasn‘t falling asleep was less painful. Over time (several months/years) my insomnia eventually went away.

Years later I listened to an audio book by The psychologist Viktor Frankl, that also helped me relax about the topic.

And of course jogging (without music) or something similar that essentially switches off your brain during day time.

Wow - So a few comments on what you beautifully wrote for me:

* I attempt to accept it and Breathe - (Some WIM HOF and I just received the book "BREATH" [0] <-- technically i just received via grabbing from my book-shelf after having bought it 6-months ago, without ever picking it up....

:-( --> :-))

[0] https://www.mrjamesnestor.com/

---

But running used to be my thing... in ~2002 I was running 15 miles a day at ~5 MPH...

In august 2021 I rode my 29"er 1,000 miles!

---

Exercise helps cancel out Depressopiona

did the running help with sleep? thats a lot of running!

I made an HUGE edit to my running but I lost it:

I was running 5 miles early AM, Lunch then Dinner... I did a 5 mile loop 3 times a day...

It was during the downturn, and I was at home 18 months playing games, traveling, and running... I was 25 years old, had been working as an IT manager in SV for ~6 years at that point.

(started at 17 at an Architecture firm as a drafter, and worked up from there)

I had a party that summer of ~2001/2002 (cant recall if 911 was before or not - but 75 people came to the BBQ... like 15 had jobs... (ALL devs and Tech (some jobs that dont really exist anymore (SGI Devs, MIPS procs devs, SUNSPARC/SUN devs... A lot of hardware people)

> One explanation may be that when the neurons shut off, they don’t require as much oxygen, so blood leaves the area. As the blood leaves, pressure in the brain drops, and CSF quickly flows in to maintain pressure at a safe level.

If this theory about the blood leaving the brain due to the neurons needing less of it during sleep is true, then possibly that could be a supporting argument for why being more relaxed (i.e less stressed) before sleep is helpful for better sleep. My explanation for the hypothesis in the previous sentence is that the neurons due to being more worked out from being stressed would require the oxygenated blood to replenish them and thus the CSF would not naturally be able to seep in early while in bed for sleep.

Open-access link: https://www.science.org/doi/full/10.1126/science.aax5440 Registration is required.

Oh cool, my wife’s in the lab of one of the coauthors. It seems like brain imaging is a really really small world.

I recently worked on a project involving segmenting clinical CT scans. Almost every python package for working with CTs was written by the same guy.

I had a friend who had a surgery to unblock the cerebral fluid paths. Not sure if "shunt" is a technical name for it, but that's what he said.

Also, he said he had insomnia before it. Really cool to see it. Internal plumbing of a human body.

Am I the only one for whom the video isn’t loading? On iPhone. Very frustrating.

(Of course, requesting the Desktop version does jacks***—-only the mobile page is displayed, as expected…)

It's loading for me on iphone, but I’m not seeing any blue waves? Just red waves in increasing frequency. Very confused.

Very cool. I'd be very curious see if there are similar effects from basic meditation and/or from very advanced meditation in yogic or Buddhist traditions.

Hello!

Because we see significant changes in CSF that can be induced by respiration, and meditation is known to decrease the frequency of respiration, I would expect there to be significant effects of CSF flow with meditation [1].

Beyond this, we know that with individuals who are good at meditating seem to induce theta waves in EEG [2]. Because there is a coupling between EEG/fMRI global hemodynamic signal/CSF flow, I would also expect there to be a change in CSF that may be coupled with the EEG rhythms.

1. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3895787/pdf/204... 2. https://pubmed.ncbi.nlm.nih.gov/19922249/

Indeed! And it might also be easier to test the various meditative states vs testing sleep states in such a machine...

Interesting but is there any comparison with CSF while awake? Is there more CSF movement while sleeping?

Hello!

This 2019 paper touches on CSF while awake -- CSF tends to be coupled with the respiration cycle. We know that you can also induce CSF flow by changing respiration [1].

Recent papers have shown that there can be coupling between the hemodynamic response and CSF while individuals are awake [2].

1. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3895787/pdf/204... 2. https://journals.sagepub.com/doi/abs/10.1177/0271678X2210746...

Wonder if transcendental meditation achieves something similar.

This comment indicated that meditation might have an effect.

https://news.ycombinator.com/item?id=30222210#30228358

Edit: here as well

https://news.ycombinator.com/item?id=30222210#30227672

(October 2019)

Year added above. Thanks!

Reply on news.ycombinator.com