Naked Science Forum
General Science => General Science => Topic started by: remotemass on 13/03/2022 01:04:36
-
The average volume of the brain is around 0.001234 m3. That is, 0.1234% of a cube with one meter.
If we consider cubic picometers (10e−12 m) close enough to the Plank scale in order to 100% digitize a human brain and assuming that each cubic picometer would need one bit of memory to be stored in a digital format, we would need 2 to the power of 1.234e+33 bits since we have about 1.234e+33 cubic picometers to digitize with zeros and ones. That is: 154,250,000,000,000,000 Petabytes, or about 150 quadrillions of Petabytes.
Would we need about 150 quadrillions of Petabytes to store a digital brain?
Please do check all my calculations and estimations and if you can come up with different ones please show yours.
- remotemass
-
Hi.
Please do check all my calculations and estimations
OK, I'll try some at least.
The average volume of the brain is around 0.001234 m3
That is what Google suggests as a sensible estimate.
... in order to 100% digitize a human brain..
You seem to be digititizing a 3-D picture or model of a brain here. Although I'm not sure exactly what you were trying to do.
You only allow 1 bit to be recorded, so that's enough to determine if there is some brain there or not. Recall that the volume of the brain only and not some rectangular space in which a brain is located was stated as 0.0012 m3. So all of your bits are set to 1 - there is some brain in every cubic pico-metre you are considering. I'm not sure how you get much useful information from this vast array of 1's, other than the fact the total volume is 0.0012 cubic metres, which you knew already.
If you really wanted to generate a 3-D model of a brain then you need to look at more than just 0.0012 m3 of space, you need to consider something like a rectangular space of 100 cm x 100 cm x 100cm. Then you can set the bits to 0 or 1 in each volume element of that space to show if there is some brain in that piece of space or not.
Also allowing just 1 bit of data for each pico-metre cubed is very limited information. Just enough to know there is some brain there. You wouldn't be able to determine if it was piece of a cell membrane for a neurone or just a small volume of CNS fluid. So, presumably you'd want to allocate a lot more then just 1 bit to each cubic pico-metre.
we would need 2 to the power of 1.234e+33 bits since we have about 1.234e+33 cubic picometers to digitize with zeros and ones.
No, I don't follow that. You've just said you were allocating 1 bit to each cubic pico-metre. So you need 1.234e+33 bits for 1.234e+33 cubic picometers ( 1 for each ) not 2 to the power of 1.234e+33.
However, since each bit could be 0 or 1, there are 2 to the power of 1.234e+33 total different combinations you can have, if that's what you meant.
Would we need about 150 quadrillions of Petabytes to store a digital brain?
Subject to previously mentioned issues.... ummm.... maybe. You could digitize a model (like a 3-D equivalent of a photograph) of a human brain this way. That's a long way short of making it a functional brain, if that's what you were really interested in.
There is research into trying to model the functional components, especially the inter-connections between neurones, of the brain. Such a map is called a Connectome.
https://en.wikipedia.org/wiki/Connectome
If you were hoping to model all the connections at the micro-scale for a human brain, then wikipedia offers this data:
The human cerebral cortex alone contains on the order of 1010 neurons linked by 1014 synaptic connections.
I hope that's of some use.
Best Wishes.
-
It is probably enough to record the type and location of every atom in the brain, and it's oxidation state.
- Recording the electron density of electrons around each atom, but the process of measuring the electron density destroys it, and disrupts nearby atoms too.
- But let's capture the connectome first, since every synapse and neuron is made of many atoms, and is easier to capture than the position of every atom (but its still a destructive process).
-
I'm pretty sure the uncertainty principle won't let you.