As passive_fist already covered, the strength of the magnetic field is mostly irrelevant in terms of safety. The human brain doesn’t suddenly stop functioning at 11 Tesla, nor does it get heated by the static magnetic field. Radiofrequency heating safety at high field is still an ongoing topic of discussion, but it should (presumably) not be insurmountable problem, and there are technologies in development that may reduce the required RF power substantially.
The bigger question, assuming you were being serious with your reference to uploading, is what MRI can actually measure. You touched upon this in mentioning contrast limitations. Keep in mind that MRI only produces a spatially and temporally varying signal intensity map. This signal arises from the average signal across all molecules in the nominal target voxel, and the nature of the signal depends on the design of the RF pulse sequence used and any contrast agents administered. For example, there is the Blood Oxygenation Level Dependent (BOLD) signal used for functional magnetic resonance imaging (fMRI) studies, which is designed to highlight the differences between areas of the brain with high blood oxygenation vs. low. That’s just a single, very specific type of contrast. Think about how many different sorts of contrast maps you might need to measure to be able to generate an accurate enough simulation of your brain to result in an accurate simulation of your personality and consciousness, and on what scale of spatial and temporal resolution, and look at how crude even the proposed new MRI system is in comparison. The level of MRI-like technology you would need to pull off such a feat would make tomorrow’s MRI systems look like slings and stones in the face of modern weapons technology.
The human brain doesn’t suddenly stop functioning at 11 Tesla
I don’t recall saying anything of the sort, and the safety risks were hashed out in the cousin thread. There’s still an occupational safety risk around strong, static magnetic fields.
assuming you were being serious with your reference to uploading
This is still LessWrong, right? I didn’t accidentally post this comment to RationalWiki?
The level of MRI-like technology you would need to pull off such a feat would make tomorrow’s MRI systems look like slings and stones in the face of modern weapons technology.
I have no idea what this sentence is supposed to mean. The paragraph preceding it seems to be arguing against the feasibility of neuron-level imaging, but then this analogy makes it seem possible.
Well, I’m not sure what else to make of the “hard limits” regarding the magnetic field that a human brain can “take” that you were referring to. What exactly did you mean?
There’s still an occupational safety risk around strong, static magnetic fields.
Such as? What occupational risks are so elevated at high field that they would pose a “problem” to human brain imaging?
but then this analogy makes it seem possible
Maybe it is, maybe it isn’t. I wouldn’t hold my breath if I were you, but I also can’t comment on what is and isn’t possible for technology hundreds or thousands of years into the future.
As passive_fist already covered, the strength of the magnetic field is mostly irrelevant in terms of safety. The human brain doesn’t suddenly stop functioning at 11 Tesla, nor does it get heated by the static magnetic field. Radiofrequency heating safety at high field is still an ongoing topic of discussion, but it should (presumably) not be insurmountable problem, and there are technologies in development that may reduce the required RF power substantially.
The bigger question, assuming you were being serious with your reference to uploading, is what MRI can actually measure. You touched upon this in mentioning contrast limitations. Keep in mind that MRI only produces a spatially and temporally varying signal intensity map. This signal arises from the average signal across all molecules in the nominal target voxel, and the nature of the signal depends on the design of the RF pulse sequence used and any contrast agents administered. For example, there is the Blood Oxygenation Level Dependent (BOLD) signal used for functional magnetic resonance imaging (fMRI) studies, which is designed to highlight the differences between areas of the brain with high blood oxygenation vs. low. That’s just a single, very specific type of contrast. Think about how many different sorts of contrast maps you might need to measure to be able to generate an accurate enough simulation of your brain to result in an accurate simulation of your personality and consciousness, and on what scale of spatial and temporal resolution, and look at how crude even the proposed new MRI system is in comparison. The level of MRI-like technology you would need to pull off such a feat would make tomorrow’s MRI systems look like slings and stones in the face of modern weapons technology.
I don’t recall saying anything of the sort, and the safety risks were hashed out in the cousin thread. There’s still an occupational safety risk around strong, static magnetic fields.
This is still LessWrong, right? I didn’t accidentally post this comment to RationalWiki?
I have no idea what this sentence is supposed to mean. The paragraph preceding it seems to be arguing against the feasibility of neuron-level imaging, but then this analogy makes it seem possible.
Well, I’m not sure what else to make of the “hard limits” regarding the magnetic field that a human brain can “take” that you were referring to. What exactly did you mean?
Such as? What occupational risks are so elevated at high field that they would pose a “problem” to human brain imaging?
Maybe it is, maybe it isn’t. I wouldn’t hold my breath if I were you, but I also can’t comment on what is and isn’t possible for technology hundreds or thousands of years into the future.