I'm really sorry to hear how many misses you've had - it sounds like you've had a lot of practitioners that are either not good at communicating, or just not good in general. I'm not going to change your mind here obviously since you've had a ton of bad experiences, but frankly doing "a little bit of sober research" isn't accessible/an awesome idea for most, and I am not convinced tools like ChatGTP on their own will ever really do any better than the extremely messy and dysfunctional system we have in place. These tools need to be integrated into practice, and I think they will be, regardless of any resistance by older physicians. Unfortunately there's a lot of nuance in medicine..again I'm not trying to change your mind, but just to address a few of the situations you brought up:

1. Your pediatrician should have said something along the lines of: "If your child's condition worsens, take them to the ER". The fact that you gave chatgpt a copy of the medical record (?) from the encounter and it told you they were dehydrated isn't really a diagnosis. You know the details of this situation better than I of course, but the bar for who benefits from IV fluids is pretty low unless you're fluid overloaded, so the fact that they were given fluids at the ER doesn't really say much to me.

2. RE diverticulitis, obviously poor communication here, but did that same provider order the CT that ended up confirming it? If so, I would call this poor communication instead of a misdiagnosis. If they told you to go home without doing anything, this would be a misdiagnosis, but it's really common for a diagnosis not to be reached until labs/imaging data comes in.

3. RE gallbladder, if it wasn't the standard of practice at the time you went in, unfortunately, that's likely what everyone is going to stick to, even if it's right before new standards come out. New research is constantly coming out, but even if you had ChatGTP as a doctor I don't really think it's appropriate for it to guess what the new guidelines are going to say before they're made.

4. There's really no reason why your primary care (?) physician should see nothing wrong, and then a surgeon should tell you the complete opposite. If it was confirmed later by MRI this was a missed diagnosis from the earlier opinion.

Other than use it for targeted marketing, make broad/vague medical recommendations (a push notification to "get more sleep" or "don't forget to take breaks!" or "vegetables and exercise help prevent dementia!"), or advise you to go see a doctor ("you seem/are depressed!")? I don't see how this data could be applied medically, just used for marketing.

I'll agree that Reddit is a massive echo chamber, a pretty toxic environment overall, and will downvote you into oblivion if you post something that isn't left-leaning. Great for cute videos of animals, not so great for civil discussion.

I'm assuming you're talking about this study: https://aip.scitation.org/doi/...

It does make the points you suggest above (and the science seems solid enough), although I think it's important to note that this research was done with mannequins and olive oil droplets, and these types of simulation experiments always need to be taken with a grain of salt when making conclusions about health and medicine.

Pretty cool design, but doesn't seem to be able to go below the cortex. The predictions made about limb position were very likely done with data from the primary motor cortex or premotor area, both very superficial and easy to access.

A lot of promise with this to be sure, but all of the disorders he listed as potentially being targets for this (with the exception of some seizures, strokes, paralysis and tissue atrophy/loss) are whole-brain disorders (i.e. focal stimulation will only do so much), or require deep brain access. The reason deep brain stimulation is popular as a treatment is that it can penetrate deep into the brain, and affect structures such as the subthalamic nucleus (in the case of Parkinson's Disease).

Also, the more complicated the disorder, the less we know about it why it's caused and the more complicated it is to treat. I'm sure you could access dorsolateral prefrontal cortex with a neuralink, but....whether you would be able to specifically treat someone's psychiatric disorder (i.e. anxiety or depression) without causing psychiatric side effects (mania or psychosis) is another story and will need another few decades of research.

I've been saying this for a while (just to my friends) - I think algorithms make these mistakes because there are no consequences for wrong answers. It is in the best interest of humans and living creatures to guess correctly, because we don't want to die. For example, tigers have "meaning" to us because they could kill us, food has "meaning" to us because we die without it, etc. etc. Nothing has "meaning" to predictive algorithms, which I think is a interesting and fundamental challenge to predictive modeling and machine learning in general.

Hmm....pretty complicated to detect things like depression with an EEG cap. Stress, may be easier...Still, I'm skeptical that this "huge boost in profits" doesn't just come from scaring people into thinking that their thoughts are being read.

I now have a bit more of an understanding of why anesthesiology is such an art - not enough of this stuff and you may paralyze someone without knocking them unconscious for a surgery. Too much, and you dangerously dampen the parts of the brainstem that control breathing and cardiac rhythms. I'm curious though as to why consciousness goes before breathing does...if these drugs are given intravenously, wouldn't they diffuse to all parts of the brain equally, and cause just as much consciousness loss as breathing loss?

http://press.rsna.org/timssnet... The official press release details the actual brain area in which this metabolism change was studied. The anterior cingulate cortex is involved in pretty much everything (attention, task-switching, self-monitoring, etc), so not super surprising but pretty cool nonetheless. Also not really "controversial", as any substantial behavioral change will necessarily produce corresponding changes in brain activity...Good to see that it can be reversed though.

To one day direct them down a spine and into some new prosthetic legs.

From this article: “No neuroscience experience is required: talent and drive matter far more,” the company says on the site. “We expect most of our team to come from other areas and industries.” Eh...sounds like the bottleneck here will not be engineering machines, but getting them to produce meaningful changes in the brain at a neurological and psychological level. It doesn't matter how sophisticated your implants are if you don't have a way of reliably changing brain activity in a way that benefits a disease, which involves an incredibly sophisticated understanding of pathology at a molecular, cellular, and systems level, etc. etc. It's not really "plug and play"

Are you a veteran? If you are, I'm sure you're aware that there are unbelievable hurdles to re-integrating with civilian society, many of which can be helped with elevated community support and a steady job. True, there are many people in this country that work very hard, and you need to constantly retrain yourself and be creative to code, but vets that are reintegrating are playing by many different rules (because of serving our country!) that civilians are not: I can't see how any initiative providing specific help to this group to find jobs and meaning in the 21st century would be a waste.

Certainly Mr. (Mrs.) tech billionaire, I share your enthusiasm for integrating the human mind with a machine and believe it is indeed within our reach now that innovators such as yourself are stepping up to the challenge. A one-time donation of $15,000,000 to my lab to pursue our groundbreaking research that we were already doing anyway will ensure that humanity will praise your name forever, parades will be held in your honor, and all other entrepreneurs will gaze at you longingly at tech events.

This is really interesting, thanks!

Until we know more about how these algorithms make predictions, it'll be tricky integrating them into medicine: "I think you have a melanocytic lesion because I graduated medical school and have trained in dermatology for six years" still carries more weight than "Our highly accurate algorithm said you scored in a particular way on the 21 dimensions that we can't quite correlate to anything tangible, but it suggests you need this invasive surgery".

Ah, this is so excellent. Thanks for sharing. I'm a neuroscience/medicine student and this is what keeps me awake at night...We'll get there one day!