Will AI Replace Physicians? - The Medical Futurist
HTML-код
- Опубликовано: 12 сен 2024
- If you've been following The Medical Futurist’s platforms, you've likely heard me discuss the million dollar question before:
Will AI replace physicians?
The rapid integration of AI across industries has sparked widespread concern about job displacement, and let’s be frank healthcare is no exception. Many wonder if AI will eventually render physicians jobs obsolete.
I firmly believe AI won't replace doctors, and in this video, I’ll outline five compelling reasons why.
I agree in the short term, but I don't agree on longer time scales. We think of AI as an "us" or "them" problem, but we will continue to merge with our technology, so in the shorter term, augmented human doctors will become the norm.
Okay, but in the long run, will artificial intelligence replace doctors and surgeons or will it be an assistant to them?
@@user-mf6tx9sg3k That depends on how you define the duration of the "long run" of the future. In the 2030s, based on Ray Kurzweil's latest book, The Singularity is Nearer," and his predictions, nanobots will be created and deployed for many internal medical procedures. Procedures that doctors can not do. Doctors will remain the preferred human interface with patients in the future.
Radiology: AI trained 3d Segmentation of the paperthin bones within the cranium will really advance the 3d printed outputs for surgical simulation. Wondering if I can use my high-fidelity 3d models to train my own AI, resulting in faster outputs for patient specific simulations. Ive seen the major segmentation service providers out there - timely, expensive and most importantly, just not good enough (rhinology, opthamology, otology, etc)
I would like to preface this comment, noting that I am a Registered Nurse of 8 years, and I am in the midst of completing a Nurse Practitioner degree program. I am also facinated and intrigued by AI, and also terrified by it. I am also a hobbyist programmer, and have been working in solo game dev on my free time (which I have recently stopped focusing on). I am incredibly worried and also incredibly facinated with AI. I applaud your optimistic views, but I disagree with every single point.
Point number 1: Hospital corporations will discard empathy over profits in a heartbeat. And I am sure they will have reasons why that is OK.
Point number 2: You are wrong on human beings expertise and understanding, in point number 2. AI is predicted to be smarter than the entire human species combined. I think you are limiting your view on how intellegent AI can be, maybe based on current AI usage. Your view on AI ability to understand abstract data from the diversity of the human species is flawed. AI could become incredibly smarter than any one physician. With superintelegence, AI can also become creative on problem solving. Frankly, AI creativity sounds likely given the current exponential increases in AI.
Point number 3 doesnt seems to have any coherent differences from point number 2. AI will be in itself an amalgamation of all of the medical professionals. Maybe AI will require a significantly smaller number of human medical professionals to review its decisions... although I imagine this number will shrink fast.
Point number 4, I think is simply a shortsighted view. Its hard to understand how intellegent AI is becoming, or will become.
Point number 5: Yes, right now technologists working with AI produce improved results. This current AI is the lowest level of intellegence AI that we currently have. AI is expected to increase on an exponential curve. There has been discussion of AI comleting research on its own. AI is in its infancy, and is expected to increase its ability 5 fold, 10fold, or even more in the next 5 years.
While you may say that my viewpoints are contradictory to my career path, I would have to agree. I truly hope that I am not waisting my time working towards being a provider. I have worked so so hard, and it is very scary to be replaced by an intellegent machine. I still value myself, and my human intuition, but I do believe that AI will easily be able to replace my human intuition before long. That makes me very sad, but I will not give up because of this.
I mean, even nursing, a very phyisically demanding job... it is not safe from being taken over by AI. There is practically an arms race to create humanoid robots happening right now! Do not underestimate hospital corporation's focus on greed. THEY DO NOT CARE, profits above all else.
I saw many machine learning/deep Learning/AI researches for almost 10 years. Every researches face the same problem, that is lacked of big data.
Even simple medical research struggles to get enough sample size. The only thing that seems possible is help reading CT or MRI. Still AI can’t do a radio-intervention or radiotherapy, that only radiologist can do.
AI is as dumb as it will ever be right now. Im pretty sure AI will be able to perform those tasks in the next couple of years pretty easily
Hi sir,
Can you pls suggest me a project topic related to eeg
Current eeg analysis and statistics
Could you create a video on how medical professionals can prepare themselves to use AI?
Interesting idea! What do you mean by comparing themselves to AI? In performance?
@@Medicalfuturist thanks for the reply. I mean how Medical doctors and AI can work together. What do doctors need to do that they are not doing now
AI can definitely replace physicians. Physicians are trained to make a diagnosis based on symptoms and test results. There is no data used by doctor that is not based on the patients medical information.
3-5 years from now, if you go to the hospital/clinic, triaging will be done through computer/ triaging by nurse, lab tests, and diagnosis will be done by AI. There is no need of doctors/physicians.
And what about surgical procedures
Tur baperor murtu 😂
There's already an algorithm that out performs 99% of radiologist, because it's been trained on hundreds of thousands scans and has lead to a whole new branch of diagnosis. That's something that's physically not possible for human to do, due to time constraints. It's just a matter of time.
And yet radiology prevails. medicalfuturist.com/the-future-of-radiology-and-ai/
@@Medicalfuturist wouldn’t they add the more couplex stuff to A.I. too?
I believe you are 100% right. I want to say, sadly; but I cant even do that. It is sad, but potentially promising?
What about pathology?@@Medicalfuturist
Chatgpt already can outperform doctors on diagnosis I really do not see justification for a 500k specialist when a nurse with health metric data can get better diagnosises from a chatbot.
Not true. Might become true in a few years for very specific tasks.
@@Medicalfuturist how is it not true. Something like perplexity can combine an LLM with search of volumes of medical information, we’re talking parsing thousands of pages in under a minute.
LLMs excel at sorting through provided data.
You’re telling me a doctor can do a better job by thinking of stuff off the top of his head? Yes as we know humans are definitely not error prone and have infinite stamina
Honestly the only barriers right now are integration, regulation, and maintenance of the status quo, not abilities.
We already have studies confirming this, the medical industry is still using fax machines, 100% an integration issue.
@@arnavprakash7991 I agree with you. Its scary to me too, because I am a medical professional. I am even in school advancing my career.
First comment 🎉
First Like 🎉
You should delete that Covid “conspiracy theories” video you did 4 years ago. It didn’t age well and it ruins your credibility. I can’t take you seriously.
Chill brother. Still no scientific consensus.
@@Medicalfuturist not according to your video…lol
I guess that’s my point
No worries either way, just saying.