Oh, growing up in West Virginia, playing in the woods or fields, there were always ticks in the summer. Sometimes there might be three or four crawling on me when I got back home.
Upstate New York is wonderful, and I am sure that once you have finished something in you will really enjoy it.
I am growing really frustrated with the tech industry’s relentless push to incorporate AI into everything, even when it is not up to the task. Not only can it potentially compromise patient care, it may also represent a security risk to the EMR. I do see it eventually having a valuable role to play in medicine, but right now it is just not ready for prime time in most fields (with screening Pap smears being an interesting exception to the rule).
The Big Tech companies have definitely overplayed their hand and you can see the backlash growing daily
Pap smear AI screening assist may be the exception that proves the rule: There are only a few systems, and they are end-to-end evaluated and cleared by the FDA for diagnostic use. My understanding is the process is so strict that changing even components of the computer would run afoul of the regulatory approval
When I see us move towards strict validation like that I might be willing to entertain broader use of AI in veterinary medical imaging!
Not only is the regulation of the automated Pap smear screening systems rigorous, they can only be used to sign out normal Paps, Any abnormality detected by the screening system gets the Pap smear flagged to be reviewed by a cytotechnologist or pathologist. And I believe right now the same is true of the automated systems being developed for radiology: they only flag suspicious findings, an actual radiologist makes the final interpretation of what the flagged anomaly means.
That's a far cry from what the AI companies are trying to sell tot he general public! We're still a long way from an AI system replacing actual physicians (either human or veterinary). And I think one of the biggest stumbling blocks will be legal: Who will be legally liable when the system makes a misdiagnosis (which is inevitable, as medicine is not perfect)?
Oh, growing up in West Virginia, playing in the woods or fields, there were always ticks in the summer. Sometimes there might be three or four crawling on me when I got back home.
Yuck! I try to extend compassion to “all creatures great and small,” but I find it challenging with disease-ridden ticks…
Upstate New York is wonderful, and I am sure that once you have finished something in you will really enjoy it.
I am growing really frustrated with the tech industry’s relentless push to incorporate AI into everything, even when it is not up to the task. Not only can it potentially compromise patient care, it may also represent a security risk to the EMR. I do see it eventually having a valuable role to play in medicine, but right now it is just not ready for prime time in most fields (with screening Pap smears being an interesting exception to the rule).
The Big Tech companies have definitely overplayed their hand and you can see the backlash growing daily
Pap smear AI screening assist may be the exception that proves the rule: There are only a few systems, and they are end-to-end evaluated and cleared by the FDA for diagnostic use. My understanding is the process is so strict that changing even components of the computer would run afoul of the regulatory approval
When I see us move towards strict validation like that I might be willing to entertain broader use of AI in veterinary medical imaging!
Not only is the regulation of the automated Pap smear screening systems rigorous, they can only be used to sign out normal Paps, Any abnormality detected by the screening system gets the Pap smear flagged to be reviewed by a cytotechnologist or pathologist. And I believe right now the same is true of the automated systems being developed for radiology: they only flag suspicious findings, an actual radiologist makes the final interpretation of what the flagged anomaly means.
That's a far cry from what the AI companies are trying to sell tot he general public! We're still a long way from an AI system replacing actual physicians (either human or veterinary). And I think one of the biggest stumbling blocks will be legal: Who will be legally liable when the system makes a misdiagnosis (which is inevitable, as medicine is not perfect)?