Technology and pathology?

[doctor90210]

Technology has revolutionized medicine. Obvious examples include rapid identification of novel drug candidates by pharma companies and new imaging technologies that detail our internal organs with astounding precision. It is in our nature as human beings to continually evolve and refine the tools that help us evade our inevitable death. It's undeniable that advancements in medical technology have improved the quality and quantity of human life. It's exciting, isn't it? I get a hard-on every time I think about it. How will such advancements impact pathology? Will computer technology in the future be sophisticated enough to scan slides for suspicious patterns of disease? Is it ridiculous to think I can someday put my surgical specimens into the new BD Biosciences Surgpathomatic Z3000 to generate N slides? Don't get me wrong... I'm not saying that hard-working pathologists should be replaced by machines. Instead, I wonder whether technology can be harnessed to make our work more efficient. Could software programs running in our computerized microscopes say: "Look here... this lymph node is totally jacked.", or "Hey... this erb-B2 probe is seriously lighting up.", or "Check this out... don't you think this looks like acid-fast bacilli?". A reasonable analogy might be the use of sophisticated computer models to predict weather patterns by meterologists. Pathologists might similarly benefit from cutting edge technology to quickly and efficiently diagnose disease. How might technological advancements change pathology over the next 25 years from what it is today?

---

Members don't see this ad.

 

[deleted113029]

Technology has revolutionized medicine. Obvious examples include rapid identification of novel drug candidates by pharma companies and new imaging technologies that detail our internal organs with astounding precision. It is in our nature as human beings to continually evolve and refine the tools that help us evade our inevitable death. It's undeniable that advancements in medical technology have improved the quality and quantity of human life. It's exciting, isn't it? I get a hard-on every time I think about it. How will such advancements impact pathology? Will computer technology in the future be sophisticated enough to scan slides for suspicious patterns of disease? Is it ridiculous to think I can someday put my surgical specimens into the new BD Biosciences Surgpathomatic Z3000 to generate N slides? Don't get me wrong... I'm not saying that hard-working pathologists should be replaced by machines. Instead, I wonder whether technology can be harnessed to make our work more efficient. Could software programs running in our computerized microscopes say: "Look here... this lymph node is totally jacked.", or "Hey... this erb-B2 probe is seriously lighting up.", or "Check this out... don't you think this looks like acid-fast bacilli?". A reasonable analogy might be the use of sophisticated computer models to predict weather patterns by meterologists. Pathologists might similarly benefit from cutting edge technology to quickly and efficiently diagnose disease. How might technological advancements change pathology over the next 25 years from what it is today?

i think you need a hobby or something.

 

[djmd]

i think you need a hobby or something.

I agree.
But to answer your question, pathology will maybe get to some of what you are talking about in 30-50 years...
(maybe the computerized scanning could be ready in 10+)

 

[yaah]

I think it has been tried for screening pap smears and the results have been not that great (I think). But even then, it is a screening tool, and it would have to be directed to look for certain things only, and thus it would only be good for these instances. I suppose it's a theoretical issue for the future, but it is going to take a while. Currently the human eye works better.

 

[djmd]

I think it has been tried for screening pap smears and the results have been not that great (I think). But even then, it is a screening tool, and it would have to be directed to look for certain things only, and thus it would only be good for these instances. I suppose it's a theoretical issue for the future, but it is going to take a while. Currently the human eye works better.

Automated scanners are being used for PAPS and for HER2 IHC.
But in both cases they are single use approaches, and the still need people to make sure they are not screwing up.

 

[LADoc00]

Im going to shock people but I think we have already begun butting up against the limits of what technology can accomplish due to cost.

We are already seeing that in molecular diagnostics. Things like prognostic indicators are decided by declining reimbursements and ability to pay.

Yes we can make HUGE leaps in how pathology is done in terms of automation. It just isnt cost effective and may never be. Telepathology, next gen image analysis are great but how do you justify having a 100K machine+technician when just slapping a pathologist there is cheaper and getting cheaper.

Guess what?? Moore's law is just about to be proved wrong. We are seeing a real limit to microelectronics barring a major revolution in nanotechnology.

I dont want to burst your bubble...Oh YES I do, considered it burst.

 

[Anna Plastic]

I think it has been tried for screening pap smears and the results have been not that great (I think). But even then, it is a screening tool, and it would have to be directed to look for certain things only, and thus it would only be good for these instances. I suppose it's a theoretical issue for the future, but it is going to take a while. Currently the human eye works better.

Totally agree with yaah, here. And I think he raises an important illustrative point: pap smears as a screening tool for cervical cancer. Certainly our collective goal in healthcare is preventing or treating disease, and we have finite resources to dedicate to this. If we want to reduce cervical cancer, are we better served in investing in more and more sophisticated imaging systems, or in providing pap smears to those who don't have access to health care? The latter would have a greater impact on morbidity and mortality from cervical cancer to be sure.

I have hopes for technology providing meaningful, useful advances to pathology, but I think its also important to keep our eye on the ball and think carefully about where we are spending health care dollars.

 

[yaah]

Has anyone actually used telepathology or virtual slides? It takes like 25 times longer to load the slide up and scan along, etc, than it does to slap a cheap piece of glass with a cheap stain under a microscope. And the microscope provides a better quality view.

Molecular stuff is an ADJUNCT to current diagnostic and treatment algorithms. It doesn't replace anything. The closest it seems to be coming in my book is the bone marrows for CML where they go directly to molecular to look for minimal residual disease. But that's because CML has a common detectable molecular abnormality. Most lesions do NOT. Molecular tests (currently and for the foreseeable future) require you to know what you are looking for. You can have a fancy molecular assay for synovial sarcoma but you don't need it if you can diagnose it with H&E +/- a couple of immunostains. Where these tests will likely have relevance is after diagnosis, to determine sensitivity to certain drugs or treatments. Maybe in 200 years you can have a test that fits in a little machine in a surgeon's office where you shove the tumor in there and it tells you what to give the patient to cure it, but not now!

 

[LADoc00]

The ONLY department in the world that successfully gotten telepathology to "work" is Univ Toronto as far I as I know. This is after the US military spent tens of millions (maybe hundreds) on Iron Mountain and simply walked away from it.

 

[mlw03]

Totally agree with yaah, here. And I think he raises an important illustrative point: pap smears as a screening tool for cervical cancer. Certainly our collective goal in healthcare is preventing or treating disease, and we have finite resources to dedicate to this. If we want to reduce cervical cancer, are we better served in investing in more and more sophisticated imaging systems, or in providing pap smears to those who don't have access to health care? The latter would have a greater impact on morbidity and mortality from cervical cancer to be sure.

I have hopes for technology providing meaningful, useful advances to pathology, but I think its also important to keep our eye on the ball and think carefully about where we are spending health care dollars.

couldn't agree more about the issue of finite resources and the healthcare community as a whole having to decide how best to spend them. in terms of cervical cancer, perhaps even better than screening pap smears for indigent patients would be getting Gardasil out to all these women.

this actually brings up a good point - how much money and effort is it worth to try and further develop pathology's diagnistic abilities? ie, is there a major disease that pathology is bad at diagnosing? what i think would be great would be some technological advance that allows pathologists to definitively say if a particular pre-cancerous lesion will or will not develop into invasive cancer. is there some molecular marker that differs in tubular adenomas that do develop on to adenocarcinoma versus those that do not? or a marker that could tell us the few basal cell carcinomas will spread? i'm guess i'm thinking of a prognistic indicator, but something truly definitive, not just something that makes it more likely. maybe some of this stuff exists and i just don't know it yet, but that's a technology that i think could really benefit patients.

 

[yaah]

There is a lot of research going into prognostics these days, i.e. markers that can predict which tumors will metastasize or behave aggressively or whatever. It's similar to using markers to predict treatment response (her2neu, etc). But if you can do an immunostain that tells you that the colon cancer you just diagnosed is more likely to spread than the average tumor, does that change management more significantly than staging the patient and doing a full standard workup? I dunno. It's all theoretical right now. But it will probably become standard on more tumors than breast cancers and GISTs to do certain stains for treatment management purposes.

 

[mlw03]

But if you can do an immunostain that tells you that the colon cancer you just diagnosed is more likely to spread than the average tumor, does that change management more significantly than staging the patient and doing a full standard workup? I dunno. It's all theoretical right now. But it will probably become standard on more tumors than breast cancers and GISTs to do certain stains for treatment management purposes.

i think the standard treatment for colon CA is so aggressive right now because we can't tell the surgeon, "yeah, that tumor you diagnosed isn't going to spread beyond the muscularis - i'm sure because the ___ stain was negative". if we could do that with the same speed a frozen section can be done, then maybe an ascending colon tumor could just get a segmental resection instead of a R hemicolectomy. i gotta think, like yaah suggested, that this is being researched out there for a lot of other tumors. i don't think we'll ever find a marker that's 100% like i hoped in my previous post, because so little in medicine is 100%.

 

[docbiohazard]

I mixed my posts a little last night, but I also am hopeful that we'll start to see more directed therapy and pharmacogenomics, so we can say "this tumor will respond best to this chemotherapy regimen, but be resistent to this alternative regimen" and hopefully things such as picking out the best drugs to achieve therapeutic effect in certain patients, or correlating side effect profiles with individual genetic markers.

TGIF
BH

 

[pathomatic]

I mixed my posts a little last night, but I also am hopeful that we'll start to see more directed therapy and pharmacogenomics, so we can say "this tumor will respond best to this chemotherapy regimen, but be resistent to this alternative regimen" and hopefully things such as picking out the best drugs to achieve therapeutic effect in certain patients, or correlating side effect profiles with individual genetic markers.

TGIF
BH

completely agree, and they are already doing it with SNP and micro-satellites and actually micro-RNA profiles these days. I am sure it will take some time before insurance companies start paying out for every person to have their signature SNP profile done. But, as medicine is moving more and more towards establishment of preventive and predictive measures, another step in that direction is to be able to predict disease outcomes and treatment outcomes.