Informatics Live Blog at USCAP 2009

13 03 2009

Well, it’s not live at this point, but I followed the 2-hour session closely with the trusty laptop; here is a re-cap:

5:30 – President of CAP, Dr. Schwartz, introduced by Dr. Monroe. Speech begins with hulu video from Fox about DTV switch – . Follows up with mention of the inevitable trajectory towards digitalization pathology. Preaching to

Dr. Schwartz - President of CAP

Dr. Schwartz - President of CAP

 the choir. Mentions the uber-populatrity of facebook. Yea, pathologists talk about facebook, so what? Then he polls audience: “Who has iPhones?” (about 50%… sheep).  Shows a ppt slide of a virtual slide on an iphone screen, asks for diagnosis… crowd dumbfounded- histiocytes? Granular cell tumor? Hmm, maybe that tech isn’t here yet… lol.

5:52 – Brings up the time-honored arguments for digital pathology: consultation for underserved areas around the world, also, need to keep up with the Radiology Jones’s; point, Dr. Schwartz.  Canada is adopting WSI because of pathologist shortage… wait does this mean my job will be obsolete/outsourced?

5:55 – The emergence of “-omics”.  Pathology needs to keep pace; EMR’s are imminent, we should integrate our PACS with those. Reminder that despite all the technological advances, in the end, there’s always room for the trusty microscope. I like it. Closes with: “when it comes to the future, there are three kinds of people: those who let it happen, those who make it happen, and those who wonder if it happened.” Viva la revolution!

6:05 – Next up, Dr. Kaplan, Mayo Clinic. The state of digital pathology today. Digital slide boxes are going mainstream. One first year resident at Mayo said that Mayo was the first place they’d used an actual microscope! What!? Far cry from Smallbany. Well, who cares about WSI? Well, WSI is superior because they are persistent, they don’t fade or break, and you don’t have to FedEx them. They are just the awesome.  PACS integration with EMR’s is the total package. Shows example of what they developed at Mayo: Q-Reads 5.0. Anecdotally, clinicians LOVE it. Moves on to image analysis. Who wants to count nuclei? (I don’t). Talks about validation of digital image analysis. Apparently, image analysis was more effective in certain situations, after all…

6:15 – Moves on to digital pathology at Mayo clinic: referrals and consults from anywhere in the country! Demographic ‘issues’ about maintaining image servers, accessioning, etc…  Says that basically any barrier makes it difficult for ppl to justify not sending it via FedEx. Is it a “niche” technology? It shouldn’t be though, because there is money to be saved. New frontiers: breast Her-2, eventually full adoption. Raises question of “out-sourcing” said he won’t talk about it… Sketch.

6:20 – Medicine 2.0… catchy. Is there a place for crowd-sourcing in medicine? Yes.  There’s a Mayo clinic youtube, facebook, blog and twitter!  Why is this important? gets a million hits per month; 10% of “e-patients” become actual patients… so it

Keith Kaplan, MD - Mayo Clinic

Keith Kaplan, MD - Mayo Clinic

 is important to pay attention to it. There are medical specific photo sharing sites mentioned as well as well as Youtube pathology lectures (one of his lectures has already been downloaded over 1,600 times!). Talks about USCAP’s move towards online posters (they’re so green!), how when he was a resident he had to bring glass slides to get in, crowd laughs.  Mentions that in the future that “small screen” technology will be big. Mentions! (go there).  Shows the HALL 2000 Venn diagrams…

6:25 – Next speaker:  Mahul Amin, world-renown GU pathologist- will be talking about the impact of informatics for us. Starts with Obama’s commitment to EMR’s. I guess it’s a start. Gives  a short history of prostate cancer advances. Haha. Ppt opens a web browser, makes bloop noise- IE popup successfully blocked! technology FAIL.  Haha does it AGAIN! Love when tech talks crash computers… Ok back to the topic. Who should receive wait-ful watching? Answer: personalized medicine! You can help delivery to the patient in a complete and convenient way using technology…

6:35 – Informatics in the next 5 years. Integration of traditional histopathology and gene expression arrays. Using these in conjunction we can tell who needs to be treated how… the integration of these determining factors will guide specific treatment, and not whether or not which to treat but what treatment to get (ie Gleevac). Additionally, you can take the basic pathology report and extract out of it survival percentages, or include the proteomic and IHC profiles IN the report. This saves everyone time and is particularly useful to the clinician in delivering care to the patient. More on molecular testing. Using gene signatures can stratify people in to risk factors group or who gets Tamoxifen or who is at genetic risk (in her family and her ovaries!). 

6:45 –  Concludes on rate limiting factors. First up QC, we can sing all the praises and throw all our trust in the computers, but these modalities need some serious QC and standardization. Plus there is a resistance by pathologists themselves (especially old ones!). The key, the driver, is the advent of the molecular era… 10 years from now we (pathologists) will be integrated with radiology and oncology (sweet!) 

6:46 – Last speaker- Ajit Singh, CEO BioImagene. Let’s learn from radiology. Either you need to automate or redefine. (Wow this is a sexy powerpoint… I guess I shouldn’t expect any less from the head of a bio-tech company… lol.) Technology leads to innovation leads to sustaining or disruptive technology.  Example: home pregnancy tests = FAIL as far as sensitivity, but are now the standard. (I like where he is going.) Pregnancy tests are so cheap tho, that women adopted them anyway, and the accuracy isn’t that off… so this is what we are talking about in pathology.  This disruptive technology happened in radiology in the past 5 years. This is when all the software came out. Allowing us to do things we just couldn’t do before. Now makes example of Nano in India… 2000 bucks and its better than the rest of the cars, but it was pitched as an upgraded motorcycle… clever. Really clever. The cost is SOO low, that it makes more sense to pitch up from the technology below it.  So there are two attributes of disruptive technology… interrupted by his cellphone going off- Fur Elise- apologies… Type 1 (improving existing tech)- So, adding PACS increases productivity by 20%- you can take more cases now (up to double) and turn-around time decreases by a factor of 10. Even screening is up. Type 2 (redefining) Improvements: What innovations do we have: 3D visualization, quant analysis, fusion of antomy and physiology, contextual acces to anatomy atlas at POC, or to similar cases (like, or expert opinion at POC.

6:58 – goes into more of the specifics of a typical PACS. Wow. Mitosis, gland detection… holy crap! Her-2 is now FDA approved. Prostate, Ki67, FISH, EGFR for lung and colon, etc… For each diagnostic test (ie a stain)  we need an algorithm (which is better interpreted by computers)…. So we can take the image, but what does it mean for the patient… (see above)  like drug targeting. Another example of personalized medicine:  1. Collect large databases, 2. Create personalized knowledge models, 3. apply these to clinical practice. So what do you do: you can lead change, follow change, or define change (I notice a common war-cry theme).

7:05 – the summary: work-flow approach from end to end. You can’t half-ass this. You need a single modality and “from information to diagnostic intelligence” It’s all about the cock-pit!

7:07 – 20 minute Q&A session.
1) how do we get paid? A: not sure, but efficiency is better, right?
2) how do you reconcile the cost of R&D? A: Stay the course. Things might cost more in the short term, but pay off in the end. Mentions “virtual trials” and “insurgents” and “how the FDA needs to sit and listen like a pupil”, hmm…
3) How do address DICOM standards when path is way more data intensive? A: there are three ways to approach this: 1 [I forget], 2 the Microsoft way, 3 the wiki way. The idea we want to support is an open community of sharing of information. Answer back question: I tired that in radiology, the request to be flexible was a deal-breaker. A: I saw open it up. We can do both and let the best man win… damn, he’s smooth!
4) Q: nobody owns wiki, where’s the profit model? A: well digital pathology has more use than slide sharing. Thank you though, my investors are in the audience… lol. We need laissez faire!
5) Q: question about contrast manipulation how it would affect the image. A: well actually digital image manipulation could improve image processing. Think about HJ&E in your own lab, It can definitely be inconsistent. Digital imaging can help cut down on artifact.
6) Q: for Dr. Amin: there are specialty laboratories offering these molecular tests… if pathology doesn’t adapt we are fucked!  A: Yes!  But this is what will provide the drive3. BUT it can also be very threatening… out-say ource-say ing-lay!! The experts are pissed because they are worried about losing business (maybe he means local experts??). In the end either adopt it our you are screwed… yikes!




Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: