Why are Automation & Innovation Critical to the Histology Lab?

I hope we can have a little bit of a discussion about automation and innovation, but what I want everybody to do right now is just look at that first slide that we have here from Leica Biosystems, Vision 24. That's exactly why we're going to talk about automation and innovation. This is what our patients have an expectation of us in the anatomic pathology lab. Within 24 hours they want results, they expect results, and as you know it's all those little boxes up there. They want some very detailed data and information from us. So, this is why I've been involved in and have been trying to make sense out of what kind of automation works for us in the histology lab and those other innovative tools, which most of you know as LEAN and Six Sigma, and they're even expanding past that. Let's get started with some of the other slides and see where we go.

Our learning objective today is to identify opportunities for automation. All of you should be well aware that’s some difficulty for us in the pathology lab. We’re a very manual multi-step, multi-critical task process. There are opportunities. We want to evaluate automation and innovation solutions. Again, there are multiple opportunities and multiple solutions here. This is why it's so difficult for us, and our third learning objective is to discuss improvement processes, because automation and some of these tools that have been around for several years will not help or fix all of our issues and create a better patient safety environment. It's a discussion as I’ll call it.

Histo-technology today

So our first one is histotechnology today, and there's even been a couple of articles out recently. We are a 150-year-old process. What we're doing today in the histology laboratory, even with all the new instrumentation and everything else, that process is over 150 years old. So we have a conflict of need and to our current processes. This is my own little timeline that I have put together for my time in the lab. When I started in the clinical lab we were basically doing a diagnosis. That was it. As you can see, we had H&E, some special stains. We had diagnosis, then we moved forward to the differential diagnosis, and IHC and then we started adding molecular, a little star here basically is that's when companion diagnostics came out and I hope all of you have heard about those companion diagnostics because that seems to be the new opportunity for us in pathology and our participation in cancer management.

Then, in 2011, we created that only with the probes, but then in my particular situation, NextGen sequencing became pretty much a standard in the clinical diagnostic process, which involves pathology. Today we moved from the digital pathology system that has now been FDA approved that finally brings us in the US to the rest of the world, and we have a system that can be used in the clinical lab. It’s not that much different the system and/or the new systems that come about. They’ve been used in there search and in other areas other than a clinical primary diagnosis.

All of this takes us from a specificity and sensitivity process, and again the acceptable type of result was developed. The process was developed in 150 years ago, and now we're being asked to do much, much more with our process. That leaves us to, where do I start? What is the demand, what is the need, and I will tell you it’s being patient-centric, being patient safe, and patient quality. Each and everything we do in the lab if it's not tied to that I will tell you, why are we considering? So, with that timeline in mind, the tried and true, but old process that we have and I think everybody needs to understand, we provide very good and accurate information. In my mind, how much longer can that be true?

So, there is many opportunities…there are types of automation in histology I would venture to guess that all the participants are using some sort of automated immunohistochemistry. My issue with starting there is that that’s at the end of the process. And there are so many things from the selection or taking of the tissue sample to getting to IHC, but that seems to be the trend for the last15, 20 years of my experience in the lab that automation as it comes out and becomes available, and it becomes affordable to us in the histology lab, typically it’s designed to be at the end of the process. The newest automation for the clinical lab is digital pathology. That’s at the very end right before and replaces that delivery of the slides. Everything seems to be skewed to the end and then it creates its own issues.

Types of innovation in histology are, how do we move from conventional tried and true methods to more small batch and rapid—and that seems to be the term everybody uses—rapid processes. It is, and there is a lot of truism to the fact that if you move into smaller batches you have a better control of your quality, quality control, and your quality assurance which leads to improved patient safety and results. Most use automation in histology. I would say the two are number one your IHC stainers and then your H&E stainers

The other innovation with the H&E stainer that has come about is that the vendors quickly have learned how to use one of the innovative tools of co-location to bring two critical tasks or two methods and processes together, they added some robotics and now a lot of us have H&E strainers that not only stain, but then move right to cover slipping, that’s the co-location. The new automation is automated and betters are coming at us, new microtomes, more of the automated microtome and there will be many more opportunities and offerings in digital pathology, so that’s the type of new automation that is coming at us. Stop and think for a quick moment. How am I going to put that into my laboratory? It’s difficult and it’s hard, it's expensive, and there is a process that you have to go through before you decide to purchase these instruments. In the last… in my career. We went from $20,000 to $40,000 for a piece of instrumentation to $200,000+ for a piece of instrumentation.

Wow, where am I going to come up with this? It has a return on investment, but it has to have a lot more thought. So, you have to focus on how you're going to connect and integrate a piece of automation with the rest of your manual process. But all of that, where you decide to start where you decide to change from manual to automation will all help you become more patient-centric because anytime we remove a manual task and replace it with automation and/or connect extra automation, it will improve the quality, it will improve safety, which means our ability to confirm that result as the pathologist continues to perform their review of the information and release it out to our clients the physicians, and then ultimately our primary client, the patient.

Four types of histology automation

There are lots of opportunities and I will tell you, I don't have the answer of where we start. Each and every lab is still so diverse and different. It is a discussion about where you're going to start at and/or what your next step in automation will be. When we look at the automation, we also as I just spent the last couple of minutes talking about only instruments. There are really four types of automation that we need to talk about. Yes, absolutely instrument automation is necessary for us to get out of the tedious and repetitious manual tasks. But then, communications. By communications, I mean how is that instrumentation connected and tied into your LIS? Data to information conversion, that is what kind of software is going to be available to us and then more with the information management is what you may have heard - - talked about in your institution the big data.

Now we want to take all of those data and everything that we're doing in histology—and right now histology is fairly non-connected to the rest of the healthcare system. We have some automation, some communication. Most of our data to information conversion is a manual process which we don't have a lot of time for, and our information management is very much in the infant stages. But what everyone is interested in is how do we get to—which is difficult terms for us—standardization and utilization of our process, to making sure that we're doing the right test at the right at the right time with the right pathologist for the right patient. That's hard for us to do while we're in this manual process, but each and every step-in automation and putting in that background process of connectivity, computers, and software, it helps us to move to being able to fully connect and integrate each and every piece of our automation.

So what do we need to do? It’s nothing that hasn't been asked of us for the last several years. We have to understand our process. What's the manual parts, what’s the automated parts? What can we automate? What will happen if we automate? But as we use automation and we continue to connect it, connect it to our systems and then connect them to each other through co-location, we get to define new standards and we can comply to those standards.

The importance of histotechnicians and technologists in Quality Assurance

We can then take the knowledge that the histotechnicians and technologists have and apply that to other areas of quality control, quality assurance, and development instead of just being required to produce and be in a production mode for a good portion of our day. This is where the patient-centric, the patient safety, the improvement in quality, will really happen for us is when we have implemented a certain level of automation. I don't believe it has to be complete automation, but when we get to a certain level at our different labs, it allows us to start concentrating on those other things that the clinical lab started almost 40 years ago. The med-techs, the medical technologists, the clinical laboratory scientist, the medical laboratory scientist, they are able to move to that next level, which is difficult if nonexistent for us in histology. That leads us to develop new innovative solutions, and address and incorporate new testing platforms, new disease management participation that they want from us. It gives us a system approach.

But you have to watch out for all the same basic fundamental obstacles. Just plopping in an instrument will not solve any issues that we have in the laboratory and depending on what automation you look at can have a time factor and can actually— not can, it will—change your whole workflow and that does not necessarily mean it's going to be for the good. So, it takes a lot of concentration, a lot of planning, and a lot of participation from us, as the histotechnicians and pathologists.

What do we know? Part of our problem is we have guidelines to our whole process. We don't have any standard, and if you had any of my workshops previously, I will tell you we have got to move. As we move to more digital pathology with algorithms, additional disease management software that is in there, the companion diagnostics and trying to make a decision between whether a patient is considered positive or negative—that's our big qualitative decision in pathology, yes or no. So when we go from yes or no to someone who has 49% positive cells or higher gets a certain new drug. Are we confident that that 49% that we give through all of these new tools is actually 49, or is it 48, or is it 47, or our 49 is 52? There’s going to be a bigger demand for us to quantitate what we do, and that will drive that we have to look at what and how we do, so that we can create these new standards.

You also have to understand that instrumentation will never replace us. The people, the process, and the performance, the endpoint of that whole system you cannot, and we cannot, take them apart. That's where I see a lot of people get frustrated at that. The easy calculation is let’s put in a couple of pieces of automation and how many techs are we going to get rid of? How many we were going to lose? How many FTEs? My whole process over the last several years and I've been fortunate that everything that we’ve done in automation and where I've tried to help other labs, it's never a reduction. It's always a reallocation which allows us to move to that next step in the laboratory process, in the patient safety process. I think all of us have to think about that.

But it does make you think about whether you have 13, 16, or 22 critical task steps in your process from specimen to result, they have to rethink those steps about which ones would give me the best benefit using automation and which ones do I need to put off to later and wait for new innovation and automation to come? Again, it’s really, it's the same process that we have been doing all these years. We have to look at all three pieces: people, process, performance. And just, it's more of a plug and play of how we're going to do this. But never remove the people from this process.

Where do we start? Just like always, you think it takes too much time. No, it doesn't. That's where I find that, the biggest obstacle for us. If you keep doing the same thing you never can expect a different result. Quick and repetitive tasks, they can add up to weeks of time that you were released back into your laboratory. When you release that time, think of it as, I'll be able to apply that time to better quality in the results that we turn out. Don’t worry about the technology, that will be the issue of the vendors. Let them develop the technology they need to hear from us, but don't worry about it, don't start with how is this piece of instrumentation, this new software, actually going…? How's it going to work in a bigger picture is what I always suggest we do.

Does your strategy support your goals?

Automation can make your life easy, it can also make it very difficult. So, you have to always build with whether it's a piece, multiple pieces, or a whole new reorganization. They have to build and always keep the end goal in mind. Other issues will clear up, they’ll start to clear up as you keep the end goal and continue to build towards that. But never expect magic, it's just not going to be that easy. You're not going to be able to put the shiny new instrument into your multi-functional process and poof, all issues go away. Automation is only as good as the strategy that you deploy. The difficulty for us in pathology, and for the foreseeable future is whatever instrumentation we are given a approval to move forward with, it has a very complicated manual process before and after, typically. Magic is not going to happen, small steps are going to happen for us, but multiple small steps start leading to big changes.

When we get to quality, it’s the quality. We get a quality and a consistency. One of the things that I’ve been always trying to use over the last year is that since there was an FDA approved digital pathology system, they prove that we have a good baseline because they proved no hypothesis. They used multiple laboratories staining the slides, multiple pathologists not necessarily connected with those laboratories reading the slides, and they had a very, very high concordance and agreement on all those results. To me that says what we're doing in histology is we're creating a very high-quality product. Consistency is always our issue. Can, if you have more than 2 techs, you have 3 techs, 10 techs, 50 technicians performing these manual tasks, it’s that consistency. And what that means is that when we're inconsistent, the pathologist has a wide range of acceptance. As you improve the quality and consistency, acceptance range for the pathologist shortens, becomes narrow, and they are able to concentrate more on what I am seeing in the image or the slide? Is it because of the patient or is it because of the process? And I think that's our biggest challenge, is how do we get there?

Certainly, moving a small batch, removing your wasted time and effort, putting in automation to time savings. The time savings are not for anybody else but the histology lab. You get to take that time savings and apply it to your patient-centric results. The more time you have working on the results and improving quality, the better the outcome for us. You get metric visibility and that term is the one that scares everybody. Yep, we need to have more data around how long it takes us, how we take it, where we're developing opportunities and then performing errors, and how do we correct them? That's a big struggle for us still in the anatomic pathology and histology lab.

Turn-around times, the big picture

You get operational efficiency just being able to walk away from an instrument. Think of the operational efficiency all of us gained when we went from hand staining IHC or hand staining of H&Es, to you load, you press a button, you walk away with the H&E you come back and you have a cover slip slide. Great operational efficiency and that can happen with almost all types of automation if it is implemented correctly. Of course, you get your reduced turnaround times, and remember that first slide. That's what everybody's trying to do, 24 hours or less. So, you talk about all of the steps and issues and what's the right fixation time and what's the right processing time? It’s hard to do, but as we keep adding automation and putting in these process improvement tools, you will get a reduction of turnaround time and that's critical for us.

The difficult part is we have the rest of the healthcare starting with our primary customer pathologist, that we don’t always realize if we reduce turnaround time by an hour we may not—in the technical process that is—we may not necessarily see that result in the result in posting of results. So, it’s somewhat removed for us. Your process, developing your process, can make that closer to you, but you have varying processes connected to our technical process.

But everything still has to be, I'm going to do something. I'm going to put in a new piece of instrumentation, a new piece of software, rewire my laboratory, connect everything. What's the benefit to the patient. One of the other things I find with the patient-centric consideration is they think we're already completely automated and connected. They have difficulty understanding why their interaction, which is hard to get on that mobile device, is typically a PDF of our Word document of the pathology report. That is what’s pushing at us, but they have an expectation that they're going to get something else from us.

Understanding your lab process

So, before automation. You've got to understand before you start and go out to talk to the vendors or go to a show and start looking at some of these new instrumentation, or start talking about what new LIS and that, you have to understand and stay in control of your lab process. It can be working, but it can be completely out of control and in chaos. You're continually fire-fighting, you can't get slides processed, you can’t get blocks processed when you need them, you can’t get slides cut, stained, you have issues, you're doing a rework, so you have to figure out where you're at. You have to get into that stabilized and aware before you can really even consider a purchase and bringing in more automation. That's where the innovation tools come in, to get you out of that chaos and move you to this last stop, which you are in continuous improvement with staff involvement.

That scares a lot of us because we have been trained and we have had confirmation from our primary customer, which is the pathologist that I want the same thing done each and every day. And they become the primary customer, the pathologist becomes very uncomfortable with continuous improvement, because that means there could be a change. Good, bad or indifferent, it's a change. So, you've got all these things, but before you really start looking at what is the next piece of automation, or the first piece of automation into the laboratory, are you out of that chaos and into the continuous improvement? If you’re not, if you’re still in - - you've got to stop. You’ve got to stop. Delay any investment that you have. It will be way too costly and can gain. What you need is making sure that you addressed and know exactly where your shortcomings lie in your process, what you can and can't do through a manual, a manipulation of that to improve it and that will lead you to where you're going to put automation in.

You may have taken something like embedding and done everything you can to quality control and quality assure that, you still have a certain level of defect creation. At some point, your manual process will have a limit to how far you can go in the improvements. Those usually lead you that I need to do something, which is the next step, which is automation, but you have to make sure to stop and think through all that because if you get approval to buy automation and invest in automation and that doesn't go well, you're going to double, triple sometimes quadruple the time to move to your next automation. And what you'll find is as you move to automation and move it to the next piece, if you have the right tools to control the manual process, it can be a much faster timeline of moving to the next step in automation, which will eventually get you to the next level of laboratory improvement which we all struggle and fight with. Always stopping, consider, don't jump into automation. Do not go into the idea that automation is going to fix the problems. Automation will just create consistency in the quality of the product that you're producing now. So, if you produce a poor-quality product, you will consistently produce that poor-quality product.

Create definition in your process, I guess. Because of the cost, you're going to have to clearly state and define what you have as a requirement for automation. You have got to speak to… the term I use most time, you've got to be able to speak up to the C suite, the carpeted offices and everything else because again, we're asking for $200,000 I mean $40,000, $60,000, $80,000 for piece of instrumentation. You're not even really truly into a budget process. You're a leftover of the budget process. Likewise, you start asking for $200,000, $400,000, in a calendar year, you’re in a very defined and data driven budget process. So, you have to be able to clearly define and state what you need and what the requirements are for that automation to work.

What you'll find is that a piece of instrument automation comes into the laboratory, we have all these other ancillary processes, interfaces to LIS is rewiring for different cabling and everything in your laboratory. So, the worst thing you can do is get into this process and then come back and say, well you know the estimate is X, but I need XX to actually make it work .And before you can really tell whether that what you think is your automation solutions, we've got to start requiring instrument simulation through testing in the laboratory. That's always been a difficult thing for us because we are so behind the gun in trying to produce the product.

Make cassettes, make slides, do the staining and get it out. So much part of our time is involved there, it is hard for us to bring in the improvement process, instrument, process software and then run it in parallel and simulate and do testing to make sure it can and will work within your system. I think that's a key point that we have to get across to all vendors as we move forward and continue to try to advance the histology and histotechnology process. We have to do very similar to what IT and the clinical lab does before they ever agree to purchase and move it in to either enhance or replace some other step in their process.

The partners in quality assurance

What it means is this whole decision and definition of what you're going to do can’t be just in your laboratory. You’ve got to have everybody in your organization and become real good friends with IT personnel, because almost everything today has that point that it needs to be connected to the data collection for your whole system. right out of the laboratory. So today, this is kind of… I just throw this in here because this is pretty similar to what most places do. We've got this big LIS system sit out there, you may or may not be participating in a completely electronic health record, that’s a patient demand right there.

You've got order entry, whether it's coming from where the specimen is being collected or it comes in and you create the order entry in your laboratory. Less than about, it’s still hanging around 10% of us have bar-coding and tracking, our histology instrumentation may or may not be connected. Slides go out to the pathologists that and the pathologists reuse it as the laboratory information system to order re-cut specials and IC, then it goes back into this manual process. This is a continual loop of how the information flows and the product flows each and every day in histology. Very simplified, but it's not what everybody, especially our patients expect us to have. The idea is that we are more integrated. The reason I used the circle as though we are integrated because the histology automation sits in there. There is a whole new pathologist process coming to us.

The histology process, the LIS, has to sift and collate everything, it has to be connected to everything, everything we do, because right after all of that data that is pouring into the LIS is everything expected to go out to this electronic health record. There will be need for us to confirm and prove our results more than a pathologist saying yes, no, or giving a semi-quantitative calculation. That process has already started and we're struggling with it, but we will have to do that so this is a completely integrated, not a use when I need and move back in. It's pushing us to how are we going to become fully automated.

Then again, you've got all of this new thing, digital pathology coming out there that is quite expensive and demands that we really look at the whole process about how do we get as connected and as automated as much as possible. So, this is our end goal is how do we get to that fully integrated and connected automated process?

These are just some examples of some of the innovative solutions we work with. There are multiple tissue processors out there. How many of these tissue processors that are pictured here are connected via bidirectional interface, that information is constantly being collected and analyzed through our information system. That's the need, that's the want, that is what we have to demand. All of these work well for us. We all have our own preferences, but from everyone outside our laboratory and in the bigger healthcare and above us in management wants to know, so how do I get all the data that's on here? It’s compliance, it really leads to compliance and risk for us, is how do we give all that data and we confirm that I pressed X button to start any of these instruments with this protocol per protocol completely ran and it's being continually monitored through an interface. That is pretty new for us, but that's the demand.

These are the embedding things, we thought we had great automation when we went to the one on the left. This was our first piece of embedding automation. Now there is other instrumentation's out there or processes such as fully automated with the specialized cassette, and that causes us all kinds of angst, to a process on an existing tissue processor that will provide a certain type, a certain type of tissue or all tissues that are ready for basically cooling and cutting.

Those are the type of innovation solutions I see are going to have to come to us more because moving to this instrument or this instrument away from this instrument buys us time in the laboratory. Time to work on our quality and our patient result. Microtomy. Microtomy still stands as a big issue. It's difficult to get a majority of technicians, technologists, to used motorized microtomes. It’s still the number one in the manual that already has a proven safety and employee issue, but we continue to use it. We have to move to new innovative, such as with this one, cooling of the block. Now there are microbes out there with lights behind them to eliminate the tissue to a programmable motorized to a fully automated instrument to all we're doing is placing the blocks in and using a connection to the LIS to get the number of slides, levels, steps, to what's been used for a lot of years is a tape system that has been used with microwaves.

I don't know what the solution is, but moving from the manual is one of the big steps we're going to have to do that gains us time to move to that next level in our process. And then also, there’s… sorry, there’s the staining. We've been very receptive to all kinds of different staining solutions. How we moved to whatever that accepted process was to moving that to further automation in histology, that's a challenge that we all have to figure out how we're going to do that.

The human effect on automation

One of the things I wanted to put forth is when you go to the automation process, when we get more. Let’s say tomorrow we are completely automated from start to finish. Number one, there won’t be less technicians/technologists. We already struggle with the number; we’ll be able to move to that next level of improving quality and patient safety and care. There's a lot of logic when we move to automated processes and it's not mixed logic. So, it's fast, simple, concise, predictable. I put a block in here, I get an X result. I put a slide in this end, and I get an X result. It’s going to be consistent and predictable.

When we throw in the human effect on automation and that is what we do every day when we have manual, automated, and manual processes we’re slower, it's complicated, it is unpredictable because we're constantly evaluating the data, and we're evaluating what the solution is before we take the action. So, this is what really slows us down in the histology laboratory and this is what we have to get to. We have to get to this automated process as much as possible, and I don't know what the endpoint is and how much, but where we can streamline that process to where it is fast, simple, concise, and predictable, that’s where we get the best game and the best, best spend of our dollar.

Pathology automation. Automation will increase our productivity, no doubt about it, but if we’re producing poor results, it just increases the number of poor results we're putting out. It will reduce system costs, whether you're buying proprietary reagents or whether you have a completely open system. I’m a big proponent for closed systems, what’s been manufactured and proven for that piece of automation which we typically do for the vast majority of IC. We still have to have a couple of open opportunities, but it reduces our system costs, and that is because we are now fully connected. We know exactly how the instrument is performing. We know exactly when it needs to be maintained and looked at. All of that happens now even with some of our most cherished automation instruments that we have. That's usually a manual decision, an individual's decision when we're going to do something.

Automation and process/quality improvements

So, automation also gives us that opportunity to improve the process and quality and that's moving to that next level to where we're not spending the bulk of our day in production. We're spending our time in quality control, quality assurance and development. And I think that's where I believe we need to be, but always remember automation will not reduce your staff. Somebody still has to maintain, somebody still has to, manage the manual to automation and move to using in different parts.

That said, we have some key considerations that were there in automation advantages. We have patient satisfaction, increase in action and accuracy, we get less employee cost and higher volume production. So, it is a complicated process and we have to make that decision of what the next step is. So, we're going to struggle for the next couple of years. New things are going to come to us and we're going to have to decide where we make our next move in automation. It’s not something to be scared of, it’s not something to shy away from, it’s something I feel is necessary and we all have to work together to get to that next level of quality improvement in the histology laboratory.

Our disadvantages? It will give us last fair versatility, but that just means that we have a more standardized process. It is a large additional investment, but I think now is the time for us to demand to participate in that process.

This is a patient safety issue, making sure that we can confirm the results that we create and that we can do those in the most efficient and reliable way. We have to demand. There are unpredictable costs when you look at a piece of instrumentation, but that's usually it's because we're not system aware, we have no idea what our needs for connectivity are. Retraining, reallocation of staff will take us time, but I think that will absolutely be an employee satisfies that we move from the bulk of your time is production to affecting quality. And depending on your lab, automation just may not be a good fit at this particular time.

But again this is our end goal. This is what we're trying to work towards. We have to find a way to move all of anatomic pathology, histology, histotechnology. We have to participate in where healthcare is going, into this future information, integration, and connectivity. And that demands that we look at how we replace our manual processes with automation and automation that’s connected. Automation can help us, it’s definitely being demanded by the patients. We just have to figure out how we're going to move into this process without creating additional difficulties for us.

Q&A

What would you say is the biggest obstacles to implementing more automation in the histology lab?

The biggest obstacle and the first obstacle is trying to decide which manual process you are going to look to, to automate. So that's the biggest obstacle because we are all trained, all of us histotechnologists are trained to do these manual tasks. So, we have to not only decide what we're going to replace as a manual, but what we're going to do with our reallocation and retraining of our staff. It’s always been easy to pick the staining because as we all like to confirm that's a walkaway. So now we have to figure out what the next block away is that will give us the same desirable results that we got with the staining aspect.

This is a comment that the person has asked for your comment on. She says it is very comforting that you have a constructive view on the need for staff, even though automation is looming, looming, looming and blooming. Any comments on that?

I would like to use the example that we're very attuned to companion diagnostics. We're going to participate in that level of precision medicine, which we all are at some point. Just think of the added time, effort, and change the workflow without automation. We need more people than we have available in the laboratory today, and we need to be able to perform at that level, that sensitivity and specificity and be able to produce the slides in a timely manner to give the pathologist all the time they need to analyze all this. This is all about there just is not going to be any reduction in staff. I will tell you that that's always a question that comes up from your finance. It’s the easy one. How are we going to and how many FTEs are we going to reduce if I spend $200,000, $400,000 on this an instrument. Let's move the conversation to how do we improve our quality and become more risk assured, and we no longer write any settlement checks to anybody outside of our system. Those are the type of conversation we're not comfortable with, but those are the ones that you have to have happened to say, I'm not going to reduce staff. I am going to maintain staff. I may even have to increase with things like digital pathology, but there is a proven and now a definable and connected result that the other part of healthcare will like.

Do you think that H&E staining will become obsolete?

Not in my lifetime. The number one primary tool of pathology and the pathologist is the H&E stain. We are looking at processes and tissue relationship and cell relationship. I don't see the H&E going away. If I remove the H&E today from my laboratory, it is a very small cost to everything else that we're doing. Just look at the cost of a special stain, the cost of an IC. If I remove the H&E I would have to move to possibly every case, every slide, every block getting some sort of standard immunohistochemistry. The information gained from making that move is not adequate, it’s not what is required, so I just never see it leaving. That is the baseline information that decides what you're going to do next in your reflexive testing. That's what we do mostly in histology we’re reflexive testing and/or whether this is a sample that needs some other consideration. It doesn't matter the amount of automation. It is the primary tool of the pathologist and will remain.

What about the automated cover-slipping of slides, is it not valuable or do you prefer manual covering?

I probably didn't make myself clear. I think that’s one of the most valuable things is where the vendors have created their staining, almost all in IHC and H&E staining and putting their cover slipping. Manual cover slipping saves you so much time, effort, money. I just don’t understand why everyone doesn't have a primary H&E stainer with a connected automated cover slipper. The next question is glass or film and you have to decide which works in your process. Good news is glass or film both work with all the new generation digital imaging instruments, but that is when we get more solutions like that co-location of cover slipping and H&E or processing and embedding. That's my, that's what I've been asking for the last five years. Who is going to come out with a tissue processor that has automation that moves right to it and better. So, I put a cassette of tissue in and it comes out ready to cut. That will be the next big automation that our co-location automation I’m looking at. I believe that whatever you're spending in time of maintenance of that on an automated cover slipping, is well worth the time and releases people to do something else. It more efficiently affects quality.

Do you think they will create an automated system where a specimen goes in one end and comes out on a digital computer with scientists monitoring the platforms at various intervals. Do you think this is possible?

I think it could be. We're looking at a future state. The difficulty is so I'm aware of - They tried to move everything to this quick all blood, tissue, and blood for all of the diagnoses. The difficulty here is throw out - - that was a bad idea all together and I think there's still a lot of data out there that says circulating cells from tumors are not the same as when we take a biopsy of the actual tumor. The tumor is its own entity, it’s its own - - You stick it with the needle, you tack it with the chemical of monoclonal radiation, and if you don't completely eradicate it, it reacts and changes. The prediction out there is that biopsy is diagnostic, biopsies are going to go up over 30% by the year 2022. That means precision medicine to us and that not only are we going to take the initial biopsy, but there will be many more follow-up biopsies so we can manage that at the disease management part. There will always be competition between how do we move these things to a blood tube. And the reason I think most people are trying to come to that automation such as we have in the clinical lab, is that is a very definable specimen management process. Think of it, anything that we receive is tissue. It is not easy to put into a tube. We have a very complicated specimen acquisition and management process, and I just don't see that going away? This is what we did, this is histotechnology and anatomic pathology. There is going to be more of a demand for us to participate in that disease management with the biopsies. Will more resections go away? Absolutely. If we're being more proactive in our health care, I would expect to see less colon resections, less breast mastectomies and everything else. And, our biopsies are growing up because we will do multiple biopsies with a patient through their disease and precision management process. I don't see that I do not see the tube replacing us any time soon.