Pro Medicus Limited (PME) Earnings Call Transcript & Summary

Thank you for standing by, and welcome to the Pro Medicus PME Full Year Results Briefing. [Operator Instructions] I would now like to hand the conference over to Dr. Sam Hupert,CEO. Please go ahead.

Good morning, everybody, and thanks for joining us for this results briefing. As most of you know, Pro Medicus is a health care IT company specializing in enterprise imaging and radiology information systems. We work in 3 jurisdictions: Melbourne, our corporate headquarters; Berlin, our R&D center for Visage products; and the U.S., where we have an office in San Diego where we -- our biggest market and staff located throughout the U.S. that traveled towards clients. I think the important thing is we are a technology-focused company with over 40 software engineers and support staff. In terms of product set, we have 2 products. Visage RIS, which was our core product before 2009. It deals with all the practice management, billing, scheduling for large radiology corporations. The key market for that product is here in Australia. And on the right, Visage 7, which is the clinical desktop that radiologists use to call up, enhance, manipulate and diagnose images electronically. In terms of the results, I think we're very pleased that all of our headline numbers headed in the right direction. Revenue, despite quite significant currency headwinds, was still up 19.5%, and profit after tax over 33%. I think the other pleasing numbers were the cash and investments, now just a touch under $62 million, up 42%, and the company remains debt-free. As a result, we have increased our dividend this year, so $0.15 a share fully franked, including the half year. So the full year amount was $0.15. So all in all, we felt that pretty much all of our key metrics are headed in a positive direction. In terms of setting ourselves up for the year and for future years, '21 was the biggest year in terms of new sales in our company history. Just prior to the beginning of the year, we had won the Northwestern deal in Chicago, but starting after that, clearly, 6 key deals, 5 in Europe and 1 in -- sorry, 5 in the U.S., sorry, and 1 in Europe with a number of key academics plus a mixture of nonacademics such as MedStar and Intermountain. So only measure, was our busiest year, most probably the busiest sales period in our industry in total. In terms of where we sit, this is an interesting slide. There is a bake-off called Best Hospitals U.S. News every year. And on the left, you see where the various clients landed in the top 20. As you'll see now, we have 9 out of the top 20 using our products, over double our nearest competitor. So certainly, we are making an impact in some of the most well-known hospitals and health systems in the U.S. In terms of the revenue split, many of you have seen this bar chart before, I think the key thing is that -- on the right-hand side, in the orange-pink, the transaction revenue grew and in numbers grew even more. As I said, we had some tailwinds with currency, which have now abated to some degree. But total revenue in pretty much all of the sections increased during the year. We've put in a new section to show the archive migrations, which is the work we do when we get an archive contract in transferring the data from the previous system to our system. So we're now showing that revenue separately. And that came in largely from MedStar and Intermountain that not only took our viewer, but also took our archive. In terms of the operational model, again, we think it is proving that it is a very good model for us. It is very fine-grained. We have the minimums, which underpin our contract value, and that's how we announced the contracts. And I think the important thing is the client grows, so do we, because we charge for every additional tests that they do. I think the key figure on this slide is that the forward revenue increased from, I think, around $270 million when we last announced it to $320 million. This is a 5-year window. So if a contract is 7 years, the final 2 years are not taken into this account. And I think it just shows the growth of our base as we add each contract on top of the ones that we have already in place. As mentioned, we had a 16% year-on-year increase in terms of transactions. It would have been 28% on a constant currency basis. We also believe the first quarter was mildly impacted by the tail end of COVID. The trough that occurred in exam numbers was really around the March, April time frame in 2020. But towards the end of the year, we still weren't quite where we were pre-COVID, but we were getting closer. So there was that first quarter, in particular, was less than the normal 100%. But pleasingly, we were able to make not only that delta up, but obviously increased transaction numbers in a positive way from there. We are telling the market that we expect this growth to be a step change or a step growth in FY '22. We now have Northwestern, NYU, MedStar and others that were implemented. In the last fiscal year, well, particularly towards the end, we'll now give 12 months of revenue, each of those, which will be quite material because they're quite large implementations. And we also anticipate additional revenue coming from those sites coming -- that we've already announced in the first half, Intermountain, which is large, 2 UC campuses and U Vermont. So that plus growth in existing clients, we are seeing organic growth across the industry has returned. There is some M&A where our clients are acquiring smaller sites, and all that adds to the mix. And then I think the other area of upside is clients adopt our new products, particularly open archive and worklist. They are also transaction-based, so we will see them contribute to the total of transactions going forward into FY '22. In terms of professional services, many of you have seen this slide. It is around project planning, training and implementation. I think the new point on this one, the data migration as part of an archive sale, which we showed in the bar chart of revenue, those fees are once-off, whereas all of the other professional services are spread evenly across the life of the contract. So more of an annuity style revenue rather -- whereas data migration is a once-off. In terms of operating leverage, clearly, this is an important thing for us. I think we have proved that the offering is highly scalable. Our revenue has increased significantly more than any of our cost base. We do not have CapEx. We don't sell hardware. It's a software-only model, a true SaaS model. Training and installation from the time that we sign the contract is all charged as professional services. So we believe the business has high operating leverage, and the margin has continued to grow as the footprint increases. We did have a significant step-up in the margins in the first half that has continued into the second half. We believe some of that is because of the reduced cost base because of COVID-less travel, conferences being virtual rather than in person. And some of those costs will come back, but we still believe that we will be able to maintain our margins higher than they were this time last year. So somewhere in between where they are now and where they were this time last year. But clearly, that needs to play out still. In terms of COVID, we were able to transition to work-from-home 100% in all jurisdictions in mid-March 2020. We have been operating at 100% capacity globally. Our sales and marketing efforts continued throughout. As I said, we had our busiest sales period ever. I think a lot of that's got to do with the thinness of our technology that we're able to demonstrate it very readily through the Internet, in large-scale demonstrations and in full diagnostic quality. And I think that sets the time for a lot of the opportunity going forward because clearly, there's a lot of work from home. Radiologists had to pivot back in March, April last year. And even though the U.S. has come back in most parts to near normal in terms of on-site presence, it's never quite got back to the full on-site presence that it used to be, much like most of the industries here. So being able to work from home on demand, seamlessly on demand, is a huge plus. And as I mentioned, exam volumes in our back at or, in some cases, above pre-COVID levels. So all in all, we were not impacted negatively by COVID, other than a few months when exam volumes dropped. That was more around the March, April time frame. But now it's back to business as usual. In terms of our product set, Visage RIS. As I mentioned, this is used by largely in Australia. We do have some clients in Canada who use it. It is a highly localized product. The key thing for us is it is progressing well. We are progressing well with our 2 key contracts with Healthcare Imaging Services, used to be Primary Health Care, and I-MED. Since we last updated the market, the Healius rollout is now complete. So that's fully up and running, and that's a transaction-based model. We are seeing some upside in the I-MED contract because of their M&A and growth, increased market interest in some new opportunities. So we clearly believe in rest in Australia, we are the undisputed market leader. So that's a good position to be in for that product. In terms of Visage 7, many of you have seen this slide before, it's what makes us, we believe, the best and most differentiated product in the market, the 3 key tenets of, number one in speed, functionality and scalability. And I think with the implementations that we've done even over the last 12 months, I think this proves these 2 -- these 3 key differentiators between us and others. One of the things that is driving a change in the market, and we are seeing increased interest and people changing their systems. Many of you would have heard me say this is the massive explosion in data, the new imaging equipment creating larger and larger data sets. And that trend is continuing unabated. And the traditional methodology of legacy systems, compress and send, where they take the file, compress it as much as they can, send it down the network to a pretty beefy work station at the other end that has a lot of software on it that then uncompresses the file, and the radiologist then manipulates the file. That methodology is now cracking at the seams because the files are just getting too big regardless of the type of network that institutions have. We, on the other hand, use a streaming technology. The files go from the modalities to a back-end server that in near real-time, does all the rendering and 3D production, and we just stream the pixels. So it literally is an on-demand service, even over consumer-grade Internet, and that's one of the key things that lets radiologists work from anywhere, including from home. They cannot almost -- they cannot tell the difference between being in the office and being at home, and that's on consumer-grade Internet. And I don't believe any other product can actually offer that. In terms of the opportunities we won, some of these slides have been out before, but I think a few key things. The last 2 points, NYU Langone. That implementation was complete, on time in June 2021. They're 100% live. We didn't say at the time we won this contract that we would be setting up a new R&D hub in New York, and that is going ahead. We were able to navigate around COVID restrictions in such a way that our people from Berlin or a person from Berlin is able to go to the U.S., and that will start later this month. So that project is going ahead and will be very exciting for us. In terms of LMU Klinikum, and we won that deal in October and the implementation was completed in December 2020. So we will receive a full half of revenue, although this is a slightly different model to the transaction-based one. But the revenue from this deal will be spread across the 5 years, and we will receive a full year's revenue from this in 2022. Zwanger-Pesiri, this was an existing client. They renewed a 5-year contract, so it was another key renewal for us. Then MedStar Health, and MedStar is an interesting one. It's -- technically, it's called an IDN and nonacademic. I think interesting for a few reasons. One, it was the first client to take all 3 Visage products. That is, our new worklist, our viewer and our archive, and it's 100% cloud-based. We think it's most probably the largest PACS implementation, 100% cloud-based, arguably anywhere in the world. And that implementation was completed towards the end of June, 1st week of July. And so again, we will receive a full 12 months revenue coming into FY '22. Intermountain Healthcare. It's the largest health care system provider in the Intermountain West region, Utah. It lies based out of Salt Lake City in Utah. Again, multiple products, viewer and archive. And that is slated for implementation in the September, October time frame this year. And then University of California, which we announced in February. This will be a staged implementation, a large 5-campus implementation, cloud-based, and the first sites will be done in the first half of this year. So 2 of the 5 will be staged in this first half and then another 2 in the second half of this year and 1 about 12 to 18 months -- 12 months away from now. So that's all going on track. And the last one, which is University of Vermont, that will be done in this coming half. So effectively, we've had an interesting time in terms of implementations. We think one of our strengths is the ability to implement quickly. We -- OSU was our first 100% remote implementation last August. We have now developed a highly optimized hybrid model, which is a mixture of on-site and remote, and we believe this has been a key differentiator. And certainly, I think we've been able to prove with the next slide that we were able to keep up our implementations despite changing restrictions because of COVID in different jurisdictions. So as mentioned, Ohio State in August was our first 100% remote; Northwestern in May of this year was on-site; NYU Langone was 100% remote; and MedStar, hybrid, where we had a team on-site plus a team remote, and we believe that is most probably the model we will use. It's the most optimized model going forward. And it's worthwhile mentioning that there are a number of other smaller sites that we did background over this period. So in terms of implementations, it was by far our busiest 12 months ever. Just moving forward, one of the key factors we think that support our pricing model, which is the premium price, is that we have a proven ROI, and I think it's not just in terms of financial ROI, but certainly proven superior clinical capabilities, which for these institutions is critically important. There is a -- this is a slide talking about a certain way of looking at implants in the ureter. And on the bottom here, Eric Pepin said that some PACS, Visage Imaging, you're able to do this. It takes just a few seconds because in most other systems, it could take minutes. And I think this is a perfect example of clinical ROI. Moving forward to growth strategy. Again, we believe we've delivered on this, and it's consistent. We have expanded our footprint. We have had existing clients grow their transactions. Our new product offerings, particularly archive and worklist are starting to keep goals. We did have a second major sale in Europe, and we are leveraging our R&D capability, which I'll show in just a minute. Pipeline, we often get asked about this. Clearly, we had a lot of opportunities in the pipeline about a year ago, those opportunities we've been able to convert to contract. But pleasingly, we've had a number of new opportunities come in as well, and they were opportunities that were in the pipeline over the last 12 months that have continued to progress. So we're very happy with not only the size, but also the mix of opportunities in the pipeline going forward. I think things that will help us in terms of size of sale will be the newer products like Visage Open Archive. And as we said, 3 of the recent sales were both viewer and archive. And we think that will be a trend that will continue, not if everybody will buy archive from the getgo. But we are seeing an increased number, which is positive. Worklist, our newest product. As I said, we've had this in production in some smaller sites. The first large-scale client go live with it was MedStar, and that's been very successful. It can be implemented on-premise or in cloud. And I think the important thing is it provides us with flexibility for those clients, one like MedStar that just wanted to deal with one vendor. We were able to do it by having the worklist component. And I think we are seeing it more and more in the mix with opportunities in our pipeline. Enterprise imaging, the ability to move the product outside radiology. We are progressing with that, if there was any of our initiatives that maybe got put a little on the backbone because of COVID. This was the one, but we are starting again with some of our academic clients. We don't yet have a material implementation across the whole hospital. But again, we believe we are progressing in the right direction with this. And we're hopeful within the next 6 to 12 months to make some more headway in this regard. Visage in the Cloud, I won't go into it too much because clearly, we see this as a huge strategic advantage. We are 100% cloud-native. A number of our opportunities, recent ones are cloud-based. As I said, we already have large implementations in there, UCs, Intermountain, UVM will all be cloud-based. And certainly, we're seeing a lot more cloud-focused RFPs coming through and opportunities where cloud is important to these large-scale operations. And the fact that we already have a number of large clients fully implemented, we think will stand us in good stead with those. Finally, I'll finish off with AI and the accelerator. Most of -- many of you would have seen this. It is a unique end-to-end solution, and it supports both research and production. It is a -- based on our Visage 7 technology. It's a sandbox version of it. We do offer this to key academics. And the main thing is, as its name implies, it allows them -- it provides a glue for these academic institutions to be able to take concept to product in a fraction of the time that would otherwise occur. So -- we see it as part of an ecosystem. We have an open API that will allow third-party and Visage-developed algorithms to work inside the Visage viewer rather than as a separate product that's not integrated, and we are looking at joint development and commercialization opportunities. Here's our team that leads our AI research: Malte Westerhoff, who's one of codevelopers and founders of Visage Imaging, co-developer of the platform; Detlev Stalling, who is the other co-developer; and MingDe Lin, who's based out of Yale, but is looking -- he's our clinical research manager in North America, and we will be building team further, given that we have got some new agreements. So in addition to the work we've been doing with Yale, we did sign 2 very important collaborations agreements, one with NYU Langone and recently, with Mayo Clinic. The one for Mayo is for all of their campuses: Rochester; Phoenix, Arizona; and in Florida. And finally, proof that the AI accelerator works. We developed our first diagnostic AI algorithm in conjunction the breast imaging team at Yale. It provides an on the fly in virtual real-time assessment of breast density. We did receive -- it was previewed at RSNA 2019 as work in progress, and we got FDA approval in 2021 February. It is implemented in one of our academic clients, and we're just assessing how and when they use it, and then we will be looking to offer commercially to other clients. So we see a particular project as a model for future AI development. Finally, this occurred last week, it shows the new world order. It's the HIMSS conference in Las Vegas. Was meant to be in February, got postponed. And I only show it because we did get registrants, but it is strange at the conference to see everybody all wearing face masks. And I think that will be the same for the RSNA later this year in November. And on that, thank you all, and open it up for questions.

[Operator Instructions] Your first question comes from Peter Meichelboeck with Select Equities.

I just want to ask about renewals. I think you said it was sort of due for renewal at some stage, but obviously, they had difficulties. I just want to see if there's any update there and also any of the other ones, and which ones are coming up for renewal this year.

Yes. Look, Sutter is still progressing. They've -- we have made some good progress with them. As you mentioned, they got particularly badly hit by not only COVID, but by bush fires and everything else. So we decided to give them some breathing space. That opportunity is a close-door discussion with us. It's not gone out to market. And we're hopeful to have something within the next few months, but it could be earlier than that.

Right. And which are the ones up for renewal this year? Sorry.

Allegheny was the other that is due during this period, and that has been renewed.

Right. Okay. Can I just ask -- you've made a particular comment around just how busy you have been in terms of the implementations and sales, obviously. But particularly in terms of the implementations, has it actually required any additional staffing to do that? And also the other part, what -- with people sort of tied up on the implementations, does that impact the sort of the R&D or the rest of the business at all?

The answer is the -- we've not hired specifically for any of these implementations. So I think the answer for that is no. It hasn't -- we haven't just gone out and hired 5 or 6 people. I think we've hired 1 or 2, but that's our normal sort of cadence that we do every year. But clearly, there has been big R&D effort and all the worklist because it's the first major implementation in the first cloud. But it hasn't prevented us from continuing with other projects and some of the AI work and also the product releases. I think we've gone through it incredibly well and the fact that we've got 4 out of the 7 are already behind us. Even with COVID, there were some juggling around that, working out and we've optimized our capability to deliver remotely. And as I said in sort of in the interviews, we've developed this hybrid model, which we actually think even in a post-COVID world would most probably be the most optimized. So it just makes more efficient again. But no, we've just hired the normal amount over the last 12 months.

Okay. And finally, just on the enterprise imaging sort of opportunity there. I appreciate that, that's still, as you said, sort of a little bit on the back burner, et cetera. And -- but I mean how should we be thinking in terms of the sort of the size of an enterprise imaging sort of opportunity versus just the sort of the radiology, I guess, in general terms?

Look, it depends on the department. So if they do department by department, some are bigger, like cardiology is roughly 25% to 30%, depending on the hospital of radiology. So it depends which one you get and what the institutions focus is because some of them are better known for one area than another. And clearly, that would be a bigger opportunity. We -- it's hard to tell because a lot of this is photo and video or reflective light. Most institutions don't know how many images they've got because these are just taken not as part of sequence of an order, as we call it, they've taken just as the patients present. So it's a little hard to tell, but we think if you had everything other than radiology and outside pathology would be an additional 30% to 50% depending on the institution.

Your next question comes from Sarah Mann with Moelis Australia.

Sam, it's David Meehan, step in for Sarah Mann. I just have a couple of questions. Firstly, a lot of the hospitals had a tough year financially with the elective procedures being delayed. Have you seen any impact to how some hospitals are thinking about new spend on technology, for example, slowing down of sales cycle? And then just following up on that, have you seen any of your hospital customers, especially the larger and better capitalized, been opportunistic in the M&A front to acquire some of these impacted hospitals? And if so, what kind of volume uptick have you seen, if any?

Yes. Look, on the first one, we haven't seen an impact. As I said, the exam numbers, which is the metric we charge on did go down in March, April last year because people didn't know what to do with COVID. Everybody was literally just shut down everything. But thankfully, it reopened up fairly quickly. And we have seen pre-COVID numbers, in some cases, even better. And when we looked at them somewhere in the early part, a little bit of catch-up for the test that went down during the shutdowns. But we are seeing organic growth back, and it's quite healthy. So at least the groups that we deal with, we haven't had -- there hasn't been an impact. And clearly, on the sales cycle, it's been the opposite. We've had our busiest year ever, and all of that was in COVID. So I think from our perspective, if anything, COVID has been a bit of a tailwind. In terms of the smaller ones, yes, look, there has been some M&A. It hasn't been major, but all of that would help. I can't really quantify it other than we do see the big always buying some of the smaller ones, and we saw that the previous year. We saw it with [ Kao ], we saw it with Mass General. To us, that's all positive because we get those additional transactions, but we see that business as usual. But we haven't had one where there was this massive M&A where it would add 60% or 70% to our volumes, but we live in [ a hut ].

That's great. And then just one last question. With the RSNA moving online last year, I'm just trying to get a better feel for how you expect to work this year and how we should think of marketing costs versus an [ ROI ].

They've said it's in person, that's 100% for sure. Unless something changes, we think marketing costs will come back. It won't be as much as the previous years because we won't send as bigger team. There will be restrictions, face masks. They've said no vaccination, don't come. So we expect the crowds maybe to be a little bit more focused, maybe not as many Europeans as we would otherwise get. But yes, look, some of that expense will come back, and that's why we think the 63% margin -- EBIT margins will most probably come back a bit. But we don't think it will go back to what it was pre-COVID.

Your next question comes from Josh Kannourakis with Barrenjoey.

Just a quick question on the -- on your cloud product. I'd love to get a little bit more context around out of the pipeline and RFPs, just how many of them are discussing public cloud deployment as an option. And maybe you could give some comments post the recent HIMSS conference around where the competitors' set are in terms of their current cloud capability.

Yes. Look, off the top of my head, I would say roughly 50% are talking cloud. And bear in mind, in some of the opportunities that we've contracted. One in particular, when we first started dealing with them, cloud was sort of in the background. It wasn't -- it was going to be an on-premise until they sort of looked at it all and then we're going cloud. So roughly 50%, could be a tad high. So there's definitely a momentum swing towards cloud. I think in terms of competitors, we only know what we hear. We don't believe anybody else is cloud-native, in other words, that can put their entire application in the cloud and it's working exactly the same way as it would work on-premise. So we think that's a very strong advantage. And -- we now have a number of cloud implementations, material ones under our belt. And in the case of MedStar, all 3 applications of ours are 100% in the cloud. In that case, it's Google DCP. But -- and the other thing about cloud is we're agnostic. So some of our opportunities are in AWS. We believe some will be in Microsoft. As you said, we will have examples in every one of the 3 big clouds. We wouldn't use a second-tier cloud provider. So I think as far as we can tell, we're the only ones that can state we have a full implementation, 100% from cloud. And with UCs and Intermountain, we'll have more -- and UVM, we'll have more.

Yes. Absolutely. And with Intermountain, I imagine that will probably be one of the largest sort of public cloud rollouts in North America, maybe globally. Like I guess I'm interested in your view around whether you're doing much outbound marketing or what the sort of funnel of inquiry has been specifically around cloud. And just -- I'm just interested in terms of especially relevant to some of that more mid-market, whether that opens up that opportunity for you in a greater capacity.

Look, I think it was a key theme at HIMSS that was just in Las Vegas last year. It is a conference that all the 3 cloud providers go to as well. So that dovetailed nicely. Look, UVM is more in that mid-tier in terms of volume. So there's a perfect example that used to be completely on-premise and just right now were going to cloud. And definitely, it has opened up opportunities, but not just in the mid-market. There are certain large health institutions that are now saying we're not IT companies, we're health care providers. And so they have a cloud-first strategy, and that plays nicely into what we're doing. But certainly, we are making it well known. I mean that's what you'd expect us to do. And as I said, not everybody's cloud and those would go on-prem may in a year or 2, migrate to the cloud or migrate their archive to cloud. So it's certainly a huge part of our strategy. And so far, it's working well.

Great. And just a quick question on your AI commercialization with some of your partners in the market there. I'd love to just get an update in terms of how we should be thinking about the timing and pipeline of any other product that could be lodged into FDA or brought to market on the next sort of 2 to 3 years' time horizon?

Yes. Look, it depends on the organization. Most of them already have projects where a lot of what I call the science or the research is already underway. And in those instances, with accelerator, we are the glue that sort of fast tracks that process. And so you may find that they complete the science and we do the FDA and what I call productize it. In other words, how you make it a product that's supportable and distributable, and then we commercialize it. So those would be maybe a little bit more near term. It could be 12, 18 months, it could be less. It just depends on FDA and how big a product it is. Those where we would do the science or it would be joint science could also be somewhere in that time frame. Some could be longer. It really just depends on the project. But I think there are a few projects that are further down the track simply because they've done the science, but don't have the wherewithal to make a product and put it through FDA. And if that works for us and works for them, then we'd be able to pick the ball up at that point and do it relatively quickly compared to the rest of the market. So we're saying it's not here and now. These are longer-term projects, but they're not 5 years away. They're hopefully 12 to 24 months away.

Your next question comes from the webcast. Will Hauser at Escala asks, "Are your margins expanding with new revenue and contained growth in costs? Can you comment on profit margin, please?"

Yes. It's Clayton here. I'll answer that. Our margins expanded with a little bit of both, obviously, new revenue coming through. Some of the contained costs were more to do with the marketing and the conference costs obviously not occurring in this financial year. As Sam mentioned, that will come back within November or the first half of this, the FY '22 financial year. So revenue had been the main driver of those -- the margin expansion, coupled with the decrease in costs for advertising. Clearly, as we look into FY '22, we think that some of those costs will start coming back to the conferences. But it will be some of the revenue that we get from the new contracts coming on board for NYU, Northwestern and Intermountain, and MedStar will start helping that normalize. So we think the margins -- it had a bit of a step change in terms of where it's going from mid-50s to over 60. We think with some additional costs with advertising and also setting up on the NYU R&D offers, the expenditure there will start to normalize.

Will Hauser with Escala also asked, "Can you comment any further traction in Europe outside the German hospital previously announced?"

Look, our commentary around Europe is it's unfortunately not one country. It' a whole lot of countries, and each one has a different health system or more importantly, different funding for health. We think outside Germany and the larger countries with more sophisticated health systems are looking at a different way of buying their technology. They're looking more towards the American model. So our commentary is Europe is a few years behind tends not to buy components like us or hasn't in the past and cloud hasn't been prevalent. We're seeing it more and more, I think, in the U.K., in particular, and parts of Europe. So we think it will be incremental. It will take a little longer. We think cloud will be a key part of it going forward. Now there's some regulatory things around cloud and things to do with the EU and data protection and cloud providers being American. But clearly, that needs to be worked through by the cloud providers. But once it does, we think that will open some opportunities. So Europe, we're looking at it. It will be more incremental and take a -- it will be more step-wise and what's happened in the U.S.

Michael Legg with Michael Legg & Associates asks, "What stops the competition changing to pixel-based communications like Visage?"

Yes. Good question. It's something we -- this platform was developed back in 2009. And obviously, we've enhanced it. But a few things. One, it's proprietary. So -- and it's all in-house. So we didn't use someone's toolkit, get 80% and then modify the last 20%. It's 100% proprietary and in-house. The second thing is it's not just one technology. We talk about the streaming because that's the easiest one to try and explain a bit like Netflix for radiology, but it's a bit more sophisticated. But there are a lot of other technologies in and around what we do because it is 2-way communications, and you can't have any jerkiness. Because the minute you have any jerkiness or hesitation in the image flow, radiologists will just go -- sorry, can't use it because they think they're missing something. And so there's certain technologies that try and optimize the streaming based on bandwidth, which keeps changing even on the best network every half second or so. So I suppose it's a few things to that question. It's proprietary. We didn't relive the road map. And they can't get a head start by just getting someone's toolkit and trying to put it together. A number have tried, haven't been successful. And then the last thing is the best form of defense is attack. So we are putting our foot down in terms of R&D and bringing out new features and new capability all the time. So we're making ourselves harder to catch. So as I said, at this point, we don't know anybody that has been able to bring out a platform that one-for-one competes with us. And I think -- and that doesn't mean one day they may not do something that's half is capable or 3 quarters or equivalent that we just don't know. But at this point, we haven't seen anything similar in the market. And I think the fact that we're winning the lion's share of the large public opportunities, and this probably supports it.

Will Hauser with Escala asks, "Are existing contracts being renewed on the old terms? Or are you managing to upsell improved terms when renegotiating new contracts?"

Yes, we've only had a few that have renewed and some have either been through additional volumes or increases in exam rates. Clearly, when we first started selling Visage, the exam rate has increased to some 65%, 70%. But some of the existing customers obviously pushed into that price. We haven't done that on a per exam basis, but there's somewhere in between. So we've been able to negotiate an exam rate that's fair and reasonable for both parties, but to bring it more in line.

Jason Yin with Lincoln Fund asks, "Are you able to comment on the list in other financial assets to $19.8 million, especially on listed investments in managed funds?"

Yes. So the other financial assets and investments of our cash holdings, so we've taken it out of pure cash and term deposits and put it into financial assets or investments. The main reason to have done that in the fixed income securities is to enhance our returns for those available funds that they're on. In terms of where they're positioned, we give those 2 some consultants that put them into the new listed and unlisted entities, and it's mainly due to get those enhanced returns.

Claude Walker with A Rich Life asks, "Could you please tell us whether you have lost any opportunities that were previously in the pipeline?"

No is the short answer. There are some opportunities that we don't get to look at. And thankfully, there are not many. But any of the opportunities where we've had a chance to compete, and when I mean compete, it is go through an RFP process, do demonstrations to the radiologists and then often there'll be a pilot or proof-of-concept where you have them use it with their own data. In that regard in terms of material opportunities, no. Now there've been 1 or 2 over the years where we haven't even got to the starting line. We don't get the RFP or we get -- the minute they do an RFI and say, what are your price. And say, well, we can't, we're not going to progress with that. But as I said, thankfully, there are few and far between. So nothing in the pipeline that we've competed on that has dropped out to a competitor at this point.

Julian Mulcahy with E&P asks, "When do you expect to receive revenue from the breast AI algorithms?"

We're hopeful that based on how it's going well, we've got it in situ in a large academic and it's going really, really well. I would tend to think realistically, second half FY '22, possibly first half '23, but that would stretch it out. So I think we'll get our first taste to some revenue, hopefully, second half of this financial year.