RadNet, Inc. (RDNT) Earnings Call Transcript & Summary

Good day, everyone, and welcome to today's call to discuss RadNet's recently announced acquisition of Aidence and Quantib and RadNet's strategy for artificial intelligence. As a reminder, today's conference is being recorded. At this time, I'd like to turn the call over to Mr. Mark Stolper, Executive Vice President and Chief Financial Officer of RadNet. Please go ahead, sir.

Thank you. Good morning, ladies and gentlemen, and thank you for joining us today to discuss our recently announced artificial intelligence acquisitions and our AI strategy. Participants in today's call include Dr. Howard Berger, RadNet's Chairman and Chief Executive Officer; Dr. Greg Sorensen, CEO and Co-Founder of DeepHealth and President of RadNet's AI Efforts; Mark-Jan Harte, Co-Founder and CEO of Aidence; and Arthur Post Uiterweer, CEO of Quantib. Before we begin, we'd like to remind everyone of the safe harbor statement under the Private Securities Litigation Reform Act of 1995. Today's prepared remarks and discussion contain forward-looking statements within the meaning of the safe harbor provisions of the U.S. Private Securities Litigation Reform Act of 1995. Forward-looking statements are expressions of our current beliefs, expectations and assumptions regarding the future of our business, future plans and strategies, projections and anticipated future conditions, events and trends. Forward-looking statements in this discussion include, among others, statements or inferences we make regarding whether Quantib's and Aidence's existing or any future products will receive European CE and U.S. FDA 510(k) clearance or other regulatory clearance and/or approval necessary for commercialization. Whether Aidence's and Quantib's existing and any future solutions will prove effective and whether RadNet's development and deployment of AI solutions will prove effective for improving the care and health of patients. Expected market acceptance for Aidence's and Quantib's products and the willingness of customers to use or continue to use Aidence and Quantib products in the future. Aidence, Quantib's and RadNet's ability to develop, maintain and increase their market positions in a competitive environment, and economic benefits and cost savings anticipated to be derived from AI products and solutions as well as anticipated importance of and impact of AI solutions to RadNet's future business operations. Forward-looking statements are neither historical facts nor assurances of future performance. Because forward-looking statements relate to the future, they are inherently subject to uncertainties, risks and changes in circumstances that are difficult to predict and many of which are outside of our control. Our actual results and financial condition may differ materially from those indicated in the forward-looking statements. Therefore, you should not place undue reliance on any of these forward-looking statements. Important factors that could cause our actual results and financial condition to differ materially from those indicated or implied in the forward-looking statements include those factors identified in the annual report on Form 10-K, quarterly report on Form 10-Q and other reports that RadNet Inc. filed from time to time with the Securities and Exchange Commission. Any forward-looking statement contained in this press release is based on information currently available to us and speaks only as of the date on which it is made. We undertake no obligation to publicly update any forward-looking statement, whether written or oral that we make from time to time, whether as a result of change of circumstances, new information, future developments or otherwise, except as required by applicable law. And with that, I'd like to turn the call over to Dr. Berger, who will make some opening remarks.

Thank you, Mark, and welcome, everyone. Today, I would like to describe to you not only the rationale behind the recent announcement of Monday this week in the acquisition of 2 additional artificial intelligence companies, but also expand on what the potential overarching opportunities are for that investment. Monday's announcement marks a seminal event for RadNet and perhaps imaging, but not limited to just the specialty of radiology and imaging, but perhaps also for the healthcare market. Artificial intelligence has been a term that has been used for quite some period of time and it has yet to develop a coherent strategy. Today's conference call is an attempt to explain some of the factors which led RadNet to make this investment and my remarks will be focused on 3 themes that I would like to expand on. Number one, consolidation of artificial intelligence around cancer screening; number two, technological advances in equipment; and number three, response to the needs for these tools heightened by the recent pandemic. Let me start with the first and perhaps the most important. When RadNet first made its entrance into the artificial intelligence world almost 2 years ago with its acquisition of DeepHealth and the leadership of Dr. Gregory Sorensen, who heads that division, we have for the last several quarters in talking about expanding screening tools for cancer. Our first entrance into this market was out of our own needs internally for RadNet, given the fact that RadNet does almost 1.6 million mammograms a year, which represents about 4% of the entire U.S. market and which we felt was a good investment that would improve both diagnostic accuracy as well as earlier detection of breast cancer. Since that time, we have been diligently searching for other partners whose products and whose strategy would complement that of DeepHealth and breast cancer screening. We were fortunate to find two such companies, both in the Netherlands which have been doing Artificial Intelligent work primarily around Quantib's efforts in prostate cancer and Aidence's efforts primarily in lung cancer. I'm pleased to report that after our own extensive due diligence, we believe that combining these 2 best-of-breed companies for these to cancer screening tools, along with DeepHealth in breast cancer gives us the platform to start a more focused strategy on cancer screening, not just as a diagnostic tool, but for population health much similar to what has already been established for the use of mammography in breast cancer and which is now something that women do on an annual or biannual basis as part of their routine health and wellness. [Audio Gap] efforts in the way of prostate and lung cancer are here today and need to be availed. With what we will do internally with these 3 outstanding teams, meaning DeepHealth, Quantib and Aidence is internally develop our own colon cancer AI screening tools. Putting the 4 of those cancers together, we believe that the screening for almost 70% of all cancers, solid tumor cancers that is, can be detectable earlier with not only artificial intelligence, but new technological advances in equipment. We believe that the opportunity for this is not limited to RadNet, but limited to KORs, health systems, and to governmental agencies that formulate public health policy. And that all of these at some point in time should be tools that much like mammography for breast cancer can be accessed directly and routinely by the general public. The second tenet that I believe is important here is that this is only possible because of additional advances that have occurred principally in the last 5 years on the equipment side. In mammography, we've had advances for digital imaging as well as better resolution on detectors that allow every radiologist to see the cancers earlier. And with CT scanning and MRI scanning, the recent advances have allowed for greater flexibility in the patient experience by significantly scanning -- shortening excuse me, shortening scanning times, primarily in MRI, which now makes the adoption of artificial intelligence and screening tools more accessible and at a lower cost. And I can't emphasize enough how important that is in marrying the artificial intelligence, which is the reading part with the technological advances in the equipment, which actually performs the scan. The third and maybe even the most important part that I want to emphasize here is that the response by us and the equipment is magnified by what the pandemic itself has exposed in the way of challenges to make certain that as these tools become adopted that we can accommodate the needs of the public to provide these invaluable tools. And I'm particularly referring to the staffing shortages, whether it's on the professional side or on the technological side, that I believe are a fact of life that we will be living with for the foreseeable future. The tools that we're developing and hopefully that others are working on will allow the access to equipment and the reliability of these tools to be performed with less dependency on the staffing that traditionally has been necessary to produce these images and their reports. This is a critical juncture for society in general, to make cancer screening a reality. The potential implication of the volumes and demand for these procedures will become self-evident and much like the impact that artificial intelligence and mammography have already begun to have on breast cancer in earlier detection, which both reduces morbidity and mortality, will become even more obvious with the other 3 major cancers that RadNet will be focused on lung, prostate and to be -- soon to be developed, hopefully, colon cancer. So the opportunity here that lies ahead of RadNet and the challenges that come there with are important not for just RadNet as a company, but I believe a response perhaps uniquely by a company that can merge the extensive resources that we have within our imaging centers and the data that we own to help facilitate the development of expanding AI tools in the cancer screening market. And I should add and parenthetically acknowledge that the tools that we will use both for artificial intelligence and the equipment to produce the images and the procedures that we do have other potential implications given the fact that the areas that we scan provide information regarding anatomical and physiological aspects of an individual's health and well-being. The implications for using cancer screening as a tool to look at other potential diseases such as diabetes, cardiovascular and other more chronic diseases should not be overlooked. That information is on the scans that we produce and it generally in the past has never been accessed. But I read articles regularly about how various artificial intelligence developers are looking at ways that all of this information can be used to help both the clinicians, the payors and governments look to manage the health and wellbeing of their populations. So the implications of this are extensive. We're excited to welcome both the Aidence and Quantib members to the RadNet family and combining that with DeepHealth. We not only are the largest artificial intelligence company within radiology. But I think, clearly, the best with this enormous talent that we've brought together. So with that, I'd like just to introduce you to Dr. Sorensen and the leaders, Arthur and Mark-Jan from DeepHealth, Aidence and Quantib to make a brief opening introduction. Hopefully, you'll be hearing from them in addition to other RadNet conferences on a regular basis. So Greg, perhaps if you can start with your comment and then turn it over, and we're ready for that.

Fantastic. Thanks very much. I'd like to just expand briefly on Dr. Berger's comments about why cancer screening with AI is the highest impact and most valuable area of focus for us today. It's just simply our best opportunity to have the biggest impact. And there's both a medical reason and machine learning or AI reason for this. And I'd like to explain both. On the medical front, there is very clear evidence that detecting cancer is early before they spread, saves lives and reduces costs. Because cancers are almost always more treatable at lower cost when we find them before the cancer has spread through the body, advisory bodies in the U.S. and around the world recommend cancer screening. In the U.S., this includes the recommendation screen for breast cancer with mammography, for lung cancer with CT scans of the chest and colon cancer also with CT scans and screening is extremely effective. In the lung, for example, the survival benefit from screening far outweighs any survival benefit from chemotherapy or even new immunotherapies, and it's much less expensive. In short, screening is a major public health benefit and it impacts tens of millions of people in our country and in our markets. On the AI front, modern machine learning techniques are very powerful at learning but mainly at learning what they've been taught, that is what the computer is actually seen examples of before. Modern AI is not as powerful at extrapolating. It best learns from examples. In screening for cancer with imaging, we get literally millions of examples. And so the algorithms can really learn well. We've shown that in some domains, the algorithm can actually do better than human experts at certain image recognition tasks. So to put it succinctly large-scale screening is a great match for modern AI. So when you put the medical value proposition of screening, together with the fact that machine learning is best suited for tasks with millions of examples, it just becomes very clear that cancer screening is where our real opportunity lies. And since, of course, at RadNet, we do literally millions of screening exams here already, we're well positioned not only to develop those tools, but also to implement these tools that will have a positive financial impact on our business and more importantly, a positive health impact on our patients as we roll these technologies out. So those are my opening comments. I'd like to now turn it over to Mark-Jan Harte from Aidence to offer his comments and we've been waiting for Mark-Jan. Take away, Mark-Jan.

Thank you very much, Greg. So from my side, Aidence has been focused on the early detection of lung cancer since we founded it in 2015. And as with many cancers, but especially lung cancer, the survival rate increases dramatically with early detection, as Greg referred to earlier as well. And this is why the U.S. is reimbursing lung cancer screening and also why the NHS in the United Kingdom is, at the moment, running a large-scale screening programs. 40 at-risk population with the intention of scaling this up to a national level. And many other countries in Europe are following the same path because they see the same benefits. And Aidence's strategy is to be the partner of choice for all of these programs. And of course, partnering up with RadNet provides us with a huge opportunity to increase our ability to do that. Furthermore, the treatment of lung cancer with powerful therapies is advancing quickly with companies like AstraZeneca, for example, obtaining approval for the early-stage treatment with their drug, Tagrisso. And since it is crucial to identify the eligible patients early for that treatment, Aidence is a very attractive partner to pharma companies that already have or are planning to bring such early-stage drugs to the market. And that was illustrated very well by the collaboration agreement that Aidence has signed with AstraZeneca in July 2021. So with that, I would like to turn it to Arthur Post Uiterweer from Quantib.

Thank you, Mark-Jan. And let me talk a little bit about Quantib and thanks, everyone, for giving me some time here. So Quantib chose to focus on MRI and prostate as our second clinical application because of 3 main reasons. It's a high-growth procedure, there is a big timing savings potential there, and there is a big potential for quality improvements. The high growth is [indiscernible] by medical associations making MRI a standard procedure prior to every biopsy causing a 5x growth in this domain. Reading these exams is very time consuming, about 20 minutes per patient. While AI has the potential to reduce this to about 3 minutes. And lastly, we observe a large variability in the reading quality between radiologists, which is largely solvable with AI tools. Quantib has created and received CE and FDA clearance for our MRI prostate solution with our first users seeing indeed enormous time savings as well as improved consistency in performance. We now bring this product to RadNet to further improve it using vastly more data that we now have available. This will positively impact the RadNet users as well as our users outside of RadNet. And we are very pleased to be part of this big team and really boost our impact on the market. Thank you for allowing me to make some comments, and I'll pass it back on to Dr. Sorensen.

Thank you, operator. It's Mark Stolper. I think we're ready to begin the question-and-answer portion of today's call.

[Operator Instructions] We'll take our first caller from Brian Tanquilut with Jefferies.

Congratulations to everyone. So I guess just a couple of questions for me. I mean Dr. Berger talked about the opportunity here in expanding testing for the different hard tumor cancer. So how are you thinking, Mark or maybe Dr. Berger in terms of commercializing this, right? I mean what are those conversations like with the payors? And how are they thinking about driving increased screening among your patient population using the tools that you're offering them?

Brian, it's Howard. I think the conversations that have already begun will certainly get amplified as a result of the announcement this week. I believe at every level, whether it's payors who were already talking about these tools or even governments that have a national health program who we've already begun discussions or discussions and relationships are already in place with both Aidence and Quantib. I think that the opportunity here is obviously not just nationally as RadNet has been focused since its inception, but also on an international on a global basis. Part of what needs to be understood and adopted is to create these cancer screening tools that become part of a public health policy much as it has already begun as Mark-Jan pointed out in the U.S. with lung cancer screening and as it has begun and talks are ratcheting up in many of the European countries. I do want to point out that even though lung cancer screening in the U.S. has been approved for reimbursement, only perhaps 10% of the people who buy risk assessment, qualify for lung CT screening are currently accessing the opportunity here. Part of the problem is that it hasn't as yet been made easy enough for people to access this without going through substantial referrals for the procedure as well as authorization. In order to make these tools more effective and universal, health plans and governments are going to have to make it a part of their health plan policies and adopt this into their reimbursement or their coverage much as they have with mammography or breast cancer screening. So that's the challenge that front of not just RadNet, but the marketplace as a whole and how to not only adopt these tools, but then improve compliance so that people understand the value that this can play. Hopefully, what we'll be able to do is develop other tools along with the screening that might give information, as I alluded to earlier about other aspects of patient's well-being and earlier detection of non-cancer diseases as well as how do you evolve the plan design to help encourage the compliance. Part of that problem again is demonstrated by mammography, which is estimated at only 50% or slightly more of all the women who qualify from mammography are getting it on an annual or biannual basis. So part of what we'll be talking about with health plans and government is how do we get the compliance where these tools that are potentially life-saving are better understood and utilized by the public. One of the attractions for the Quantib and Aidence acquisitions was that they are rooted obviously, in Europe, specifically the Netherlands. And most of the European countries have a national health care program. So to the extent that we can speak directly with the public health officials and the leadership in the health care systems in Europe, and they adopt these programs we're essentially bringing in 85% to 90% of the population under these kind of opportunities. So that indeed, in and of itself represents a far different opportunity than here in the United States where the Medicare or CMS covers maybe 20% of the population, and we have to speak individually with all of the other commercial payors in order for each of them to adopt this. So there will be a multipronged attack on these efforts, which both Aidence and Quantib have already established significant commercialization opportunities that will not only facilitate further discussion as we add mammography and eventually colon cancer screening to our tools for Europe. But also as we further adopt these and implement it into our centers and other centers across the country. I want to point out that it's important to remember that RadNet has extensive joint ventures with some of the largest healthcare systems in the country, and the adoption of these tools will be greatly facilitated by those relationships that we have, not only for the centers which are part of our joint ventures, but all of these systems have other hospitals that RadNet currently doesn't have more specific relationships with. So putting all of these tools together and a full court process, if you will, is a commitment that RadNet is making to rolling these out.

No, that makes a lot of sense. And then I guess my follow-up, Mark, as I think about the capital requirements for these businesses going forward. Any color you can share with us on that?

Sure, Brian. So DeepHealth, as you're probably aware, is running at a loss in 2020 and 2021. You know those losses are embedded in the financial results that we've been -- that we've been reporting each quarter. The acquisitions of Aidence and Quantib will increase that loss. And when we issue our guidance in or about March 1 when we report our year-end 2021 results, we'll give you some more transparency as to what those losses will be like. And we're having some discussions internally and with our auditors about further disclosures that we'll be able to make on a quarterly basis that will essentially separate the core imaging business from the financials of the AI divisions, inclusive of DeepHealth, Aidence and Quantib. But for your own kind of thoughts right now, DeepHealth has been running around $4 million to $5 million loss. We expect that to continue and be embedded in our financials. And I would expect that Quantib and Aidence will add to about that loss in the $10 million to $12 million range, and that's reflective of further investments that we're going to be making in those companies and in personnel, especially to try to accelerate the -- both the commercialization of new products to further development of those products as well as increasing the commercial and sales teams of those organizations. So we'll have a lot more to talk about it financially in several weeks.

Next, we'll move on to Sarah James with Barclays.

Congratulations. This is a very exciting strategic move on a lot of different levels. So just to follow up on the last question, as we think about it today, are these 2 new properties going to be revenue accretive for you? And then should they get approval in the U.S.? How do you think about that revenue model working for selling externally or the efficiencies that they could offer you internally? I know you've talked about mammography saving about 25% efficiency. So I'm not sure how these compare.

Well, good morning, and thank you Sarah for joining us. I'm sorry, Rebecca. The opportunity here for commercialization will eventually over or dominate the opportunity for RadNet from a revenue standpoint. The importance on the time savings or the benefit for our radiologist is really twofold. One is not only to reduce the amount of time that it would take to read these scans, but us improving the accuracy, I think as Dr. Sorensen indicated here, all of these tools should be capable of diagnosing cancer earlier. Now by that, I mean, it's not that the radiologists are inadequate. But from an intuitive standpoint, cancer, which theoretically starts with one abnormal cell then proliferating should be detectable from a programming and artificial intelligence standpoint, much earlier than the best of radiologists. So in the ability that we have currently in mammography to separate highly suspicious from nonsuspicious mammograms that has already led to about a 20% improvement in productivity. When it comes to prostate and lung cancer, the tools that we're talking about will reduce the amount of time that a radiologist typically spends circling lesions, identifying them, determining size and density and automate this process, which could reduce their workload by perhaps 50% or more. The backdrop of all of that is that the current shortage of radiologists qualified for these highly specialized tools will allow the current radiologist team to read these scans faster and more accurately and ultimately produce benefit to RadNet from an operational standpoint. So I think your question is well designed in that eventually, we hope within a couple of years, the AI division will actually become cash flow neutral for us as we expand to markets well beyond RadNet centers. And during this interval, we will get significant improvement from an operational standpoint on the efficiency with which both our centers are capable of doing the exams as well as our radiologists in improving their reading and their accuracy.

Got it. Great. And can you give us any guidance on what the FDA approval process looks like, what a reasonable time frame would be for you guys to hear back on that?

Greg, why don't you respond to that?

Sure. I'd be happy to. Thanks for the question. Both Aidence and DeepHealth have made submissions to the FDA in the last month or 2, and the FDA typically has its own internal 90-day clock that under MDUFA they promised to essentially get you an answer within -- or an answer to 95% of the companies within 90 days. But of course, that clock can stop and it usually does during kind of a month or 2 long Q&A period. So I think realistically, we can't expect the FDA to do anything this quarter. But I would say that probably in the next 2 quarters, it would be highly likely that we'll not only have substantive engagement with the agency, but -- and of course, this is modulated by the COVID slowdown that so many groups in the bureaucracy have been forced to deal with. But assuming that, that doesn't get worse, I think sometime in Q2 or Q3 is when we expect to see approval of these products.

Fantastic. And last question, how does Quantib MB's prostate product work with what you were already developing into the prostate market?

I think you were referencing or go ahead. Go ahead, Howard. You might have probably place, but a follow-up on it.

And then I'll expand just necessary.

Okay. Yes. When DeepHealth first became part of the RadNet family, we already had the vision that the prostate would be an area of interest for us as we described in the press release and as Dr. Berger pointed out in the initial part of the call, although prostate cancer does not have a screening paradigm the way with imaging, the way these other modalities do. There's -- as Arthur said, there's very good evidence that if you've got cancer from -- or suspected to have cancer for some of the reasons and are going to undergo biopsy, imaging is playing a critical role. And we think it's a matter of the science continuing develop -- to develop, that will eventually there will be a role in certain high-risk populations for prostate cancer screening. So that's why 2 years ago, we already knew prostate was important. What we get with Quantib is someone who's been working on this for more than 5 years and already has a big customer base and a lot of experience. And so it essentially lets us leapfrog what the efforts we were going to do and now is part of the attraction for us for acquiring this company and their team, very talented team.

Maybe if I just amplify a little bit on that, Greg. Once we acquired DeepHealth and its mammography AI tools, the decision to expand into other significant cancers, namely prostate, lung and colon became a matter of buy or build, if you will. And the decision at that time since we weren't at all invested in prostate or lung tools was to look for a good acquisition opportunity or opportunities in this case, to rather than build it internally ourselves. The turnaround time for developing a product and submitting it for approval to the FDA or CE mark in Europe is probably close to 2 years between all the testing that you do before you submit to the governmental agencies and then the time that they take to finally give you approval. So we felt that the urgency for us to take advantage of the platform that we established with DeepHealth, and rather than invest the capital to build these to find best of breed in at least 2 of the other categories of cancer related to prostate and lung was a better way for us to invest our money and time to buy rather than build. The fourth leg of this stool for colon is something at this time that we have not found anybody else doing at -- with the type of screening tools that we think are capable today with both AI as well as technology and equipment. And so that one, we're beginning to develop a strategy to develop ourselves. So really, the leap into the acquisitions did not augment any development that we had internally for cancers other than what we were already deeply involved with breast cancer.

That makes more sense.

[Operator Instructions] And next, we'll take our question from Mitra Ramgopal with Sidoti.

Just a couple for me. I was just curious if you could provide a little more color on the acquisitions themselves in terms of how the process was initiated and sort of how these 2 companies showed up on your radar given that they're both outside of the U.S. and historically, the acquisitions tended to be domestic. And also aside from expanding the AI platform, was there a determination you also wanted to expand your geographical presence?

Mitra, thank you. The world of artificial intelligence and the players inside of that is rather small even on a global basis. So both of these companies have been around for 7 to 10 years. And so they were fairly well known in the industry as to their efforts in both of these cancer initiatives. So as we screen the available market, or acquisition opportunities, they quickly popped up on our radar. And then along with our physician advisers internally, they helped us look at their current products and give, so to speak, a green light as to the quality of these. I'm happy that in both cases, our physician advisers were very enthusiastic about not only the quality of these tools from a diagnostic standpoint, but from the relative ease of operation that will allow them greater efficiency. So the universe of people out there is relatively small, and we began the process probably about 7 months ago, back in maybe the summer around the August time frame, where we met with the companies and began discussions about acquiring them. Both of the companies, as you may know, were venture capital back. So that, along with the fact that they are domiciled in the Netherlands where both the financial as well as legal structure can be substantially different than here in the U.S. Maybe a bit more of a challenge, but the challenge was actually beneficial to the company because we got to know the individuals more carefully. And I'm happy and perhaps most pleased to report that the cultural fit with these organizations both as demonstrated by their leadership with Mark and Arthur as well as their Chief Technology Officers and development teams and commercial teams are at the highest levels that we have seen. So for us, it was an easy decision to make, but a little bit more difficult to execute. So that's the breadth of that. What was the second part of your question, Mitra?

Yes, Howard, the geographic presence expanding outside of the U.S.

Yes. I think the geographic opportunities really will help further commercialize not only our artificial intelligence initiatives but also our eRAD information technology tool. We plan on integrating these tools to be far more effective together than either is on a stand-alone basis. And we think that, that will be a substantial offering and attraction in not only outside of the U.S. but inside the U.S., where RadNet doesn't own center. So this opportunity ultimately will be what we call capital light in the sense that it will not be heavily investing in equipment, but looking at ways that we essentially will transform a part of our business to be a tech company. So part of what Mark has talked about in terms of reporting the results of AI separately from the operations and other opportunities that we have to heighten the potential value opportunity of artificial intelligence and IT platforms will become a little bit more evident. But we do plan on expanding anywhere where these tools can provide benefit to the population, whether it's internationally or nationally.

And we have no further questions at this team. I'd like to turn the call back over to your presenters today for any additional or closing remarks.

Thank you, operator, and thank you all for joining us. This is a very important and potentially game-changing opportunity here for everybody. And I think we are making a very bold statement as to where RadNet can contribute that only, as Mark mentioned, to our current stakeholders, but stakeholders anywhere that want to try to provide better health and better outcomes for its population. We expect more to come from us as we evolve this opportunity, but recognize that RadNet not just as a company but as a societal opportunity is committed to better health. And as I have said in previous conference calls, good medicine is good business. With that, we look forward to our first quarter -- excuse me, fourth quarter earnings call in early March and further updating you on our opportunities. Thank you all, and stay safe.

That does conclude today's teleconference. We do appreciate your participation. You may now disconnect.