[00:00:00] Speaker 00: Good morning. [00:00:01] Speaker 00: Our next case for this morning is a model LLC versus new ones. [00:00:05] Speaker 00: Communications ain't um, counselor bell. [00:00:10] Speaker 00: You requested to reserve two minutes of rebuttal time and maybe we. [00:00:17] Speaker 00: Thank you, your honor. [00:00:18] Speaker 03: Good morning and may it please the court under any reasonable interpretation. [00:00:23] Speaker 03: The Tyra prior art teaches the sole limitation at issue in this appeal. [00:00:28] Speaker 03: identifying latent information required for generation of a particular report. [00:00:33] Speaker 03: And there's really one dispositive and I submit uncontested fact that shows that this is true. [00:00:39] Speaker 03: TIRA teaches generating reports that must include at least one necessary piece of information and that is the head or topic of the TIRA reports such as a medical problem. [00:00:50] Speaker 03: And that's exactly the same type of required latent information that the patent describes. [00:00:55] Speaker 03: for example, medical problems. [00:00:56] Speaker 03: And that requirement in TIRA is established in advance based on expert-defined models that target that specific information. [00:01:04] Speaker 03: And that's at pages A, 1702 and 1706 of the appendix. [00:01:09] Speaker 03: Therefore, we submit even under nuances improperly limited view of what it means to be required information, TIRA teaches that disputed limitation. [00:01:19] Speaker 03: And that theory we further submit was fairly and clearly spelled out in the petition. [00:01:24] Speaker 03: The petition states that Tyra uses, quote, expert-defined models to target specific information. [00:01:30] Speaker 03: What are you reading from now, Mr. Bell? [00:01:33] Speaker 03: And that is at page A, 1046, the petition, at pages 1046 and 47 of the appendix. [00:01:41] Speaker 03: And there the petition is quoting Tyra and says, Mr. Bell? [00:01:47] Speaker 04: Mr. Bell, this is Judge Stolls. [00:01:49] Speaker 04: So it's your view that here at 1046, where [00:01:53] Speaker 04: It talks about targeting specific information, but that means it's required. [00:01:57] Speaker 04: I mean, I think it's, you know, as I understand the difference, first of all, can I just ask you this? [00:02:06] Speaker 04: Is the difference between what you're asserting and what the board is saying is that the board seems to take the position that TARA creates the form based on information that is found [00:02:21] Speaker 04: you could create different reports based on the information found. [00:02:25] Speaker 04: While the patent and the claim is understood by the board is that you're going out and you're seeking specific information required by a form. [00:02:33] Speaker 04: Is that how, just stepping back, is that how you think the board understood Terra to work and that was the difference that the board saw? [00:02:43] Speaker 03: I think that fairly encompasses the difference between the board's view and our view and also between Nuance's view [00:02:51] Speaker 03: in our view, yes, Your Honor, that the board doesn't allow for the kind of one-off day of example of a report that somebody asks for certain information. [00:03:03] Speaker 03: So, for example, in TARA, if a doctor steps up and says, I want a report on anomaly, that is the required information because the report that's generated, if there is one, must have information about the anomalies, in particular, the head or topic. [00:03:20] Speaker 04: which I don't think that's been... Okay. [00:03:22] Speaker 04: I understand what you're saying. [00:03:23] Speaker 04: Now, can I ask you something else? [00:03:25] Speaker 04: If we turn back to, like, page 1046, I think it's 1046 through 48, I don't see that specific explanation here in the petition. [00:03:33] Speaker 04: But what do you think is your best... What would you point to as being your best support for your position? [00:03:40] Speaker 04: And, by the way, are you relying on Tara alone on appeal, which it seems that you are? [00:03:48] Speaker 04: Or are you relying on also TARA in view of Buchanan as teaching the required element? [00:03:55] Speaker 03: On appeal, Your Honor, we focused on TARA because we think that is sufficient. [00:03:59] Speaker 03: And in those pages that Your Honor identified, 1046 to 48, I would first note that this all comes under the heading of required information. [00:04:08] Speaker 03: So it recites the entire limitation. [00:04:10] Speaker 03: So all of the discussion is in that context, and then it [00:04:14] Speaker 03: proceeds through there and explains why the required information in TIRA, and it doesn't use the word required again until the end, and I'll come to that, but it talks about targeting the expert defined models with specific information. [00:04:29] Speaker 03: And then on 1048, and I should note that those expert models are defined in advance. [00:04:35] Speaker 03: You know in advance what type of information you're looking for, and you need that information to generate a report. [00:04:40] Speaker 04: What is your position that you needed to explain that more in the petition? [00:04:45] Speaker 04: I mean, what if, you know, like the board were to say, well, you needed to explain that. [00:04:49] Speaker 04: You needed to say, like, when you're talking about targeting specific information or important medical information, you have to say that that means it's required. [00:05:00] Speaker 03: So I would point, Your Honor, to page 1048, where it's the conclusion of this discussion. [00:05:06] Speaker 03: And it talks about how [00:05:08] Speaker 03: both Tyra and Buchanan, but let's focus on Tyra, acknowledge that they include latent information required for generating a particular report and then it says why? [00:05:17] Speaker 03: Because unstructured reports, unstructured dictation reports could not be submitted to insurance companies or regulatory agencies. [00:05:25] Speaker 03: So there again it's saying that there is certain information that must appear in these structured reports in Tyra, this latent information, otherwise it wouldn't be useful. [00:05:34] Speaker 04: And then if you trace through the steps... Can I step back and ask you another question about this? [00:05:39] Speaker 04: I see what you're talking about where it says a POSA would have been motivated. [00:05:43] Speaker 04: I think the board said this wasn't sufficient because it was part of your motivation statement. [00:05:50] Speaker 04: And I understand your response to be, well, it's not just a motivation statement. [00:05:54] Speaker 04: It's also talking about combining Terra and Buchanan to include latent information required for generating a report. [00:06:02] Speaker 04: Is that right? [00:06:03] Speaker 03: Precisely. [00:06:05] Speaker 04: Okay. [00:06:05] Speaker 04: Now why, what about this? [00:06:07] Speaker 04: Now it doesn't say Terra alone, it says Terra and Buchanan. [00:06:11] Speaker 04: So how do you respond to that? [00:06:13] Speaker 03: Well, it also says each of Terra and Buchanan, right? [00:06:17] Speaker 03: It says both references acknowledge that there would be this problem with unstructured information and therefore for certain purposes like insurance, you would want to structure it in a certain way, which would entail, given the context of these three pages discussion, [00:06:32] Speaker 03: What we're getting at is what's the required information for inclusion in these reports. [00:06:37] Speaker 03: And then that same required information, that latent information, you see it again throughout the steps of the patent as we walk through in the petition subsequently. [00:06:46] Speaker 01: Mr. Bell, can I just double check something? [00:06:49] Speaker 01: The sentence that we've just been focusing on that begins on 1047 and carries over to 1048 has two citations behind it. [00:06:57] Speaker 01: One is to Buchanan. [00:06:59] Speaker 01: The other, I think, is maybe to your expert, is it Baygee or something, which is not in the Joint Appendix, but there's no citation there to Tara or Tyra. [00:07:08] Speaker 01: That's correct. [00:07:09] Speaker 03: That's correct. [00:07:11] Speaker 03: So the expert in his declaration made the same point, that both of them acknowledge that there's problems with unstructured information, and therefore you would want to structure it in a certain way for other purposes. [00:07:24] Speaker 01: And can you get back to the thing that you started the argument about, about something about a heading? [00:07:30] Speaker 03: Yes. [00:07:31] Speaker 03: So the key piece of information, and we made this point in our opening brief at pages 41 and 42, and then again in our reply when we submit it wasn't contested in the red brief, the head or the target of TARA. [00:07:44] Speaker 03: And you can see this, for example, at A1701, where figure one depicts a sample [00:07:50] Speaker 03: frame of Terra, and it includes the head or topic. [00:07:56] Speaker 03: And that is a required piece of information. [00:07:58] Speaker 01: And just to be clear, is that head latent information sifted out of the data stream? [00:08:11] Speaker 01: Yes. [00:08:13] Speaker 01: How do we know that? [00:08:15] Speaker 03: Because the topic in that example is a math. [00:08:19] Speaker 03: an abnormal mass. [00:08:21] Speaker 03: And so what Tara says is we need to extract that information out of the incoming information. [00:08:29] Speaker 01: This may be too kind of picky, but isn't the information that is being extracted out of, say, the doctor's, you know, orally recorded report, you know, some number like, you know, five grams, [00:08:47] Speaker 01: but not the word mass as a topic? [00:08:52] Speaker 03: No, I respectfully disagree, Your Honor. [00:08:55] Speaker 03: First you get the topic, and then you know what else you need to look for. [00:08:59] Speaker 03: And so that's why we say without that topic, now it could be one of three, and Tara says on page A1706, and I'll just quote from it, says each frame, each frame has two things, a specific topic, e.g. [00:09:13] Speaker 03: a mass, together with descriptions of select property. [00:09:16] Speaker 04: So it's extracting all of that. [00:09:18] Speaker 04: Mr. Bell, this is Judge Stoll. [00:09:20] Speaker 04: Um, I'm going to be able to follow you better and really have more comfort in your argument. [00:09:25] Speaker 04: If you point to me where these sentences are that you're relying on and just slow down a little bit, because this is probably the most important part of your argument. [00:09:34] Speaker 04: So if you could please do that, point out where you're relying on on 1706 and also you had referred to 1701. [00:09:41] Speaker 04: But if you could point out specifically what you're pointing to, I'll follow you better. [00:09:45] Speaker 04: Thank you. [00:09:47] Speaker 03: Absolutely, Your Honor, of course. [00:09:48] Speaker 03: I will start with A1706, and I'm looking at the right-hand column, three lines down under the frame constructor heading, where Tyra says, each frame represents knowledge about a specific topic, e.g. [00:10:03] Speaker 03: math, that's that abnormal finding we discussed, together with descriptions of select properties. [00:10:10] Speaker 03: So that sentence right there, and that's a sentence we quoted in the petition, [00:10:15] Speaker 03: That's a sentence that the board itself acknowledged. [00:10:20] Speaker 03: And I don't think it's disputed that to have a report in TERA, you must, at a minimum, have a certain piece of required information, i.e., that topic, which can be one of three things. [00:10:31] Speaker 03: And I'm, again, looking at A, 1706, the fifth lines on down. [00:10:36] Speaker 01: And just to be clear, the frame is something that's in the report. [00:10:41] Speaker 01: It's essentially a template category. [00:10:46] Speaker 01: That's correct. [00:10:46] Speaker 03: The frame then leads to the structured report in TARA. [00:10:51] Speaker 04: And so anything that you see. [00:10:53] Speaker 04: It's shown in Figure 5. [00:10:55] Speaker 04: Figure 5 is a sample frame, right? [00:10:58] Speaker 04: That is correct. [00:10:59] Speaker 04: OK. [00:11:01] Speaker 04: And you had said earlier that you first have to pick the topic. [00:11:05] Speaker 04: And you cited page 1701. [00:11:07] Speaker 04: Was there something specific on 1701 that you think supported that? [00:11:14] Speaker 03: It was at, I think it was, maybe I misspoke, I think it was perhaps 1702. [00:11:20] Speaker 04: I might have missed, I just misunderstood, I might have misremembered, but please yeah, we're on 1702. [00:11:25] Speaker 03: 1702 talks about creating this expert defined report and I'll point you to the first column. [00:11:33] Speaker 03: It's under that diagram, one, two, three, four, five, six lines down. [00:11:37] Speaker 03: And so it says that the task here first, [00:11:40] Speaker 03: is creating an expert-defined data model of the targeted information that is to be contained within the report. [00:11:48] Speaker 03: So that tells us that the expert-defined model is something done in advance. [00:11:52] Speaker 03: And this is why we submit that even under a narrower view of what is required, Tyra definitively teaches it for exactly that reason. [00:12:01] Speaker 03: There is this predefined, called institutional advanced definition of what you are looking for, and then Tyra goes on [00:12:08] Speaker 03: to find it. [00:12:09] Speaker 03: And I should note that the dispute here on claim construction is an alternative way that the court could resolve this, because we think that the boards and nuances definitions are far too narrow of required. [00:12:21] Speaker 03: Consider the following situation. [00:12:24] Speaker 03: If I could finish the sentence. [00:12:28] Speaker 00: Yes, you may. [00:12:29] Speaker 03: Thank you, Your Honor. [00:12:30] Speaker 03: Consider the following situation. [00:12:31] Speaker 03: A judge asks a clerk for a report on the most recent cases citing KSR or Chevron. [00:12:39] Speaker 03: I think any clerk you ask would say that that is required information for generation of that report, but that's the type of day of one-off, Dr. Axe asks for an X report that the nuanced definition and the board's definition would improperly exclude. [00:12:56] Speaker 01: Mr. Bell, what if the judge says, please give me a report on any recent KSR siting case? [00:13:08] Speaker 01: then there is no particular and the clerk being diligent generates a report with 37 such cases, no one of which actually was required by the judge's request. [00:13:25] Speaker 01: I kind of think that's sort of what the board was saying here. [00:13:30] Speaker 03: I agree, Your Honor. [00:13:31] Speaker 03: I think that is what the board said and I think that's wrong. [00:13:34] Speaker 03: If you look at the specification, [00:13:36] Speaker 03: let alone the claims that don't provide any sort of restrictions on what is required. [00:13:41] Speaker 03: But you look at the specification, it says any reports can be generated. [00:13:45] Speaker 03: That's at A53, column 10, line 62 to 64. [00:13:49] Speaker 03: It also says you can have user-defined reports. [00:13:53] Speaker 03: In other words, the user gets to decide what's the required information. [00:13:56] Speaker 03: So whether it's a judge saying any case, in which case that's the required information. [00:14:00] Speaker 03: You have to have at least one case. [00:14:02] Speaker 03: That's a requirement. [00:14:03] Speaker 03: Or a doctor saying, please give me a report on all of the anomalies identified in my free text narration from the last week. [00:14:12] Speaker 03: Therefore, those anomalies are required. [00:14:14] Speaker 03: And we pointed that out in our brief at 33 and 34. [00:14:18] Speaker 03: And Nuance's response to it I think is revealing at page 54. [00:14:22] Speaker 03: The only response Nuance has to that hypothetical [00:14:26] Speaker 03: is that, no, that doesn't work because you need, quote, specific information predefined as being required for accurate and complete reporting on a type of encounter. [00:14:36] Speaker 03: And that requirement is not part of the word required. [00:14:40] Speaker 03: It's not found in the claims. [00:14:41] Speaker 03: And we submit it's refuted by the specification. [00:14:44] Speaker 03: With that, I'll reserve the rebuttal of my time, unless the court has questions. [00:14:49] Speaker 00: OK. [00:14:49] Speaker 00: We thank you, Mr. Bell. [00:14:52] Speaker 00: Let's hear from Counselor Speed. [00:14:55] Speaker 02: Good morning, Your Honors. [00:14:56] Speaker 02: Nathan Street on behalf of the Appellee of Nuance Communications. [00:14:59] Speaker 02: The board's final written decision should be affirmed because the board correctly found that EMOTO's petition failed to provide a detailed explanation of how or why the tire reference discloses this relevant limitation that we've been discussing today. [00:15:14] Speaker 02: This flaw in the petition was fatal below. [00:15:16] Speaker 02: It should be fatal on appeal as well. [00:15:18] Speaker 02: The relevant limitation that we're talking about is the identifying latent information limitation. [00:15:25] Speaker 02: And it recites in full, processing the input data stream to identify latent information within the data stream that is required for generation of a particular report. [00:15:35] Speaker 02: And Modal's petition addresses this. [00:15:37] Speaker 04: Hi. [00:15:38] Speaker 04: This is Judge Stoll. [00:15:40] Speaker 04: Why do you think that the language, for example, at page 8, 7, over and under 10, isn't sufficient? [00:15:50] Speaker 04: the sentences opposed it would have been motivated to combine Terambican and to identify information, comma, including latent information required for generating a particular report, because both references acknowledge that reports could not be submitted to insurance companies or regulatory agents or unstructured dictation. [00:16:11] Speaker 04: So why isn't that enough? [00:16:15] Speaker 04: when the difference is whether someone's decided what the information, you know, whether information's required or just important, why isn't this sentence here enough for that claim element? [00:16:28] Speaker 02: Hey, Your Honor. [00:16:29] Speaker 02: That paragraph, the entire paragraph there is devoted to addressing a weakness in the Tire reference that EmModel was hoping to prebut in their petition. [00:16:38] Speaker 02: The argument that that paragraph in its entirety is presenting is [00:16:42] Speaker 02: should the board conclude that Tyra's structured frame, which is what Imodal pointed to as the claims report, should the board conclude that that structured frame is not actually a report as required in the claim, it would have been obvious to take that structured frame and use it as a document template such as that just purportedly disclosed in Buchanan. [00:17:02] Speaker 02: There's nothing in that paragraph that provides the critical detailed explanation as to how or why any information in Tyra [00:17:12] Speaker 02: is required for generating a report, whether that report be in a certain frame. [00:17:16] Speaker 04: I understand your argument on that. [00:17:18] Speaker 04: I understand it. [00:17:20] Speaker 04: One thing that's puzzling me is that when I read the board's opinion, when it says the petition's deficient, it's not just saying it's deficient for purposes of showing whether TYRA discloses this element. [00:17:34] Speaker 04: It says it's also different on whether TYRA, in view of CAN, discloses this claim element. [00:17:40] Speaker 04: That's, for example, on page 8-8, and I believe it's also on page 8-22. [00:17:45] Speaker 04: So, I mean, what is your response to that? [00:17:49] Speaker 04: Just, I mean, what is your response to that? [00:17:54] Speaker 02: Yes, thank you, Your Honor. [00:17:56] Speaker 02: The issue is even if there's no, there's simply no explanation of how the sub-frame property, which is what Emoto pointed to in his petition as the latent information, how [00:18:07] Speaker 02: those properties would be required. [00:18:10] Speaker 04: Let me ask my question more specifically. [00:18:13] Speaker 04: Do you think the board erred when it said that this sentence doesn't provide an explanation for why Terra in view of the canon teaches the required element? [00:18:25] Speaker 02: No, I don't think that the board erred in this regard. [00:18:28] Speaker 02: I think that the board [00:18:30] Speaker 02: properly read this paragraph as setting forth a motivation for combining Tyra in light of Buchanan, which Enmodo has concluded they no longer are relying on on appeal, a motivation for combining those two references so that you would take the property frames in Tyra and somehow use them in the document template of Buchanan because that document template would have been a report as claimed. [00:18:54] Speaker 02: There's still nothing in that analysis that Enmodo provided that explains why [00:18:59] Speaker 02: Any particular property subframe entire so say the existence or location of a mass live that information would be required for this document template and you can there's no as you're on it noted earlier, there's no citation attire in here. [00:19:14] Speaker 02: to suggest that TIRRA would have motivated that or would have led a pose to somehow turn the information in TIRRA and make it from optional, which is what it is in TIRRA, and somehow make it required simply with a document template if we can. [00:19:28] Speaker 02: There's no explanation. [00:19:29] Speaker 04: What about the bot quote that Mr. Bell was referring to on page A1047 where it's saying talking task of automatically structuring radiology reports and it says the first [00:19:43] Speaker 04: step is to create expert-defined data models of the targeted information contained within the report. [00:19:49] Speaker 04: And he's saying that what this means is that basically, as I understand it, is that you're picking your topic. [00:19:58] Speaker 04: And so because you're picking your topic, then you should know the information that would have to be included for that topic. [00:20:09] Speaker 02: suggested if we turn to the 1046, the paragraph before that, which is the first paragraph in the petition that EMOTO uses to identify what in TIRA is the identified latent information. [00:20:21] Speaker 02: And they point to, they quote from TIRA, and the quote is, the existence, properties, location, and diagnostic interpretation. [00:20:28] Speaker 02: And those properties are the sub-frame properties. [00:20:31] Speaker 02: And so they were, EMOTA was clearly pointing to the sub-frame properties, not to the header of the topic that they point to in their reply brief. [00:20:38] Speaker 04: And you don't need to take my... So you don't think that it's enough that they quoted this block, quoted this particular paragraph? [00:20:50] Speaker 04: Correct. [00:20:51] Speaker 04: Correct. [00:20:54] Speaker 04: to specifically explain what they meant by it. [00:20:58] Speaker 02: Is that right? [00:21:01] Speaker 02: Yes, you're right. [00:21:02] Speaker 02: To the extent that there was any explanation as to what that block quote was going to mean, it's the paragraph, the prose that they wrote leading into it. [00:21:10] Speaker 02: And that prose, as they concede in their own reply group at page eight to nine, they point to this exact paragraph on 1046, they point to the exact quote of existence, properties, location, and diagnostic interpretation, [00:21:22] Speaker 02: And they say that that is a reference to the subframe of Tira, not to its top of her head, which, sorry. [00:21:29] Speaker 04: When I read a patent, I'm supposed to read it from the perspective of a person of ordinary skill in the art, right? [00:21:35] Speaker 04: When I read a petition, what perspective should I read that from? [00:21:38] Speaker 04: Is there a perspective? [00:21:44] Speaker 02: I don't. [00:21:46] Speaker 02: I suspect that it would be from a reasonable reader with some understanding of how petitions are supposed to be put together to comply with the board's rules. [00:22:00] Speaker 02: I think that your question kind of gets to the reason why this is an abusive discretion standard. [00:22:05] Speaker 02: This is the board applying its own rules and saying, this petition is not sufficient. [00:22:12] Speaker 02: It did not provide us, as the board said, [00:22:14] Speaker 02: a coherent explanation of how any information they identified is actually required for the report that they identified. [00:22:23] Speaker 02: It's somewhat similar to the case, the Garmin versus Log Logan Tree case that we studied at page 38 of our opening brief. [00:22:32] Speaker 02: In that case, you had the reference, the challenge claim we said is a microprocessor capable of doing one or two things. [00:22:40] Speaker 02: And the petition identified a microprocessor in the prior and then it identified disclosures that they alleged did the one or two things. [00:22:48] Speaker 02: And the board ultimately said, that's not enough because you haven't explained to us either by a reference, a citation to the reference through our argument of inherent anticipation or through an obviousness challenge as to how or why we should believe that that microprocessor is the thing in the system that actually does the one or the two processes that were claimed. [00:23:08] Speaker 02: And this court affirmed [00:23:09] Speaker 02: and pointed out that a petitioner has the burden before the board and that they can't satisfy that burden with mere implication. [00:23:16] Speaker 02: And that's exactly what we have here. [00:23:18] Speaker 02: At best, a modal has mere implication. [00:23:20] Speaker 02: That's just because they pointed to the subframe properties. [00:23:24] Speaker 02: And again, I just want to emphasize this idea that they pointed to the topic or the head. [00:23:27] Speaker 02: There's nothing in pages 40 to 42 to suggest that they would have or that they did. [00:23:33] Speaker 02: They pointed to the information and then they pointed to the report and they left it to implication that that information was required for that report. [00:23:43] Speaker 02: And we would respectfully submit that in a scenario like this where you have the required for language was actually added during prosecution so that the claim went from saying simply identifying latent information to identifying latent information required for a report. [00:23:58] Speaker 02: That they're leaving that critical limitation to implication [00:24:01] Speaker 02: is insufficient as the board would have been proper to find it would be insufficient to carry their burden of proof. [00:24:08] Speaker 02: And it was certainly not an abuse of discretion for them to conclude that it fails to comply with their rules that require a detailed explanation of how and why the prior art meets the claim limitation. [00:24:21] Speaker 02: On the claim construction issue, just briefly. [00:24:25] Speaker 01: Can you just stay on this for one more minute? [00:24:29] Speaker 01: And I realize we've been going over [00:24:32] Speaker 01: different aspects of the same ground. [00:24:34] Speaker 01: But if one, and I'm, I guess, returning to where Mr. Bell started about this idea of a header, isn't the sentence that starts at 1047 and goes to 1048, the one that refers to Tyra and Buchanan, in the use of the term unstructured, [00:25:01] Speaker 01: dictation or reports are really not satisfactory. [00:25:05] Speaker 01: Doesn't that make clear that there has to be a structure which, and it's a little hard to see how there's a structure without a header? [00:25:16] Speaker 02: I think you could certainly have structure without a header per se. [00:25:25] Speaker 02: So if you have, the analogy that I find most useful is [00:25:29] Speaker 02: thinking of the form that you submit when you're going to pay for some clothes you purchased online. [00:25:36] Speaker 02: That form has many fields, and so every field could be entirely optional. [00:25:42] Speaker 02: Typically, the fields have little stars next to them to indicate when particular information is required before you can submit the report, but you could theoretically have a field that has no little stars, and so whatever information you happen to put into the billing field would get generated in the report, and it would be structured at the back end. [00:26:01] Speaker 02: So if you just put in your middle name and your apartment number and you hit submit, it would be structured in the sense that you have a billing report that identifies those two, [00:26:10] Speaker 02: but no one would say that that information was required to generate that report in the first instance. [00:26:16] Speaker 01: And at least related in my mind, though I'm not quite sure how related, putting aside what's in these pages on the petition and just looking at Tyra, where it says mass in, what is this, I guess the figure one on page 1701, and I think that's, [00:26:40] Speaker 01: the same thing being referred to as the head or topic in referring to the frame. [00:26:47] Speaker 01: Is that word a piece of what the petition identified as the latent information in Tyra? [00:27:01] Speaker 02: I think, again, turning to pages 40 to 42 of the petition, I do not see any [00:27:07] Speaker 02: suggestion that they were relying specifically and separately on them. [00:27:10] Speaker 02: Right. [00:27:10] Speaker 01: I think I asked the question incorrectly. [00:27:12] Speaker 01: Forget about the petition. [00:27:16] Speaker 01: There is some, except in so far as follows, the latent information that is being sifted out of, I guess, the dictated radiology reports in Tyra, does that information include [00:27:37] Speaker 01: the word mass when the information is about the mass of some body part? [00:27:48] Speaker 02: Your Honor, the way that TIRER works is it takes a report and it structures every single sentence in that report. [00:27:55] Speaker 02: And so as TIRER explains at 17, appendix site 1705, [00:28:06] Speaker 02: that the head is just the topic of the sentence. [00:28:09] Speaker 02: And so each sentence has a topic, but what's identified is important. [00:28:16] Speaker 01: And is that topic actually drawn always from a word in the sentence? [00:28:22] Speaker 01: Or is it somehow identified by the computer that even though the word mass is nowhere in the sentence, we know it's about mass. [00:28:31] Speaker 01: And so we're going to call the topic mass. [00:28:34] Speaker 02: So in the examples provided in Tyra, it's always a word that's actually in the sentence. [00:28:41] Speaker 02: And I think that gets to the question of whether or not that is latent information. [00:28:46] Speaker 02: And this goes, the reason, and modal campsite expert testimony on it, and I can't point your arm to it either, is that this just simply wasn't an argument that they presented before the board. [00:28:56] Speaker 02: I've shown you in pages 40 to 42 that this argument is not detailed in the petition. [00:29:01] Speaker 02: And I think it's important to understand that in the reply paper, [00:29:05] Speaker 01: When they offered an alternative argument for how higher... Can you give me a JA site, what you're about to refer to? [00:29:13] Speaker 02: The board's decision at 25 would be the site I was going to refer to. [00:29:18] Speaker 01: Yeah, but you were referring to the reply paper, so what's the JA site? [00:29:22] Speaker 02: I believe it's 1278. [00:29:24] Speaker 02: I can get it for you one second. [00:29:26] Speaker 01: Well, it starts at 1281, I think. [00:29:29] Speaker 01: But you must be thinking about a particular page. [00:29:33] Speaker 02: That's 1298, Your Honor. [00:29:35] Speaker 02: Thank you. [00:29:36] Speaker 02: And at 1298, what we'll see is that they point to a speculated mass report in TIR, which is a subsequently generated report that TIR references in its disclosure. [00:29:50] Speaker 02: They pointed to that as the report for which the latent information would have been required. [00:29:56] Speaker 02: So before the board, when they wanted to articulate a theory under our allegedly narrowed construction, [00:30:02] Speaker 02: They pointed to, they didn't point to the header topic. [00:30:05] Speaker 02: They pointed to the speculated match report. [00:30:07] Speaker 02: And at page 25 of the decision, the board found that argument to have been waived. [00:30:13] Speaker 00: Okay. [00:30:13] Speaker 00: Any other questions from my colleagues? [00:30:16] Speaker 01: No, thanks. [00:30:17] Speaker 00: All right. [00:30:18] Speaker 00: Thank you. [00:30:20] Speaker 00: Yes. [00:30:21] Speaker 00: Let's hear back from Mr. Bell. [00:30:24] Speaker 00: If you have three minutes to rebuttal. [00:30:26] Speaker 03: Thank you, your honor. [00:30:28] Speaker 03: I'd like to start with the last topic of discussion, and if I could point the court to A, 1056. [00:30:34] Speaker 03: This will answer, Judge Strana, your question about whether, in the petition, the topic was ever identified as part of the latent information. [00:30:42] Speaker 03: At the top of 1056, and this is in one of the subsequent steps, or one of the steps that the claims use to identify the latent information, and this is the top of 1056, it says, identify latent information, specifically the class, [00:30:57] Speaker 03: topic or property. [00:30:59] Speaker 03: So that is latent identified information. [00:31:03] Speaker 03: Likewise, in Tyra itself, I would point the court to A1702 and about 10 lines down the second column, it's talking about parsing a particular sentence in the input. [00:31:16] Speaker 03: For example, consider the following sentence. [00:31:18] Speaker 03: A mass is seen in the right, lower, lobe, et cetera, et cetera. [00:31:22] Speaker 03: That is the sentence that's parsed and it says right after that, note that the frame consists of a head or topic and set of property. [00:31:32] Speaker 03: So that head or topic, in this case, example, a mask, is required information. [00:31:38] Speaker 03: I don't think there's any doubt about that. [00:31:41] Speaker 03: And that required information is what's identified as the latent information throughout the petition at the pages we've discussed at length. [00:31:50] Speaker 03: including at 1047 where it talks about a detected mass, at 1056, which I just referenced, where it said the topic could be a mass, and also at other pages throughout. [00:32:01] Speaker 03: So we think that there's really no doubt that there is at least one piece of required information for Tyra's report. [00:32:08] Speaker 03: To put it somewhat differently, if you end up with a report in Tyra, it must have at least something filled in [00:32:15] Speaker 03: in that topic area, whether it's an abnormal medical finding or a procedure or something of the like, that is required. [00:32:24] Speaker 01: Now. [00:32:25] Speaker 01: And just to be clear, the sentence you read that starts at 1055 to 1056 does seem to include topic as one of the kinds of identified latent information. [00:32:37] Speaker 01: Did you ever call out to the board [00:32:40] Speaker 01: either in the reply or at the oral argument before the board, this notion, which is now kind of the centerpiece? [00:32:54] Speaker 03: Well, I think just to turn back a few pages in the petition, Your Honor, I think at page 1047, you specifically said that a detected mask falls within the category of latent information that we're identifying. [00:33:08] Speaker 03: And then in those subsequent pages, [00:33:10] Speaker 03: I don't think there's any doubt that we're referring to that as a topic or header, and that is part of the required information. [00:33:18] Speaker 01: Well, I would be hard pressed to say that there's no doubt about it. [00:33:21] Speaker 03: Maybe you're right, but... Well, apologies for the overstatement. [00:33:26] Speaker 03: I don't take it as contested at least by the other side that if entirely you get a report, it must as a minimum have that piece of information in it, and therefore under any reasonable definition of the word required, [00:33:40] Speaker 03: just as when a judge requires a report on something, even if it's not specific information you know in advance, you know there is a certain type of information required. [00:33:50] Speaker 03: And under the plain meaning of required, we don't think anything more is required. [00:33:55] Speaker 03: So we ask the court to reverse the board's erroneous determination under any reasonable interpretation that Tyra does not teach this contested limitation and remand the board to address whether the prior art teaches the other limitations. [00:34:09] Speaker 03: If there are no further questions. [00:34:12] Speaker 00: Okay. [00:34:13] Speaker 00: We thank counsel for the arguments in this case and this case is now submitted. [00:34:18] Speaker 00: Thank you.