WEBVTT 1 "Johnny Hadlock" (1470405376) 00:00:05.819 --> 00:00:22.470 Friends of Nagara happy Thursday and welcome to today's Sponsor Solution series presentation Engineering intelligent records. AI enabled document capture for governance, compliance, and scale, presented by Nagara thought leader. 2 "Johnny Hadlock" (1470405376) 00:00:22.470 --> 00:00:37.800 Rico document scanners. I'm johnny Hadlock, the executive director of Nagara, and on behalf of our entire association, we welcome you to today's presentation. Now before we begin, there's just a couple of housekeeping items and announcements to go over and I will keep it brief. 3 "Johnny Hadlock" (1470405376) 00:00:37.800 --> 00:00:57.800 1st, today's presentation is being recorded just like all of Nagara's virtual presentations. And by tomorrow morning, all of you will have in your inbox an email letting you know when the recording along with our presenter's PowerPoint presentation is ready for review and sharing with your colleagues. So be on the watch for that email tomorrow morning. 4 "Johnny Hadlock" (1470405376) 00:00:57.800 --> 00:01:17.990 And second, where you welcome questions. So if you've got any of them, please look for the chat bubble at the bottom right hand side of your Webex screen. Feel free to enter your questions there and we'll get to them at the end in a dedicated question and answer session. No holds barred. Let us know what your questions are. We will be happy to answer them. 5 "Johnny Hadlock" (1470405376) 00:01:17.990 --> 00:01:38.180 I'd also like to begin by thanking Nagara's 2026 corporate sponsors whose company logos are shown here on the screen. Nagara's got a very ambitious mission. We've got a growing organization, and without the support of these corporate sponsors, it would be just far more difficult for Nagara to achieve all that we do. So in advance. 6 "Johnny Hadlock" (1470405376) 00:01:38.180 --> 00:01:54.030 I think our sponsors including Rico for their great support of Nagara and encourage all of you when you have a few minutes of time, go to the visit, go visit the Nagara website and check out their company profile pages there to learn more about the products and services that they offer you. 7 "Johnny Hadlock" (1470405376) 00:01:54.030 --> 00:02:14.030 We also have some upcoming events to highlight briefly. Next month, in the month of August, we've got two webinars in the Nagara webinar series. The 1st on 6 August will be leading AI readiness, essential skills for records and information professionals, and then the following week on Tuesday, 11 August, one. 8 "Johnny Hadlock" (1470405376) 00:02:14.030 --> 00:02:30.150 Called digital evidence management and ensuring the timeliness for prosecution purposes. We hope you'll take advantage of all the great professional development offerings Nagara has on the schedule. You can see what those are by visiting nagara.org slack events. 9 "Johnny Hadlock" (1470405376) 00:02:30.150 --> 00:02:50.150 And speaking of events, in less than 48 h time, nagara's staff will arrive on the ground in Philadelphia to prepare for next week's largest signature event, the 2026 Nagara annual conference. It is a sold out event this year. We are delighted for what's in store. Everything is going according to plan. 10 "Johnny Hadlock" (1470405376) 00:02:50.150 --> 00:03:10.150 Talk on Wood so far and we are delighted to see many of you next week in Philadelphia, including Rico, who will also be presenting and is another sponsor of that great event as well. For those who can't attend, we understand not everybody can. Next year's event will be in Denver Colorado in July, so save the date and plan to attend then. 11 "Johnny Hadlock" (1470405376) 00:03:10.150 --> 00:03:30.380 And that's all I have by way of announcements. I now want to introduce both our sponsor and our speaker for you today. As I mentioned, today's presentation offered by Nagara thought leader, Rico document scanners. Rico is a company that delivers fatchy compliant imaging solutions for government. 12 "Johnny Hadlock" (1470405376) 00:03:30.380 --> 00:03:49.950 In archives and records management. Their advanced technology ensures accurate, high quality digitization that meets federal standards for preservation and accessibility. Rico empowers agencies to safeguard historical documents while streamlining workflows for secure efficient records management. 13 "Johnny Hadlock" (1470405376) 00:03:49.950 --> 00:04:09.950 And presenting for Rico today, I welcome my friend Vaughn Minger. Vaughn has more than 27 years of experience in the document imaging and records management industry. He is he served as a past aim chapter president, bringing deep practical knowledge of document scanning and captured technologies. Vaughn, it's a delight to have you. 14 "Johnny Hadlock" (1470405376) 00:04:09.950 --> 00:04:18.165 Our presenter today, the rest of the hour is yours, my friend, you may now take the screen share away and have a good time. 15 "Vaughn Minger" (1577446400) 00:04:18.165 --> 00:04:37.237 Thank you sir. I always love your introductions, Appreciate everyone's time. We're gonna get started here in a second as soon as Webex lets me do my thing, you should see my screen in about 2 s. 16 "Johnny Hadlock" (1470405376) 00:04:37.237 --> 00:04:39.357 Looks good to me. 17 "Vaughn Minger" (1577446400) 00:04:39.357 --> 00:04:54.659 Firm, thanks sir. Yep. Again, for those who've been around as as long as I have, we are part of the old Fujitsu document scanner team, now branded as Rico and have joined. 18 "Vaughn Minger" (1577446400) 00:04:54.659 --> 00:05:09.809 The red Rico flag across the country. So happy to be here. Look forward to seeing everybody next week. You're going to hear a lot about AI. You're going to hear a lot about intelligent document processing. 19 "Vaughn Minger" (1577446400) 00:05:09.809 --> 00:05:25.139 You might hear it called intelligent data processing. The records that you are all in charge of, are the foundation of every AI engine out there without that accurate data. 20 "Vaughn Minger" (1577446400) 00:05:25.139 --> 00:05:45.139 That AI engine is producing poor results, incorrect answers leading people astray sometimes, so what we have said with our internal AI and as we're part of the AI process scanning those documents, trust would verify. 21 "Vaughn Minger" (1577446400) 00:05:45.139 --> 00:06:03.929 The integrity of that data, you know, is such an important piece of what we've been doing since since microfilm, not to age myself out, but I'm sure John Burgiss is laughing at me right now. So AI output suffers without high quality input. 22 "Vaughn Minger" (1577446400) 00:06:03.929 --> 00:06:21.089 You know, AI is, for lack of a better word, a database of information. If if your records are converted to digital, uploaded to that database, you know, if some of the eyes look like ones and the Rs look like S's and threes. 23 "Vaughn Minger" (1577446400) 00:06:21.089 --> 00:06:36.299 The intelligence of the AAI drops off and your accuracy rates start to fail rather quickly without that accurate conversion of your paper records. So. 24 "Vaughn Minger" (1577446400) 00:06:36.299 --> 00:06:51.389 And it's not because the models are weak, but it's really is about the data that we're dumping into, the AI engines. You know, unreliable capture, if, if you start with bad information or a crack foundation. 25 "Vaughn Minger" (1577446400) 00:06:51.389 --> 00:07:08.099 You know, the building doesn't stand up too long. So as we're all getting into the AI bucket, our scanners, processes, softwares, the ability to capture with accurate. 26 "Vaughn Minger" (1577446400) 00:07:08.099 --> 00:07:25.079 Information the 1st time, has never been as important as it is currently. So we're gonna add structure to the unstructured and talk about a complete solution. You know, some of the challenges that. 27 "Vaughn Minger" (1577446400) 00:07:25.079 --> 00:07:42.389 Of everyone experiences, you know, whether it's a small SM, you know, small business to, to county courthouse, to, to federal agency, you know, we all see different document types, different document formats. 28 "Vaughn Minger" (1577446400) 00:07:42.389 --> 00:08:02.219 You know, marketing groups get a hold of documents and they create their own special size or their own special font or watermark security backgrounds etc. We need some of that backgrounding and colors and watermarks to go away so that we can just give you the data. 29 "Vaughn Minger" (1577446400) 00:08:02.219 --> 00:08:22.219 And with our solution we're able to do that. As you look around the market and see what different softwares are working, what different capture platforms are out there, you know, how easy is the setup? You know, is, are you looking at professional services by the, by the week? 30 "Vaughn Minger" (1577446400) 00:08:22.219 --> 00:08:40.469 Or is it, install the software, you know, maybe go through a five or six step setup wizard and start scanning with accuracy, great images, and data extraction that's, more accurate than ever before. 31 "Vaughn Minger" (1577446400) 00:08:40.469 --> 00:08:57.539 Some of the problem with, you know, starting to out push out AI is the fragment fragmented capture systems. By that I mean, you may have a, a current system, you may have yesterday's system, you may have, we took over. 32 "Vaughn Minger" (1577446400) 00:08:57.539 --> 00:09:17.539 You know, the building next door and they were using this system, you know, how to how to get those images, maybe the paper, maybe the digital digital images that are already there, we can talk about the ability to import those images that may have been created off of a copier, may have been created. 33 "Vaughn Minger" (1577446400) 00:09:17.539 --> 00:09:50.429 It off of an a competitive scanner, but if you've got a file folder or a hard drive full of images, we can show you how to import those and get those ready for IDP and AI processes as well. You'll hear some talk about the, you know, what's the cost of this? Are there hidden costs? What's the front of keeping your information internal, security wise, you know, it's not I'm Sorry, it did not forward it's dilemma. 34 "Vaughn Minger" (1577446400) 00:09:50.429 --> 00:10:07.769 You know, we're trying to get into AI, everyone is, we're trying to play our part to make sure that everything is accurate, but you know, AI teams are struggling with or or part of their issues are, let's add AI, but let's not compound issues. 35 "Vaughn Minger" (1577446400) 00:10:07.769 --> 00:10:26.039 Let's make sure it's a solution and let's not add risk. We don't want to add shadow copies, personal information leaked out. There's a lot of pressure on the market to do something with AI. It's real. I, I it's not gonna go away, I don't believe. 36 "Vaughn Minger" (1577446400) 00:10:26.039 --> 00:10:41.669 There's way too much investment from Google to copilot, Gemini, you know, the top ten cloud, all the above AI is not gonna go away, so. 37 "Vaughn Minger" (1577446400) 00:10:41.669 --> 00:10:56.939 Do do we need to re architect our capture workflow? Or do we just need to add some intelligence to it and get that data off the form, off the paper, and show its value into that AI system? 38 "Vaughn Minger" (1577446400) 00:10:56.939 --> 00:11:14.579 We can expand, you know, without security, the, the talking points get long, but as you're discussing, you know, maybe a new capture platform that has the IDP, that has the, what we used to call advanced capture. 39 "Vaughn Minger" (1577446400) 00:11:14.579 --> 00:11:31.229 You know, automatically recognize a form, pull out that invoice number, pull out that social security number, pull out that HR onboarding document with all the relative fields, without any interaction with the keyboard or mouse. 40 "Vaughn Minger" (1577446400) 00:11:31.229 --> 00:11:47.009 All those things are, are possible now. And you might hear a lot about one size fits all. That's probably not true with any capture or hardware device out there anymore, solutions. 41 "Vaughn Minger" (1577446400) 00:11:47.009 --> 00:12:05.459 They think they are best at everything, tend to have a weak spot and can compromise, data, which then adds cost what we were trying to avoid in the 1st place. From input to action, this is just kind of a. 42 "Vaughn Minger" (1577446400) 00:12:05.459 --> 00:12:22.709 A picture of what we're talking about, you know, paper, going through a document scanner, going through an MFP, image processing, you know, we're trying to clean up that image. We want to make sure it looks exactly like the original piece of paper did. 43 "Vaughn Minger" (1577446400) 00:12:22.709 --> 00:12:41.969 All the way down to the square edges, NO black borders, you know, automatically rotated so it's face up and readable. If you want to, we can delete blank pages to save some space. All that's done at the, at the point of scan, making sure that that data is captured accurately the 1st time. 44 "Vaughn Minger" (1577446400) 00:12:41.969 --> 00:12:59.729 So that we're not rescanning the paper, we're not going back to the box in two weeks to find page six of batch 20, because it got cropped or cut off or was too dark, smudged, whatever it might be. Classification and extraction. 45 "Vaughn Minger" (1577446400) 00:12:59.729 --> 00:13:15.659 You know, the ca the capture solutions that are available now, can look at those documents, do the OCR, maybe do ICR, do OMR, you know, image capture, image capture recognition. 46 "Vaughn Minger" (1577446400) 00:13:15.659 --> 00:13:35.489 Optical mark recognition if you're doing some bubble forms, all those things can be automated now at the touch of a button. If we do recognize document type number one, you know, which workflows that need to be sent through, what's it gonna kick off and start the process? Where does that image need to end up? 47 "Vaughn Minger" (1577446400) 00:13:35.489 --> 00:13:51.359 And what metadata fields are, are required, you know, back end system integration, you know, the capture piece needs to be flexible, needs to have a, the openness or the capabilities to route that image to. 48 "Vaughn Minger" (1577446400) 00:13:51.359 --> 00:14:11.219 Wherever it might be on premise, cloud, maybe you have a different capture, backend that you want these images to get to. All that needs to be ease of use should be considered. And then the actionable business intelligence, what are we actually gonna extract from those forms? 49 "Vaughn Minger" (1577446400) 00:14:11.219 --> 00:14:27.269 From your, your, your bankers boxes full of documents or your fire file cabinets, full of documents. We believe the AI should be inside of Capture before the data even gets passed off. 50 "Vaughn Minger" (1577446400) 00:14:27.269 --> 00:14:46.859 To its permanent destination. So documents, you know, we're gonna intelligently clean up the image so that it's usable. We can extract the data and then pass that along with all the metadata analytics and automation possible. 51 "Vaughn Minger" (1577446400) 00:14:46.859 --> 00:15:04.979 Based on your rules, and your use case. Something that we're able to talk about and do, AI achieved superior results across. 52 "Vaughn Minger" (1577446400) 00:15:04.979 --> 00:15:23.549 All the IDP classifications, OCR, document classification, processing, et cetera, of, we're gonna talk about a hybrid version where you can scan documents that maybe aren't a permanent record, that maybe are just. 53 "Vaughn Minger" (1577446400) 00:15:23.549 --> 00:15:43.549 A simple, anything from a letter to, you know, something outside the normal, you can scan those without running those through the AI engine with our product. Not everything has to have that higher cost per image because we're not gonna do the data extraction and all the IDP processes. So as you do your dock prep. 54 "Vaughn Minger" (1577446400) 00:15:43.549 --> 00:16:02.369 You are in control of your total cost of operations. IT security, we're definitely on board, you know, working through certifications on all of our products to make sure that we can stand up, and meet all those. 55 "Vaughn Minger" (1577446400) 00:16:02.369 --> 00:16:19.319 Personal information require requirements that are out there, but we do encrypt, you know, our scanners do not hold on to or keep any image. We're not duplicating, and keeping copies anywhere, anything like that, et cetera. 56 "Vaughn Minger" (1577446400) 00:16:19.319 --> 00:16:40.909 The the safest place to apply AI is where capture already happens. So your documents are that AI data. We just need to scan them, extract them, and make that the value of that data come across. 57 "Vaughn Minger" (1577446400) 00:16:40.909 --> 00:16:58.679 As it improves your AI return your AI search engines and as they get smarter, based on that accurate data that was 1st created from the conversion, everyone's happier downstream because your accuracy rates are higher. 58 "Vaughn Minger" (1577446400) 00:16:58.679 --> 00:17:14.309 And you're ready to return, to help your business quicker, better, faster. So we are we're looking to. 59 "Vaughn Minger" (1577446400) 00:17:14.309 --> 00:17:29.399 And I would I was gonna use an old quote and I hesitated for a second and AI is basically garbage in, garbage out. So if you're aware of that. 60 "Vaughn Minger" (1577446400) 00:17:29.399 --> 00:17:46.679 You know, working with your IT folks, your organizational chart, your chief data officers, with the solutions that are on the street today, you can make sure that we're passing along accurate data to make them not have to worry about integrity. 61 "Vaughn Minger" (1577446400) 00:17:46.679 --> 00:18:06.679 Of that data that's being passed on, a little bit about how our version of AI works, you know, various input formats, you know, receipts, business cards, HR forms, invoices, HICCA forms, insurance. 62 "Vaughn Minger" (1577446400) 00:18:06.679 --> 00:18:23.219 Forms, you know, daily case files, whatever it might be. We're able to scan them. We're able to import an image that may have been created ten years ago on a different device, maybe a copier. 63 "Vaughn Minger" (1577446400) 00:18:23.219 --> 00:18:39.209 We can run those through capture, clean up that image, and give you the important data off that page going forward. AI value with reduced adoption risk. 64 "Vaughn Minger" (1577446400) 00:18:39.209 --> 00:18:55.289 What does that mean? Great question. We're not reinventing the wheel. We're just expecting an updating capture and solutions to work better with AI the 1st time. 65 "Vaughn Minger" (1577446400) 00:18:55.289 --> 00:19:10.319 We have a process that I'm gonna talk off, off slide here for a second, but after scan correction, if you stop by the table next week, we're happy to show it to you, but with our solution, you're only gonna scan the paper once. 66 "Vaughn Minger" (1577446400) 00:19:10.319 --> 00:19:30.319 Of if you're not happy with the image quality or you want it in black and white instead of color, we're able to do that without having to rescan the paper. So it's a great demo. I can show you how much time it's gonna save and it's less risks less risk to your documents too. One time through a device versus well that corner. 67 "Vaughn Minger" (1577446400) 00:19:30.319 --> 00:19:49.619 Got cut off or this one should have been color but not gray scale. It's a great demo, the ease of use and higher productivity rates are obvious once you see it of AI with governance build into your capture platform. 68 "Vaughn Minger" (1577446400) 00:19:49.619 --> 00:20:05.639 You know, we're not, IT likes our operational value. They like ease of use, installation, you know, scanner setup, software set up. We're not looking to add to your risk. 69 "Vaughn Minger" (1577446400) 00:20:05.639 --> 00:20:21.269 Create any ghost images, ghost copies, etc. Some of the use cases, mail room, you know, we're able to separate files every time we see an envelope. 70 "Vaughn Minger" (1577446400) 00:20:21.269 --> 00:20:37.259 We can set the software up so it knows that that is the beginning of a new page or a new file. Excuse me, a lot of vpos, service bureaus, you know, your back end records rooms, batch scanning, you know, all day long. 71 "Vaughn Minger" (1577446400) 00:20:37.259 --> 00:20:57.259 If you need to run through the AI engine, great, click that button. If you don't, go ahead and scan as a normal batch through the batch process. Banking and finance, and obviously government and public sector. We know where the paper is, we try to. 72 "Vaughn Minger" (1577446400) 00:20:57.259 --> 00:21:14.249 Make you as productive as possible with all of our solutions. A little more specific, you know, there a capture platform scanner connected via USB, to a host PC. 73 "Vaughn Minger" (1577446400) 00:21:14.249 --> 00:21:32.099 Of text recognition, forms identification, field extraction, out through pa out through Capture Pro premium, and there's a just a sampling of some of the connections and connectors that were already built in use in service. 74 "Vaughn Minger" (1577446400) 00:21:32.099 --> 00:21:48.269 We had the ability to get that data and the image to whatever destination you may need it to go. Enterprise class software, you know, secure. 75 "Vaughn Minger" (1577446400) 00:21:48.269 --> 00:22:03.929 Cloud native architecture or on prem, you know, we try to stay consistent across our solutions where the ease of use, set up, control. 76 "Vaughn Minger" (1577446400) 00:22:03.929 --> 00:22:19.469 Security settings. Those are all customizable. You're not locked out of, of any setting with our, with our solution. And keeping with that theme. 77 "Vaughn Minger" (1577446400) 00:22:19.469 --> 00:22:39.469 You know capture AI, we're trying to trying to follow the the feedback and you know what what the ease of use and what the end users are asking for, you know, keep the costs down simple, easy to adopt and scale out, you know, one scanner, three scanners, ten scanners. 78 "Vaughn Minger" (1577446400) 00:22:39.469 --> 00:23:12.049 Sometimes it might be, you know, desktop scanners, everyone gets their own and works day forward, otherwise, you know, it maybe three larger scanners in a records room, you know, scanning large batches all day long. And then enterprise ready, you know, rational services, set up, onboarding, all the support available, to help you get off the ground and get moving. Just marketing slide one on one, but keep. 79 "Vaughn Minger" (1577446400) 00:23:12.049 --> 00:23:33.419 Point, you know, we've, we win the awards. Our cameras have always created a a bright and crispy image. It's never been a question. Scanners are very reliable, not always the lowest price on the market, but definitely total cost of ownership, less rework, less rescan. 80 "Vaughn Minger" (1577446400) 00:23:33.419 --> 00:23:51.059 Less jamming, makes you more productive so that that upfront costs quickly disappears as you start scanning. Yeah, and where it fits, you know, scanning the document, maybe importing. 81 "Vaughn Minger" (1577446400) 00:23:51.059 --> 00:24:11.059 You know, you, you took over a different agency and you've got terabytes of data off of hard drives or a network folder that you now have access to. We can import those images automatically. We'll capture it the way you want it, cloud extraction classification or do it all on premise. 82 "Vaughn Minger" (1577446400) 00:24:11.059 --> 00:24:32.069 On prem if it's a security risk or security question. Traditional, you know, batch scanning, extract an OCR field read a barcode et cetera. All that to say it's a complete solution with great image quality that makes your AI engines more accurate. 83 "Vaughn Minger" (1577446400) 00:24:32.069 --> 00:24:48.119 Which should make everyone happier day forward with the AI solution. And with that, Johnny I can't see the chat, but if you had any questions for me. 84 "Johnny Hadlock" (1470405376) 00:24:48.119 --> 00:25:08.219 Hey Vonne, let us take a look in the chat. For those who maybe joined during the presentation, we welcome any questions you have about what Yvonne's been talking about. Enter that into the chat. That's the bottom right hand side of Webex. Look with a little bubble there. And I'll just take a look here. I'm not seeing any direct questions to me, which is good. Put them into the main chat. 85 "Johnny Hadlock" (1470405376) 00:25:08.219 --> 00:25:23.907 And we'll just give it a few minutes here to see what might come through. 86 "Vaughn Minger" (1577446400) 00:25:23.907 --> 00:25:56.397 And like Johnny said, we'll be there all week next week setting up on Tuesday and we'll have a table with a couple of scanners. We'll be doing some live demos, stop by with any questions as well. I'll have my sales engineer for anything really technical, so if you have a, a direct question or a, you know, a use case scenario that you have questions about, let us know. 87 "Johnny Hadlock" (1470405376) 00:25:56.397 --> 00:26:17.489 Alright, I see there's 77 of us here. No questions just yet, but let's let's wait another minute or two. Sometimes it takes a second for you to think about what to write, but we welcome all questions. 88 "Johnny Hadlock" (1470405376) 00:26:17.489 --> 00:26:37.489 Still NO questions, another 30 s. There's one gay I knew if I just waited long enough it would come in. Day. 89 "Johnny Hadlock" (1470405376) 00:26:37.489 --> 00:26:53.623 Man asks this one, how does the system learn which value corresponds to which fact in the schema? So you use the social security number as an example and how does the system know the digits and the value is the social security number specifically? 90 "Vaughn Minger" (1577446400) 00:26:53.623 --> 00:26:59.369 That's a great question and I can take you onto the hood, next week. 91 "Vaughn Minger" (1577446400) 00:26:59.369 --> 00:27:19.369 But the algorithms and the actual AI behind the scenes, social security numbers are, you know, 324 in format. Those rules are established and picked and if that's the, you're given the chance to scan it out. 92 "Vaughn Minger" (1577446400) 00:27:19.369 --> 00:27:35.839 Document, see what we pull, what the AI engine pulls from it, agree to keep it or not keep it. And once we tell it that rule, you can scan documents all day long, day forward, seven days a week, and it's gonna follow those rules for you. 93 "Johnny Hadlock" (1470405376) 00:27:35.839 --> 00:27:41.885 Fair enough, but do you know of any edge cases that you might be able to discuss a bit here? 94 "Vaughn Minger" (1577446400) 00:27:41.885 --> 00:27:44.279 Edge cases? 95 "Johnny Hadlock" (1470405376) 00:27:44.279 --> 00:27:53.875 Yeah, Dan writes. Let's go through an edge case, which I'm assuming maybe it's not a324 or something, not a rule for it yet, he says. 96 "Vaughn Minger" (1577446400) 00:27:53.875 --> 00:28:12.795 Gotcha. I don't have one, I don't have anything built to show you, but I can, I can easily show you, the rules and the behind the scenes that we have, next week if you're gonna be on site or or send me an email and I'd be happy to schedule a teams to show you how we do it. 97 "Johnny Hadlock" (1470405376) 00:28:12.795 --> 00:28:37.143 Yeah, and tomorrow and tomorrow morning when I send out the thank you email when the recording's available, I'll also include Von's email address. So if you have direct follow up questions, you can certainly have that information to reach out to him directly. Christine asks another question. Thanks again Dan. Vaughn, do you use a commercial AI agent such as copilot or is it proprietary? 98 "Vaughn Minger" (1577446400) 00:28:37.143 --> 00:28:54.137 The the capture engine does not use copilot. We do personally use Copilot internally, here at Rico. 99 "Johnny Hadlock" (1470405376) 00:28:54.137 --> 00:29:09.784 Thanks for that. And now Ian asks, is there a success rate percentage available for the optical character recognition function of scanning the document specifically around handwriting? He's representing a medical entity. 100 "Vaughn Minger" (1577446400) 00:29:09.784 --> 00:29:42.229 Great question standard text has reached the level of high 90 %, like 9890 9 % of ICR engines based on the dr.. I I can't give you a great number. You know, if it's if we get to 80 % of all handwriting I I think that'd be great. I mean, personally, I can't get it to read my my print, my check and scratch, but I would. 101 "Vaughn Minger" (1577446400) 00:29:42.229 --> 00:30:01.709 Tell you, overall, the numbers are getting better of every year because the engines are improving. But yeah, when the nurse writes off the edge of the chart or the dr. writes off the edge of the script, yeah, I I don't know if we're gonna get a hundred percent ICR rates. 102 "Vaughn Minger" (1577446400) 00:30:01.709 --> 00:30:18.599 Healthcare finance, there seems to be some different, different groups that like to write a lot by hand or right off the edge or, you know, they squeeze two lines of handwriting in the space of one line of. 103 "Vaughn Minger" (1577446400) 00:30:18.599 --> 00:30:33.779 Machine print text, so the engines, I won't say struggle like they used to, but I, I can't guarantee you, you know, 95 % accuracy on handwriting either. 104 "Johnny Hadlock" (1470405376) 00:30:33.779 --> 00:30:48.499 All right, great. Thank you. For that, let's see. He follows up, Ian follows up, NO worries. It makes sense. Is there an exception section that lets users see where a human review is needed, where paper stream isn't available to understand or learn the information? 105 "Vaughn Minger" (1577446400) 00:30:48.499 --> 00:31:20.359 Yes, a hundred percent. Yes, that is part of the the capture process. You'll get to see the image live. You get multiple index fields that you've agreed that are important that you want to keep. And then there's accuracy rates, you know, if it's green, it's, it's good to go. We believe we've read it correctly. If it's red, you know, maybe it's a digit short or there's a four that may have been a, you know, a G and at that point, the operator can interject and correct and move on to the next. 106 "Vaughn Minger" (1577446400) 00:31:20.359 --> 00:31:27.955 This field, but yeah, a hundred percent, you get to see what was captured and adjusted if necessary. 107 "Johnny Hadlock" (1470405376) 00:31:27.955 --> 00:31:42.455 Thanks. You you may have answered this, but Ivy asks, if you do identify something that's being read wrong, are you able to manually correct it? And then if so, does the system begin to learn that manual change and begin to correct it? 108 "Vaughn Minger" (1577446400) 00:31:42.455 --> 00:32:07.416 It, it learns off of every change that we make, number one, and yeah, a hundred percent you get to correct the field, you know, a comma versus a decimal, you know, an I versus a, a T, whatever it might be, but you get that interaction doing the during the QC and index process to fix anything that maybe incorrect. 109 "Johnny Hadlock" (1470405376) 00:32:07.416 --> 00:32:10.201 Right, thank you. All right on John. 110 "Vaughn Minger" (1577446400) 00:32:10.201 --> 00:32:44.078 Would oh sorry johnny. No, NO NO NO. You go on. I was just gonna add that we can do the same thing with the image itself. So if there's shadow, you know, the color, the gray scale black and white confusion, without your scan correction, with us it's point and click. You make adjustments to the image or the index field. So I just wanted to be clear that we can adjust the image on the fly as well as correct the indexing, the indexer metadata fields as well. 111 "Johnny Hadlock" (1470405376) 00:32:44.078 --> 00:32:56.281 Thank you. Now John asks if a federal agency wants to procure some of these Faji capable devices, who should they contact? Is that gonna be Rico directly or a resal reseller? 112 "Vaughn Minger" (1577446400) 00:32:56.281 --> 00:33:11.378 Of reseller. A hundred percent. We do not sell direct. I work for the manufacturer and I I work side by side hand in hand with our resellers. Okay. 113 "Johnny Hadlock" (1470405376) 00:33:11.378 --> 00:33:28.983 And Mylo asks, for after scan correction, if we've inherited a large collection of poor scans, what corrections can be created? Is it all garbage in garbage out or is there an ability to improve a poor scan? 114 "Vaughn Minger" (1577446400) 00:33:28.983 --> 00:33:50.459 That depends on the format and how it was captured the 1st time. If it's, you know, 200 DPI black and white, there's not as much as we can do with maybe a 300 DPI black and white or a 300 DPI color document of if it's already, you know, if it's 200 DPI. 115 "Vaughn Minger" (1577446400) 00:33:50.459 --> 00:34:10.459 Straight black and white and there's a lot of speckling and backgrounding. I cannot make that pristine. I can make it better but if the original is in gray scaler color, the the cleanup and intelligence piece can do a lot of work for you and make that image better and make that data. 116 "Vaughn Minger" (1577446400) 00:34:10.459 --> 00:34:23.524 Get more accurate. So it really depends on the the how the file was created originally to how much benefit you're gonna get from the cleanup engines. 117 "Johnny Hadlock" (1470405376) 00:34:23.524 --> 00:34:29.438 Great, thank you for that. And that was also Milo's assumption that he notes here. 118 "Vaughn Minger" (1577446400) 00:34:29.438 --> 00:34:30.878 Excellent. 119 "Johnny Hadlock" (1470405376) 00:34:30.878 --> 00:34:49.069 All right, I'm looking into the chat. I'm not seeing anything else but happy to stay on and wait for those questions to arrive if you've got them. Let's give it another 30 s or so to see if any new questions pop up, and as a reminder, as we're waiting for any last minute questions to arrive. 120 "Johnny Hadlock" (1470405376) 00:34:49.069 --> 00:35:09.809 Five. Again, this session was recorded. It will be available in the sponsor solutions series archive tomorrow morning as well as this PowerPoint presentation in a PDF form for your review. So if you missed any part of it or want to go back and review, you can and we'll also provide Vaughn's contact information directly if you've got questions about anything, you can reach out to him. 121 "Johnny Hadlock" (1470405376) 00:35:09.809 --> 00:35:31.729 Alright, that was the best philabuster I could do to try to feel time to give you more chances to ask questions and all I'm getting now is a thank you from Sheree. Hey is good to, good to see you here. With that, I think what we'll do is thank all of you for attending today. We'll thank Rico and Vaughn for the presentation and wish you all the very best and hope to see you. 122 "Johnny Hadlock" (1470405376) 00:35:31.729 --> 00:35:46.644 Next week and if we don't visit agar.org/events for our virtual events lineup and we'll see you right back here for another virtual presentation next time. Until then, everyone be well. Thank you. See you next time. Thanks Vaughnon. 123 "Vaughn Minger" (1577446400) 00:35:46.644 --> 00:35:48.869 Thank you. Save travels.