1 00:00:00,719 --> 00:00:04,000 Exciting things are happening at Becker's Healthcare. Stay 2 00:00:04,000 --> 00:00:06,240 ahead of industry trends with the new Becker's 3 00:00:06,240 --> 00:00:08,419 CFO plus Revenue Cycle podcast, 4 00:00:08,800 --> 00:00:11,119 your go to source for insights from top 5 00:00:11,119 --> 00:00:14,154 healthcare finance leaders. Tune in wherever you get 6 00:00:14,154 --> 00:00:15,054 your podcasts. 7 00:00:15,914 --> 00:00:18,475 And don't miss the tenth annual health IT 8 00:00:18,475 --> 00:00:22,154 plus digital health plus RCM conference, happening September 9 00:00:22,154 --> 00:00:24,474 30 to 10/03/2025 10 00:00:24,474 --> 00:00:25,214 in Chicago. 11 00:00:25,755 --> 00:00:26,879 Join thousands of 12 00:00:27,279 --> 00:00:30,079 executives, engage with industry leaders, and explore the 13 00:00:30,079 --> 00:00:31,619 future of health care innovation. 14 00:00:32,320 --> 00:00:36,719 Learn more about our upcoming events at beckershospitalreview.com. 15 00:00:36,719 --> 00:00:37,539 See you there. 16 00:00:38,159 --> 00:00:40,239 This is Laura Dierda with the Becker's Healthcare 17 00:00:40,239 --> 00:00:42,354 podcast. I'm thrilled today to be joined by 18 00:00:42,354 --> 00:00:44,914 doctor Clara Lynn, vice president of digital health 19 00:00:44,914 --> 00:00:47,234 and informatics as well as chief medical information 20 00:00:47,234 --> 00:00:49,394 officer at Seattle Children's. Doctor Lynn, it's a 21 00:00:49,394 --> 00:00:51,174 pleasure to have you on the podcast today. 22 00:00:51,634 --> 00:00:53,234 Hi, Laura. Thank you so much. Glad to 23 00:00:53,234 --> 00:00:53,975 be back. 24 00:00:54,510 --> 00:00:56,429 Absolutely. And, you know, it's always so fascinating 25 00:00:56,429 --> 00:00:58,670 to talk with you because Seattle Children's really 26 00:00:58,670 --> 00:01:01,149 is doing some innovative things and, is on 27 00:01:01,149 --> 00:01:02,590 the forefront of a lot of what's happening 28 00:01:02,590 --> 00:01:04,750 with digital technologies, AI, and health care and 29 00:01:04,750 --> 00:01:06,989 more. And I know you recently partnered with 30 00:01:06,989 --> 00:01:09,310 Google on some technology developments as well, so 31 00:01:09,310 --> 00:01:11,354 I'm excited to learn more about that. But 32 00:01:11,354 --> 00:01:13,534 before we dig in, for those who 33 00:01:13,994 --> 00:01:16,314 may be newer to the podcast, can you 34 00:01:16,314 --> 00:01:18,234 tell us a little bit more about yourself 35 00:01:18,234 --> 00:01:20,494 and Seattle Children's? What makes it unique? 36 00:01:21,275 --> 00:01:21,754 Yeah. 37 00:01:22,234 --> 00:01:24,590 Seattle Children's is a coronary 38 00:01:25,049 --> 00:01:27,069 referral hospital in the Pacific Northwest. 39 00:01:28,090 --> 00:01:29,530 One of the things that we're super proud 40 00:01:29,530 --> 00:01:31,609 of of among other things is that we 41 00:01:31,609 --> 00:01:34,090 are ranked as one of the top 10 42 00:01:34,090 --> 00:01:36,694 children's hospitals in the country by US News 43 00:01:36,694 --> 00:01:39,575 and World Report, and we are unique in 44 00:01:39,575 --> 00:01:41,494 that we serve a very large region. We 45 00:01:41,494 --> 00:01:45,174 cover the Washington State, Alaska, Montana, and Idaho 46 00:01:45,174 --> 00:01:45,674 region, 47 00:01:46,215 --> 00:01:48,774 as the referrals center for kids in this 48 00:01:48,774 --> 00:01:51,140 area, and we're also the only freestanding children's 49 00:01:51,140 --> 00:01:52,520 hospital in in the state. 50 00:01:53,540 --> 00:01:56,180 Well, that's fantastic. And, you know, could you, 51 00:01:56,740 --> 00:01:57,859 dive in and tell us a little bit 52 00:01:57,859 --> 00:01:59,620 about the accomplishment that you're most proud of 53 00:01:59,620 --> 00:02:00,760 from the last year? 54 00:02:01,219 --> 00:02:02,100 Yeah. I think, 55 00:02:03,034 --> 00:02:05,515 we've done so many cool things. We really 56 00:02:05,515 --> 00:02:07,775 try to keep Seattle Children's as a leader 57 00:02:08,074 --> 00:02:11,055 in innovation, particularly when it comes to pediatric 58 00:02:11,115 --> 00:02:13,275 health care and be an advocate for for 59 00:02:13,275 --> 00:02:14,735 kids when it comes to, 60 00:02:15,354 --> 00:02:18,155 the the national and and global trend of 61 00:02:18,155 --> 00:02:18,655 developing 62 00:02:19,810 --> 00:02:21,109 and innovative technology. 63 00:02:21,409 --> 00:02:22,930 But that being said, if we go back 64 00:02:22,930 --> 00:02:24,370 to sort of the bread and butter, the 65 00:02:24,370 --> 00:02:26,289 really basic stuff, some of the things that 66 00:02:26,289 --> 00:02:28,209 we're super proud of are, of course, our 67 00:02:28,209 --> 00:02:29,269 quality of care, 68 00:02:29,569 --> 00:02:31,155 the fact that we are we've been 69 00:02:31,634 --> 00:02:34,115 ranked and certified by US News World Report 70 00:02:34,115 --> 00:02:36,115 as one of the top 10 children's hospitals 71 00:02:36,115 --> 00:02:38,055 as I mentioned earlier. But, also, 72 00:02:38,675 --> 00:02:40,834 we did our class survey for the first 73 00:02:40,834 --> 00:02:42,775 time in three years in 2024. 74 00:02:43,155 --> 00:02:44,275 The last time that we did it was 75 00:02:44,275 --> 00:02:45,395 in 2021 76 00:02:45,395 --> 00:02:46,615 right after we implemented 77 00:02:47,000 --> 00:02:49,000 a new EHR in the middle of a 78 00:02:49,000 --> 00:02:49,500 pandemic. 79 00:02:49,879 --> 00:02:51,479 And then this is the first time since 80 00:02:51,479 --> 00:02:53,340 we've come out of that sort of stabilization 81 00:02:53,719 --> 00:02:55,739 phase and moving into more of an optimization 82 00:02:55,879 --> 00:02:56,939 and innovation phase. 83 00:02:57,400 --> 00:02:58,520 This is the first time we've done a 84 00:02:58,520 --> 00:03:00,914 class survey, and we're so proud that our 85 00:03:01,155 --> 00:03:03,254 both our physician and our nursing 86 00:03:03,634 --> 00:03:06,215 EHR satisfaction score went up significantly. 87 00:03:07,394 --> 00:03:09,555 My role as the chief medical information officer 88 00:03:09,555 --> 00:03:11,715 is of course with our providers and making 89 00:03:11,715 --> 00:03:12,354 sure that, 90 00:03:13,040 --> 00:03:15,920 that the experience with health IT in general 91 00:03:15,920 --> 00:03:18,719 is a good one. And the physician and 92 00:03:18,719 --> 00:03:19,219 provider 93 00:03:19,520 --> 00:03:20,020 class, 94 00:03:20,480 --> 00:03:23,120 EHR satisfaction score went up by 30 points 95 00:03:23,120 --> 00:03:24,560 in the three years, and that's just something 96 00:03:24,560 --> 00:03:26,175 that that we're so proud of. I mean, 97 00:03:26,175 --> 00:03:28,495 there is so much AI, so much cool 98 00:03:28,495 --> 00:03:30,254 stuff out there, but really the bread and 99 00:03:30,254 --> 00:03:32,194 butter things to make it 100 00:03:32,495 --> 00:03:34,814 a good environment for our clinicians to work 101 00:03:34,814 --> 00:03:36,895 in, to make it not a, 102 00:03:38,139 --> 00:03:40,699 cumbersome some experience for them to provide care 103 00:03:40,699 --> 00:03:42,780 to to the kids in the region. That's 104 00:03:42,780 --> 00:03:45,659 really what my job is, and, really excited 105 00:03:45,659 --> 00:03:46,960 to see that that outcome. 106 00:03:47,900 --> 00:03:49,180 That's great to hear. And, you know, when 107 00:03:49,180 --> 00:03:50,780 you look at those changes you've made and 108 00:03:50,780 --> 00:03:52,844 the transformation that brought you to this point, 109 00:03:53,564 --> 00:03:55,824 where you're seeing such an increase in clinician 110 00:03:55,965 --> 00:03:56,465 satisfaction, 111 00:03:57,004 --> 00:03:58,365 what what did you do? What can you 112 00:03:58,365 --> 00:04:00,044 attribute to some of those things? Is it 113 00:04:00,044 --> 00:04:02,205 just the advanced technology, or are there other 114 00:04:02,205 --> 00:04:04,604 ways that you have really been able to 115 00:04:04,604 --> 00:04:05,550 connect with the, 116 00:04:06,110 --> 00:04:08,590 clinical leaders in in team and, 117 00:04:09,310 --> 00:04:11,870 improve that satisfaction that that you have with 118 00:04:11,870 --> 00:04:12,530 the EHR? 119 00:04:13,310 --> 00:04:15,310 Yeah. I think it's it's a little bit 120 00:04:15,310 --> 00:04:15,870 of both. 121 00:04:16,670 --> 00:04:17,970 AI and innovation 122 00:04:18,750 --> 00:04:20,509 is definitely top of our mind. It's top 123 00:04:20,509 --> 00:04:23,444 of everyone's mind. But really the basic stuff 124 00:04:23,444 --> 00:04:25,125 the going back to the basics the bread 125 00:04:25,125 --> 00:04:28,185 and butter stuff is also really really important 126 00:04:28,245 --> 00:04:29,785 making sure that our EHR 127 00:04:30,485 --> 00:04:33,305 is optimized in from an informatics perspective, 128 00:04:33,990 --> 00:04:38,149 from a usability perspective, from just a supporting 129 00:04:38,149 --> 00:04:41,189 and designing workflows that make sense. I think 130 00:04:41,189 --> 00:04:43,910 all of those things helping helping every role 131 00:04:43,910 --> 00:04:45,829 practice at the top of their license, helping 132 00:04:45,829 --> 00:04:48,404 supporting our physicians and our providers to do 133 00:04:48,404 --> 00:04:50,245 what they're trained to do to help, 134 00:04:50,564 --> 00:04:52,724 provide the best quality of care. I think 135 00:04:52,724 --> 00:04:54,564 all going back to the basics is what 136 00:04:54,564 --> 00:04:57,224 we really focused the last three years, 137 00:04:57,845 --> 00:04:58,504 on doing, 138 00:04:58,805 --> 00:05:00,724 and, and it's really paying off in in 139 00:05:00,724 --> 00:05:02,104 that in that survey result. 140 00:05:03,149 --> 00:05:05,649 That's helpful to know. And and really, definitely 141 00:05:05,709 --> 00:05:07,870 centering to have an understanding of just, you 142 00:05:07,870 --> 00:05:09,970 know, getting back to those basics and understanding 143 00:05:10,029 --> 00:05:12,209 what is gonna be most helpful and beneficial 144 00:05:12,269 --> 00:05:14,805 for the clinicians and bringing that forward. I'm 145 00:05:14,805 --> 00:05:16,084 curious. Where do you see some of the 146 00:05:16,084 --> 00:05:17,925 growth opportunities for your team in the next 147 00:05:17,925 --> 00:05:19,144 twelve months or so? 148 00:05:20,004 --> 00:05:22,084 Yeah. I think the from a from a 149 00:05:22,084 --> 00:05:23,144 health IT perspective, 150 00:05:23,845 --> 00:05:24,584 like everybody 151 00:05:25,125 --> 00:05:27,685 else, the biggest growth opportunity is, of course, 152 00:05:27,685 --> 00:05:29,944 artificial intelligence and what it can offer, 153 00:05:30,750 --> 00:05:32,829 and not just the the the potential that 154 00:05:32,829 --> 00:05:34,509 it can bring to the care that we 155 00:05:34,509 --> 00:05:36,769 deliver and the experience of our our, 156 00:05:37,550 --> 00:05:39,329 workforce, but it's also 157 00:05:39,949 --> 00:05:41,569 in terms of how we can 158 00:05:42,605 --> 00:05:45,485 help our internal workforce skill set grow when 159 00:05:45,485 --> 00:05:47,644 it comes to artificial intelligence. And we're not 160 00:05:47,644 --> 00:05:49,964 just talking about the IT teams growing and 161 00:05:49,964 --> 00:05:52,444 not just talking about giving, you know, cool 162 00:05:52,444 --> 00:05:55,850 generative AI to our our every, you know, 163 00:05:55,990 --> 00:05:58,490 employee to to write emails or make PowerPoints. 164 00:05:58,550 --> 00:06:01,610 That that is great. But I think what 165 00:06:01,910 --> 00:06:04,649 another really big opportunity here is we can 166 00:06:04,709 --> 00:06:06,810 help our workforce become more comfortable 167 00:06:07,269 --> 00:06:09,750 in using artificial intelligence, and they can have 168 00:06:09,750 --> 00:06:12,045 the tool built and the framework that they 169 00:06:12,045 --> 00:06:14,125 need in order to, and kind of a 170 00:06:14,125 --> 00:06:16,125 mental model that they that we can provide 171 00:06:16,125 --> 00:06:17,105 them to evaluate 172 00:06:17,644 --> 00:06:20,045 AI technology when they're interacting with it on 173 00:06:20,045 --> 00:06:21,725 a day to day basis. So evaluating it 174 00:06:21,725 --> 00:06:23,884 critically. Am I using it right? Is this 175 00:06:23,884 --> 00:06:24,704 tool fishy? 176 00:06:25,009 --> 00:06:26,930 You know, do I feel like there's a 177 00:06:26,930 --> 00:06:29,009 concern here? How do I do how how 178 00:06:29,009 --> 00:06:31,009 do I use this particular tool on my 179 00:06:31,009 --> 00:06:33,410 desktop, you know, responsibly? I think those are 180 00:06:33,490 --> 00:06:35,729 that's that's where I think we need as 181 00:06:35,729 --> 00:06:37,810 an organization and as an industry need to 182 00:06:37,810 --> 00:06:40,365 grow in is to start to disseminate this 183 00:06:40,365 --> 00:06:41,024 type of, 184 00:06:41,404 --> 00:06:43,485 this mental model, this framework to help them 185 00:06:43,485 --> 00:06:45,185 critically evaluate these technologies, 186 00:06:45,884 --> 00:06:47,104 on a regular basis. 187 00:06:48,204 --> 00:06:49,884 Got it. And that and it's fascinating to 188 00:06:49,884 --> 00:06:51,120 hear it. And really cool, 189 00:06:52,079 --> 00:06:54,019 to see some of those the AI technology 190 00:06:54,160 --> 00:06:54,899 come into 191 00:06:55,279 --> 00:06:56,899 the day to day use spaces, 192 00:06:57,279 --> 00:07:00,000 and understanding what the workforce needs, supporting them 193 00:07:00,000 --> 00:07:01,439 the right way, but then also of scaling 194 00:07:01,439 --> 00:07:03,134 them into a a place where it, 195 00:07:03,855 --> 00:07:06,735 really becomes most valuable and useful. I'm wondering 196 00:07:06,735 --> 00:07:08,014 too, could you tell us a little bit 197 00:07:08,014 --> 00:07:09,134 about some of the things that you've been 198 00:07:09,134 --> 00:07:10,574 working on with Google and, 199 00:07:10,895 --> 00:07:13,375 just, you know, early what that experience has 200 00:07:13,375 --> 00:07:14,675 been like and and the outcomes? 201 00:07:15,134 --> 00:07:16,735 Yeah. So that was a that was a 202 00:07:16,735 --> 00:07:18,300 really cool experience. So, 203 00:07:18,779 --> 00:07:20,319 about a year ago when, 204 00:07:21,339 --> 00:07:24,159 artificial intelligence and particularly generative AI 205 00:07:24,459 --> 00:07:26,479 became sort of the talk of the town, 206 00:07:26,539 --> 00:07:28,699 Seattle Children's, like everyone else, started to think 207 00:07:28,699 --> 00:07:30,620 about sort of where our use cases are, 208 00:07:30,620 --> 00:07:32,625 where our needs are. So we're not we're 209 00:07:32,625 --> 00:07:34,384 not really going out there to look for 210 00:07:34,384 --> 00:07:36,305 technology and then trying to retrofit it into 211 00:07:36,305 --> 00:07:38,704 our environment. We're really starting with what the 212 00:07:38,704 --> 00:07:41,345 frontline users are asking for and what their 213 00:07:41,504 --> 00:07:43,264 where their needs are. And so, 214 00:07:43,824 --> 00:07:44,324 we 215 00:07:44,930 --> 00:07:46,610 we came up with a list of use 216 00:07:46,610 --> 00:07:48,470 cases that we thought would be, 217 00:07:49,250 --> 00:07:51,649 would be very beneficial to the organization from 218 00:07:51,649 --> 00:07:54,770 an operational efficiency standpoint, from an improving the 219 00:07:54,770 --> 00:07:56,709 quality of the care perspective. 220 00:07:57,295 --> 00:08:00,335 And this particular one that we developed with 221 00:08:00,495 --> 00:08:01,795 in partnership with Google, 222 00:08:02,574 --> 00:08:04,735 ended up being sort of a proof of 223 00:08:04,735 --> 00:08:07,855 concept for us, an early use case that 224 00:08:07,855 --> 00:08:10,560 we wanted to see if it's possible in 225 00:08:10,560 --> 00:08:12,480 a in a sort of fail fast model. 226 00:08:12,480 --> 00:08:14,100 Can we get this technology 227 00:08:14,560 --> 00:08:15,779 using Google Gemini, 228 00:08:17,040 --> 00:08:19,040 to to a state where we think it's 229 00:08:19,040 --> 00:08:21,939 clinically acceptable and from a workflow's perspective, 230 00:08:22,644 --> 00:08:24,884 that is usable and from a technology standpoint 231 00:08:24,884 --> 00:08:25,704 that is scalable. 232 00:08:26,165 --> 00:08:28,084 And so we can of the four or 233 00:08:28,084 --> 00:08:30,644 five, use cases that we ended up coming 234 00:08:30,644 --> 00:08:32,164 up with, we you know, for each of 235 00:08:32,164 --> 00:08:34,024 the use case, we partner with a different 236 00:08:34,164 --> 00:08:34,664 different 237 00:08:35,009 --> 00:08:37,730 team outside of Seattle Children's, and this particular 238 00:08:37,730 --> 00:08:38,790 one was with Google. 239 00:08:39,170 --> 00:08:41,570 And, what we've done is I I need 240 00:08:41,570 --> 00:08:43,170 to take us back a little bit further. 241 00:08:43,170 --> 00:08:44,290 So in 2010, 242 00:08:44,290 --> 00:08:44,790 our, 243 00:08:45,330 --> 00:08:48,334 Seattle Children's created this group called the clinical 244 00:08:48,475 --> 00:08:51,195 effectiveness program, which is under our center of 245 00:08:51,195 --> 00:08:51,695 quality. 246 00:08:52,074 --> 00:08:54,954 And what the clinical effectiveness program does is 247 00:08:54,954 --> 00:08:56,634 that it looks at all of the medical 248 00:08:56,634 --> 00:08:59,195 literature out there. There's millions upon millions. Right? 249 00:08:59,195 --> 00:09:01,914 The medical literature doubles every 70 something days 250 00:09:01,914 --> 00:09:04,889 these days. And the clinical effectiveness team sort 251 00:09:04,889 --> 00:09:07,230 of gathers the most up to date information 252 00:09:07,370 --> 00:09:09,470 and the guidelines that are out there for 253 00:09:10,009 --> 00:09:12,329 each and every disease across the 70 diseases 254 00:09:12,329 --> 00:09:14,169 or conditions that that we have on our 255 00:09:14,169 --> 00:09:16,409 catalog. And for each one of these condition, 256 00:09:16,409 --> 00:09:17,070 they develop 257 00:09:17,625 --> 00:09:19,085 a guideline, a pathway 258 00:09:19,465 --> 00:09:21,625 that we think is the standard provides the 259 00:09:21,625 --> 00:09:24,024 best of care in a standardized fashion. And 260 00:09:24,024 --> 00:09:26,044 so now that we have these 70 something 261 00:09:26,105 --> 00:09:26,605 conditions 262 00:09:26,985 --> 00:09:28,924 sort of created and 263 00:09:29,304 --> 00:09:32,079 provided a guideline and a pathway to to 264 00:09:32,079 --> 00:09:33,919 our clinicians, we can standardize the best of 265 00:09:33,919 --> 00:09:36,559 care for every kid regardless of which provider 266 00:09:36,559 --> 00:09:38,480 you see at Seattle Children's. And so that's 267 00:09:38,480 --> 00:09:40,320 kind of the basis that the shoulders of 268 00:09:40,320 --> 00:09:42,559 the giants that we're standing on. So this 269 00:09:42,559 --> 00:09:45,215 particular tool that we created with Google called 270 00:09:45,215 --> 00:09:46,274 Pathway Assistant 271 00:09:46,654 --> 00:09:48,815 sort of is an evolution to that. So 272 00:09:48,815 --> 00:09:51,855 we took these thousands of pages of really 273 00:09:51,855 --> 00:09:52,674 great content 274 00:09:53,134 --> 00:09:56,014 developed by the clinical effectiveness team under our 275 00:09:56,014 --> 00:09:56,995 center of quality, 276 00:09:57,339 --> 00:10:00,139 and we layered basically Google Gemini on top 277 00:10:00,139 --> 00:10:02,059 of it. And it's not also just the 278 00:10:02,059 --> 00:10:04,539 straight of Gemini that we can see on 279 00:10:04,539 --> 00:10:07,259 online, but it's really a Gemini that's been 280 00:10:07,259 --> 00:10:09,519 fine tuned specifically to these, 281 00:10:10,695 --> 00:10:12,455 these pathways that we see our children's has 282 00:10:12,455 --> 00:10:14,795 developed. So it's only referring to these pathways. 283 00:10:15,335 --> 00:10:17,894 We did it purely through prompt and fine 284 00:10:17,894 --> 00:10:18,955 tuning of Gemini. 285 00:10:19,654 --> 00:10:21,975 And, right out the box, it had a 286 00:10:21,975 --> 00:10:22,475 decent 287 00:10:22,934 --> 00:10:25,049 accuracy, but not good enough for clinical use. 288 00:10:25,049 --> 00:10:27,450 And we kept tuning fine tuning and prompting 289 00:10:27,450 --> 00:10:29,610 and changing, you know, the the prompts and 290 00:10:29,610 --> 00:10:32,409 and making sure that that Gemini understands what 291 00:10:32,409 --> 00:10:33,850 we were trying to get out of it. 292 00:10:33,850 --> 00:10:36,429 And the end product is a conversational 293 00:10:36,970 --> 00:10:38,970 chatbot that can ask you questions if your 294 00:10:38,970 --> 00:10:40,945 prompt is not specific enough, and then it 295 00:10:40,945 --> 00:10:42,784 gives you an answer back. So I can 296 00:10:42,784 --> 00:10:44,625 ask prompt the the chatbot, for example, the 297 00:10:44,625 --> 00:10:46,865 pathway assistant, for example. Hey. I have a 298 00:10:46,865 --> 00:10:48,384 four year old in front of me in 299 00:10:48,384 --> 00:10:50,804 the emergency room with an asthma exacerbation. 300 00:10:51,184 --> 00:10:53,664 What should I do next? The chatbot will 301 00:10:53,664 --> 00:10:56,879 basically ask you questions about the patient. And 302 00:10:56,879 --> 00:10:58,240 then in the end, it doesn't tell you 303 00:10:58,240 --> 00:10:59,600 what you should do next even though that 304 00:10:59,600 --> 00:11:01,459 was the question I asked as a physician. 305 00:11:01,600 --> 00:11:02,959 But what it will do is it will 306 00:11:02,959 --> 00:11:05,299 collate all of that evidence based information 307 00:11:05,759 --> 00:11:07,875 from the pathways that we've created 308 00:11:08,254 --> 00:11:09,795 over the last fifteen years, 309 00:11:10,175 --> 00:11:12,495 by these, you know, dozens of medical experts 310 00:11:12,495 --> 00:11:14,975 here at Seattle Children's. And then they present 311 00:11:14,975 --> 00:11:17,695 that information to the request or to the 312 00:11:17,695 --> 00:11:19,529 to the person that's chatting with and says, 313 00:11:19,769 --> 00:11:21,470 this is here all that information. 314 00:11:21,850 --> 00:11:24,009 Here is the page of guideline that where 315 00:11:24,009 --> 00:11:26,730 I'm referencing from, and here's what we the 316 00:11:26,730 --> 00:11:29,290 guideline suggests your next step of evaluation should 317 00:11:29,290 --> 00:11:31,690 be. And so the human ultimately is still 318 00:11:31,690 --> 00:11:33,230 the one that's making the decision, 319 00:11:33,705 --> 00:11:36,184 and ultimately deciding if this is the information 320 00:11:36,184 --> 00:11:38,024 that is most appropriate for the patient in 321 00:11:38,024 --> 00:11:40,904 front of them. But we using pathway assistant 322 00:11:40,904 --> 00:11:44,345 effectively shrunk the information searching time to just 323 00:11:44,345 --> 00:11:45,965 seconds for for the clinician. 324 00:11:47,399 --> 00:11:50,039 Wow. That's fascinating. You know? And really amazing 325 00:11:50,039 --> 00:11:52,059 what you're able to do bringing all that 326 00:11:52,279 --> 00:11:52,779 information, 327 00:11:53,639 --> 00:11:55,179 in in data and, 328 00:11:55,879 --> 00:11:58,519 pathways together to then, you know, in the 329 00:11:58,519 --> 00:12:00,539 moment, find what will be a recommendation, 330 00:12:01,654 --> 00:12:03,975 that is personalized to the patient. That's really 331 00:12:03,975 --> 00:12:06,134 cool. And and seems like it has a 332 00:12:06,134 --> 00:12:08,454 lot of benefits for clinicians as well as 333 00:12:08,454 --> 00:12:09,914 patients in the future. 334 00:12:10,534 --> 00:12:12,934 Oh, yeah. We're we're really excited. And and 335 00:12:12,934 --> 00:12:15,014 one of the things that that we use 336 00:12:15,014 --> 00:12:16,774 this also as a proof of concept to 337 00:12:16,774 --> 00:12:19,490 see, can generative AI and large language models 338 00:12:19,870 --> 00:12:23,470 cat catalog internal documents like this in that 339 00:12:23,470 --> 00:12:25,950 sort of a breadth. And also, it wasn't 340 00:12:25,950 --> 00:12:27,950 just text because we also fit part of 341 00:12:27,950 --> 00:12:30,375 the a lot of the guidelines are flowcharts 342 00:12:30,514 --> 00:12:34,195 and and spreadsheets and images and, you know, 343 00:12:34,195 --> 00:12:36,115 references and and things like that. And so 344 00:12:36,115 --> 00:12:36,915 it was really, 345 00:12:37,315 --> 00:12:40,514 a multimodal ingestion into the large language model. 346 00:12:40,514 --> 00:12:42,434 And so this was the proof of concept 347 00:12:42,434 --> 00:12:44,195 that we needed to show that, you know, 348 00:12:44,195 --> 00:12:44,675 it can 349 00:12:45,360 --> 00:12:46,980 technologies like this can accommodate 350 00:12:47,360 --> 00:12:48,100 our needs, 351 00:12:48,720 --> 00:12:51,120 beyond sort of the the the the pathway 352 00:12:51,200 --> 00:12:53,620 what what the pathway assistant itself can do. 353 00:12:54,320 --> 00:12:56,159 It's fascinating to hear. Thank you so much 354 00:12:56,159 --> 00:12:57,764 for sharing with us that, 355 00:12:58,165 --> 00:13:00,404 the technology and the innovation journey you've been 356 00:13:00,404 --> 00:13:02,804 on. I I think it meet makes for 357 00:13:02,804 --> 00:13:04,644 a lot of huge opportunities in the future. 358 00:13:04,644 --> 00:13:06,485 We'll just keep an eye on how things 359 00:13:06,485 --> 00:13:08,105 continue to evolve and grow. 360 00:13:08,565 --> 00:13:10,485 I'm curious. You know, we talked a lot 361 00:13:10,485 --> 00:13:11,764 about some of the cool things that are 362 00:13:11,764 --> 00:13:13,740 happening, but what are the challenges that you're 363 00:13:13,980 --> 00:13:16,720 anticipating, especially with technology and AI as well? 364 00:13:17,500 --> 00:13:19,899 Yeah. So I've been thinking about that a 365 00:13:19,899 --> 00:13:20,720 lot lately. 366 00:13:21,740 --> 00:13:24,399 I mentioned earlier that we selected 367 00:13:24,700 --> 00:13:26,480 a small handful, three or four, 368 00:13:27,144 --> 00:13:29,884 AI solutions that we developed ourselves, 369 00:13:30,745 --> 00:13:33,404 starting over the last year with with different, 370 00:13:34,345 --> 00:13:36,764 different teams and different people that we collaborated 371 00:13:36,904 --> 00:13:37,404 with. 372 00:13:37,865 --> 00:13:39,945 This was just one of one of the 373 00:13:39,945 --> 00:13:41,820 three or four that we developed. But then 374 00:13:41,820 --> 00:13:44,000 at the same time, the organization 375 00:13:44,300 --> 00:13:46,139 hasn't slowed down. Right? Like and and the 376 00:13:46,139 --> 00:13:48,220 industry didn't slow down. And so so the 377 00:13:48,220 --> 00:13:51,100 industry is booming, including our own EHR our 378 00:13:51,100 --> 00:13:53,740 EHR epic. You know, they're developing a ton 379 00:13:53,740 --> 00:13:56,174 of AI AI solutions. And then out there, 380 00:13:56,174 --> 00:13:58,254 there's also a lot more off the shelf 381 00:13:58,254 --> 00:14:00,674 products that a lot of our teams are 382 00:14:00,815 --> 00:14:03,455 evaluating and trying to incorporate into their workflows 383 00:14:03,455 --> 00:14:05,154 and into into the organization. 384 00:14:05,535 --> 00:14:07,774 So this is only a growing technology, and 385 00:14:07,774 --> 00:14:10,269 it's growing quite rapidly. So the biggest challenge 386 00:14:10,269 --> 00:14:12,049 is how do we support the technology 387 00:14:12,590 --> 00:14:14,529 from an infrastructure standpoint 388 00:14:14,910 --> 00:14:16,290 all the way up to governance. 389 00:14:17,230 --> 00:14:19,330 How do we make sure that we are 390 00:14:19,470 --> 00:14:19,970 evaluating, 391 00:14:20,750 --> 00:14:21,809 you know, implementing, 392 00:14:22,350 --> 00:14:22,850 scaling, 393 00:14:23,485 --> 00:14:24,304 and then maintaining 394 00:14:24,605 --> 00:14:26,065 that growing portfolio, 395 00:14:26,524 --> 00:14:29,004 whether it's something that's purchased or licensed off 396 00:14:29,004 --> 00:14:30,945 the shelf, or something that we've developed, 397 00:14:31,404 --> 00:14:34,684 upfront. And so so that is what I've 398 00:14:34,684 --> 00:14:36,845 been thinking about a lot and really trying 399 00:14:36,845 --> 00:14:39,000 to figure out because I think that ultimately 400 00:14:39,000 --> 00:14:40,679 will be the challenge, not just for Seattle 401 00:14:40,679 --> 00:14:43,320 Children's, but everybody out there in health care 402 00:14:43,320 --> 00:14:43,820 IT, 403 00:14:44,519 --> 00:14:46,919 that is seeing sort of a rapid expansion 404 00:14:46,919 --> 00:14:49,335 of their their AI portfolio right now. 405 00:14:50,215 --> 00:14:51,335 That's such a great point. And I I 406 00:14:51,335 --> 00:14:53,175 think that is a lot of organizations that 407 00:14:53,175 --> 00:14:54,875 are are really truly seeing the 408 00:14:55,254 --> 00:14:57,995 broad leaps in their AI technologies and portfolios 409 00:14:58,054 --> 00:14:59,894 or or pilots that they're working on. And 410 00:14:59,894 --> 00:15:01,735 then as you mentioned, just trying to figure 411 00:15:01,735 --> 00:15:03,254 out, you know, what's gonna make the most 412 00:15:03,254 --> 00:15:06,000 sense and, and keep that a growing portfolio 413 00:15:06,220 --> 00:15:07,419 under control is, 414 00:15:07,899 --> 00:15:10,299 easier said than done, I think. Yeah. For 415 00:15:10,299 --> 00:15:13,200 sure. Fantastic. Well, before we wrap up here, 416 00:15:13,259 --> 00:15:14,860 could you talk a little bit about the 417 00:15:14,860 --> 00:15:16,700 future? What is the number one thing that 418 00:15:16,700 --> 00:15:19,019 you're doing right now to set Seattle Children's 419 00:15:19,019 --> 00:15:20,320 up for long term success? 420 00:15:21,304 --> 00:15:23,304 I think that's exactly it. Is is how 421 00:15:23,304 --> 00:15:25,725 do I how do we, at Seattle Children's, 422 00:15:26,504 --> 00:15:27,644 set up an ecosystem 423 00:15:28,105 --> 00:15:31,325 so that we can accommodate this rapidly 424 00:15:32,184 --> 00:15:34,024 evolving field? Because we don't one is that 425 00:15:34,024 --> 00:15:34,764 we don't know 426 00:15:35,210 --> 00:15:36,570 we don't know where this is all going 427 00:15:36,570 --> 00:15:38,090 right now. It's such a fast moving train. 428 00:15:38,090 --> 00:15:39,309 It's super exciting. 429 00:15:39,769 --> 00:15:42,509 Everybody's on board. Most people are on board. 430 00:15:43,289 --> 00:15:45,870 But how do we how do we support 431 00:15:45,929 --> 00:15:48,250 that? Right? And so what I'm hoping to 432 00:15:48,250 --> 00:15:50,154 do now is to sort of set the 433 00:15:50,154 --> 00:15:53,595 organization up, not just from a people and 434 00:15:53,595 --> 00:15:54,894 skill set perspective, 435 00:15:55,355 --> 00:15:56,574 but also getting, 436 00:15:57,274 --> 00:15:59,534 like I said mentioned earlier, getting the workforce 437 00:15:59,914 --> 00:16:02,174 to a point where they can evaluate 438 00:16:02,759 --> 00:16:05,580 critically their experience with artificial intelligence 439 00:16:06,120 --> 00:16:08,120 on a day to day click by click 440 00:16:08,120 --> 00:16:08,620 basis 441 00:16:09,000 --> 00:16:10,860 so that, so that they know, 442 00:16:11,480 --> 00:16:13,879 hey, that just seems fishy or, hey, that's 443 00:16:13,879 --> 00:16:16,040 a great opportunity for me. So, you know, 444 00:16:16,040 --> 00:16:18,105 that that to me is is a really 445 00:16:18,105 --> 00:16:18,605 critical 446 00:16:18,985 --> 00:16:21,164 step in in democratizing, 447 00:16:21,544 --> 00:16:24,264 I guess, this technology to to every every 448 00:16:24,264 --> 00:16:25,945 area of the workforce. Because I I think 449 00:16:25,945 --> 00:16:27,464 it's only a matter of time that we're 450 00:16:27,464 --> 00:16:29,625 gonna start to see AI incorporated in everything 451 00:16:29,625 --> 00:16:30,524 that we use. 452 00:16:30,839 --> 00:16:32,199 So I think that that is in and 453 00:16:32,199 --> 00:16:34,199 that is, sort of what we're what we're 454 00:16:34,199 --> 00:16:35,639 working on next is how do we set 455 00:16:35,639 --> 00:16:38,519 up our workforce, both structured teams and also 456 00:16:38,519 --> 00:16:39,579 this AI literacy 457 00:16:40,039 --> 00:16:41,980 to to prepare us for the future. 458 00:16:43,184 --> 00:16:46,225 Absolutely. That it is amazing to hear. Doctor 459 00:16:46,225 --> 00:16:47,504 Lin, thank you so much for joining us 460 00:16:47,504 --> 00:16:49,105 on the podcast today. This has been a 461 00:16:49,105 --> 00:16:51,264 really fantastic conversation, and I look forward to 462 00:16:51,264 --> 00:16:52,804 connecting with you again soon. 463 00:16:53,345 --> 00:16:54,725 Thank you so much, Laura.