1
00:00:07,910 --> 00:00:11,170
Welcome  to Hard Problems, Smart
Solutions - the Newfire Podcast, where

2
00:00:11,170 --> 00:00:14,750
we dive into complex challenges and ways
to solve them with top industry leaders.

3
00:00:15,170 --> 00:00:18,599
I'm Gordon Wong, VP of Data and
AI at Newfire Global Partners,

4
00:00:18,599 --> 00:00:19,659
and your  host for this episode.

5
00:00:20,300 --> 00:00:23,190
In each episode, we bring you
conversations with top innovators

6
00:00:23,380 --> 00:00:26,359
and decision-makers, tackling the
biggest issues across industries.

7
00:00:27,000 --> 00:00:30,080
Whether you're looking for insights to
drive your own strategies or to learn

8
00:00:30,080 --> 00:00:31,650
from the best, you're in the right place.

9
00:00:32,130 --> 00:00:32,880
Let's get started.

10
00:00:35,560 --> 00:00:36,230
Hi, everybody.

11
00:00:36,615 --> 00:00:38,935
Welcome to the first episode
of Hard Problems, Smart

12
00:00:38,935 --> 00:00:40,525
Solutions, the Newfire podcast.

13
00:00:41,005 --> 00:00:44,205
Today, I'm thrilled to introduce
our guest, Emily Lindemir, Executive

14
00:00:44,205 --> 00:00:47,265
Director of Data and Healthcare
Innovation at JPMorgan Chase Co.

15
00:00:47,884 --> 00:00:50,754
Emily has led data science
initiatives at Cityblock Health and

16
00:00:50,754 --> 00:00:53,914
Wellframe, focusing on innovative
AI solutions to improve healthcare.

17
00:00:54,275 --> 00:00:57,005
She holds a PhD in medical
engineering and medical physics

18
00:00:57,155 --> 00:00:59,825
from the Harvard MIT Division of
Health Sciences and Technology.

19
00:01:00,365 --> 00:01:02,575
Emily has been instrumental in
advancing how we use data to

20
00:01:02,575 --> 00:01:03,765
improve healthcare outcomes.

21
00:01:04,125 --> 00:01:07,490
This conversation is especially
timely, as many organizations Including

22
00:01:07,490 --> 00:01:08,740
us and our clients at Newfire.

23
00:01:08,740 --> 00:01:11,590
We'll look at how AI can move
the needle from analytics to

24
00:01:11,640 --> 00:01:13,040
actionable insights in healthcare.

25
00:01:13,500 --> 00:01:14,490
Welcome, Emily.

26
00:01:15,540 --> 00:01:16,400
Thanks, Gordon.

27
00:01:16,410 --> 00:01:17,250
Great to be here.

28
00:01:17,259 --> 00:01:18,870
Excited for the conversation.

29
00:01:19,370 --> 00:01:19,960
Yeah, me too.

30
00:01:19,960 --> 00:01:21,340
I've been looking forward
to this for weeks.

31
00:01:21,840 --> 00:01:25,460
Um, so a little bit of background
for those of you who are listening.

32
00:01:26,429 --> 00:01:29,820
Emily and I actually have crossed
paths multiple times within the

33
00:01:30,210 --> 00:01:31,890
Boston healthcare ecosystem.

34
00:01:32,200 --> 00:01:35,510
Uh, we've actually worked at some
of the same firms, we've consulted

35
00:01:35,510 --> 00:01:36,669
with each other as colleagues.

36
00:01:36,690 --> 00:01:38,790
And so I have a lot of respect for Emily.

37
00:01:38,820 --> 00:01:40,999
So I'm, again, I feel really
privileged to have this opportunity

38
00:01:40,999 --> 00:01:42,770
to have this conversation with her.

39
00:01:43,140 --> 00:01:47,060
And so today's topic is, you know, when
we were looking for guests to explore

40
00:01:47,060 --> 00:01:50,090
this topic, Emily is really one of the
first people that came to mind, because

41
00:01:50,090 --> 00:01:56,170
of her focus on data science and AI and
really, impactful, actionable insights.

42
00:01:56,520 --> 00:01:59,200
Today, we're going to dig into AI's
role in healthcare transformation.

43
00:01:59,780 --> 00:02:02,619
Right, so Emily, you've been at
the forefront in this industry for a

44
00:02:02,619 --> 00:02:06,599
while, and arguably, your entire career
and education has led to this point.

45
00:02:07,130 --> 00:02:11,710
So, how do you see AI evolving from
really just analyzing data and producing

46
00:02:11,710 --> 00:02:14,829
descriptive analytics to driving
actionable business focused solutions?

47
00:02:16,505 --> 00:02:17,675
It's a great question, Gordon.

48
00:02:17,695 --> 00:02:21,185
This is one of my favorite topics
and probably one of the things

49
00:02:21,185 --> 00:02:22,665
I get asked about the most.

50
00:02:23,095 --> 00:02:26,715
Before I answer, let me give a little
bit of an overview of kind of like

51
00:02:26,735 --> 00:02:30,475
where my journey in AI healthcare
has been just for listeners to

52
00:02:30,475 --> 00:02:32,505
see kind of like what I've seen.

53
00:02:32,985 --> 00:02:37,960
Um, I, I started my career in the
imaging world, which is where AI,

54
00:02:37,960 --> 00:02:41,310
I think, really started showing
its first promise in healthcare.

55
00:02:41,680 --> 00:02:44,580
Um, it is the perfect application of AI

56
00:02:44,610 --> 00:02:52,025
images are, um, you know, we
think back to, like, Identifying

57
00:02:52,035 --> 00:02:53,355
cats versus dogs, right?

58
00:02:53,365 --> 00:02:57,205
Like the early, early applications
of AI that were across the world.

59
00:02:57,215 --> 00:02:59,845
Like you can apply them
to medical images as well.

60
00:03:00,375 --> 00:03:05,154
Um, and so I started in neuroimaging
and one of my first jobs out of grad

61
00:03:05,154 --> 00:03:06,834
school was working at IBM Watson Health,

62
00:03:06,975 --> 00:03:10,395
where I was part of their AI imaging
team trying to make new algorithms

63
00:03:10,395 --> 00:03:13,845
that we were bringing to market,
um, in mammography, chest CT.

64
00:03:14,315 --> 00:03:18,775
My career slowly transitioned out of
imaging more to broader health systems,

65
00:03:18,775 --> 00:03:21,795
more broad healthcare applications,
kind of working with digital health

66
00:03:21,795 --> 00:03:25,945
companies, working with, uh, providers
like Cityblock, where Gordon and I

67
00:03:25,955 --> 00:03:30,065
crossed paths, um, and really kind
of seeing more at a global scale

68
00:03:30,155 --> 00:03:34,255
rather than singular applications, how
AI,  and just data science in general

69
00:03:34,255 --> 00:03:35,955
was evolving across the industry.

70
00:03:36,555 --> 00:03:41,465
And so just to say, I did all of my
training like way before LLMs came out.

71
00:03:41,565 --> 00:03:46,565
Um, there is, we are in a new generation
of AI when we're thinking about

72
00:03:46,595 --> 00:03:50,795
LLMs and generative AI and, um, and
so I have a lot of thoughts there.

73
00:03:51,335 --> 00:03:56,775
I think that where we are today is that
there's a lot of excitement about AI.

74
00:03:56,885 --> 00:04:01,760
People are making discoveries, training
algorithms to be incrementally better

75
00:04:01,760 --> 00:04:05,380
than the last and sometimes more than
incrementally at like light speed.

76
00:04:05,960 --> 00:04:11,020
But if you know healthcare, you also
know that healthcare kind of like

77
00:04:11,110 --> 00:04:15,270
exists on somewhat of a delayed timeline
when it comes to data and technology.

78
00:04:15,310 --> 00:04:20,265
A lot of our data and infrastructure
in the industry isn't there yet

79
00:04:20,325 --> 00:04:23,645
to support these super advanced
technological applications.

80
00:04:23,995 --> 00:04:29,865
So, I am kind of coming at this
conversation from a place of, um, I've

81
00:04:29,865 --> 00:04:34,785
kind of, I've seen under the hood of what
our data infrastructure often looks like,

82
00:04:35,035 --> 00:04:38,505
and, and I'm coming at it with this view
of a little bit of skepticism, honestly.

83
00:04:38,895 --> 00:04:43,455
So I'll, I'll give you that as sort
of a backdrop for my perspective.

84
00:04:43,455 --> 00:04:46,165
And it really is just my
perspective that I'm sharing here.

85
00:04:46,475 --> 00:04:51,770
So let me, after I say that, let me give
the question back to Gordon and kind

86
00:04:51,770 --> 00:04:53,320
of where do you want to go with that?

87
00:04:54,150 --> 00:04:54,980
Yeah, absolutely.

88
00:04:55,010 --> 00:04:57,450
So that actually resonates
with me quite a bit.

89
00:04:57,890 --> 00:05:02,700
Um, because as a, uh, someone who's
been building data platforms for

90
00:05:03,200 --> 00:05:06,800
more years than I care to count,
um, we've been asked do many

91
00:05:06,800 --> 00:05:09,880
cutting-edge things, but frequently
the organizations are not ready for it.

92
00:05:09,900 --> 00:05:12,500
So I use the term pragmatic
cutting-edge a lot, and I suspect

93
00:05:12,500 --> 00:05:13,640
that might resonate with you a bit.

94
00:05:14,560 --> 00:05:15,220
I love that.

95
00:05:16,070 --> 00:05:18,060
So let's, let's really
drill into that then, right?

96
00:05:18,110 --> 00:05:22,300
So we, we are really looking
for, you know, solving for

97
00:05:22,350 --> 00:05:24,250
actionable business problems.

98
00:05:24,510 --> 00:05:24,880
Right?

99
00:05:25,120 --> 00:05:28,885
So, you know, what, what, where are
you seeing progress in terms of you

100
00:05:28,915 --> 00:05:30,845
using AI to actually move the needle?

101
00:05:32,385 --> 00:05:36,435
So I think about AI and healthcare
in like these two broad camps.

102
00:05:36,475 --> 00:05:38,575
There's administrative
and there's clinical.

103
00:05:38,575 --> 00:05:42,265
Um, and people are interested in
both and there's really, really

104
00:05:42,265 --> 00:05:43,575
important applications in both.

105
00:05:44,275 --> 00:05:48,650
If we start with administrative,
administrative are the things like, rev

106
00:05:48,650 --> 00:05:54,780
cycle management, billing, coding, um,
even just scheduling patients, like, all

107
00:05:54,780 --> 00:05:58,940
of, all of these things in healthcare that
aren't directly caring for the patient,

108
00:05:58,950 --> 00:06:01,120
but cause a lot of burden on the industry.

109
00:06:02,040 --> 00:06:05,770
Clinical applications are the things
that I think we think about as like

110
00:06:05,770 --> 00:06:09,830
these really sexy, like, going to move
the needle and change patient care.

111
00:06:10,155 --> 00:06:14,345
Those are things like automatically
diagnosing a patient or, you

112
00:06:14,345 --> 00:06:17,305
know, prescribing like the
absolute best treatment regimen

113
00:06:17,305 --> 00:06:18,705
based on a patient's history.

114
00:06:19,745 --> 00:06:23,355
Um, I think that where we are today
is really just the first camp.

115
00:06:23,605 --> 00:06:27,285
The real promise and the early
promise of AI and healthcare is going

116
00:06:27,285 --> 00:06:28,815
to be on that administrative side.

117
00:06:29,675 --> 00:06:34,355
I say that, and it's maybe not as
exciting, like, Oh, why can't we do

118
00:06:34,355 --> 00:06:36,095
like these crazy futuristic things?

119
00:06:36,675 --> 00:06:37,385
I'll get to that.

120
00:06:37,385 --> 00:06:40,645
I think the industry is just not
there yet, but I also wanted to

121
00:06:41,075 --> 00:06:44,725
say with positivity about, or
speak with optimism, I should

122
00:06:44,725 --> 00:06:47,375
say, about those administrative
tasks and the promise there.

123
00:06:47,835 --> 00:06:52,965
And the reason is that part of the reason
that the US healthcare system is just so

124
00:06:52,985 --> 00:06:59,065
complicated, so expensive, so burdensome,
is the tremendous amount of administrative

125
00:06:59,065 --> 00:07:00,835
burden that we have placed on it.

126
00:07:01,525 --> 00:07:07,620
And so actually using AI to alleviate
some of that burden is a great thing.

127
00:07:07,650 --> 00:07:12,120
And I think will actually pay really
meaningful dividends in the end for

128
00:07:12,180 --> 00:07:14,430
patients, providers, the system at large.

129
00:07:14,860 --> 00:07:18,480
Um, so I can go into details there, but
just to say, like, I think about those

130
00:07:18,490 --> 00:07:21,680
two camps, and I really think about the
administrative side is where we're going

131
00:07:21,680 --> 00:07:23,690
to see change in these next few years.

132
00:07:25,450 --> 00:07:27,320
I, you know, I actually
really agree with you.

133
00:07:27,320 --> 00:07:29,150
I see that, I, I see that too.

134
00:07:29,150 --> 00:07:31,880
And I've, I've, I've seen the
statistic that perhaps something

135
00:07:31,890 --> 00:07:34,890
like 40 percent of our healthcare
spend goes into administration.

136
00:07:34,980 --> 00:07:36,800
So it feels like of opportunity there.

137
00:07:37,690 --> 00:07:38,310
That's right.

138
00:07:38,340 --> 00:07:44,050
And I mean, you have to, I think,
believe that a lot of things will

139
00:07:44,050 --> 00:07:49,050
chain together and, and lead to that
trickling down to patients having

140
00:07:49,050 --> 00:07:50,740
cheaper and more affordable care.

141
00:07:51,080 --> 00:07:54,110
But that's what I think
the goal should be.

142
00:07:54,430 --> 00:07:58,120
As an, or as an industry, when we think
about, okay, we're going to really

143
00:07:58,120 --> 00:08:02,670
tackle this administrative problem with
AI, I think that we should say more

144
00:08:02,670 --> 00:08:05,320
than just: let's make the burden lower.

145
00:08:05,340 --> 00:08:08,980
I think that the goal should really
be, and make it cheaper for patients.

146
00:08:09,010 --> 00:08:13,720
Make the whole system, patients included,
less burdened by this extreme cost.

147
00:08:13,730 --> 00:08:16,370
Because you're right, it is up
to 40 percent administrative

148
00:08:16,370 --> 00:08:17,370
costs in some cases.

149
00:08:18,380 --> 00:08:21,710
So at Newfire, I'm heading
up data analytics and AI.

150
00:08:21,740 --> 00:08:25,170
So I have the opportunity to speak to
a lot of our prospects in terms of how

151
00:08:25,170 --> 00:08:28,310
to use AI within their organizations.

152
00:08:28,620 --> 00:08:32,960
And so if, you know, what's your
guidance to them, like what problems

153
00:08:33,190 --> 00:08:36,980
should they look at first in terms
of using AI to, to improve things?

154
00:08:38,090 --> 00:08:41,690
Um, so my advice there is, I think
this is an obvious question for a

155
00:08:41,690 --> 00:08:44,520
lot of business leaders, but it's
worth saying is, you know, what is

156
00:08:44,530 --> 00:08:46,720
the ROI if you solve this problem?

157
00:08:47,180 --> 00:08:50,690
Um, a lot of people, like I said,
they really want to tackle these

158
00:08:50,690 --> 00:08:55,580
really exciting clinical things, and
I don't want to discourage from that,

159
00:08:55,895 --> 00:08:59,455
but we actually don't know what
the ROI is often of solving

160
00:08:59,455 --> 00:09:00,925
some clinical problems with AI.

161
00:09:00,985 --> 00:09:05,155
I think that the ROI is much more
clear for some of these administrative.

162
00:09:05,155 --> 00:09:07,485
So if you're a business leader, I
mean, that really should be the first

163
00:09:07,485 --> 00:09:08,845
question you're asking yourself.

164
00:09:09,435 --> 00:09:13,945
Um, you know, the second question
is kind of this technical

165
00:09:13,945 --> 00:09:15,605
feasibility question, right?

166
00:09:15,695 --> 00:09:21,315
If you are thinking of building and
selling an AI product as a company,

167
00:09:21,850 --> 00:09:25,190
you have to think about do all of
your customers have the technical

168
00:09:25,200 --> 00:09:26,750
infrastructure to support this?

169
00:09:26,750 --> 00:09:28,610
Do they have, like, the
data to support this?

170
00:09:28,640 --> 00:09:31,610
And that is often not the
case right now in healthcare.

171
00:09:31,940 --> 00:09:34,830
There are a lot of health systems
out there who are, like, just

172
00:09:35,060 --> 00:09:36,930
migrating to the cloud, you know.

173
00:09:37,260 --> 00:09:41,790
They, they don't have the tools to
load all their data into some kind of

174
00:09:41,840 --> 00:09:44,500
AI algorithm and get an answer back.

175
00:09:45,010 --> 00:09:45,530
Um.

176
00:09:46,055 --> 00:09:47,135
So that's the second thing.

177
00:09:47,715 --> 00:09:50,515
I think the other thing is
generalizability, you know, is what

178
00:09:50,515 --> 00:09:56,505
you're building something that can
generalize, can safely exist outside

179
00:09:56,505 --> 00:09:59,495
of some very, very small testing
grounds that you've built it in.

180
00:09:59,755 --> 00:10:03,535
That's a trap that a lot of people fall
into, I think, with AI, is they can

181
00:10:03,545 --> 00:10:07,615
build something that performs amazingly,
and then they take it and they try to

182
00:10:07,955 --> 00:10:11,325
have it perform somewhere else and it
doesn't do as well, and I think the

183
00:10:11,325 --> 00:10:16,025
trust starts to really erode in the
industry, um, with AI for those reasons.

184
00:10:16,565 --> 00:10:20,815
And I think the last thing is, like, keep
an eye towards the regulatory concerns

185
00:10:20,865 --> 00:10:22,375
around what you're trying to build.

186
00:10:22,725 --> 00:10:27,245
Those administrative tasks that I
mentioned are often not so, they're

187
00:10:27,245 --> 00:10:31,425
not so prone to regulatory challenges,
whereas the clinical side of things,

188
00:10:31,695 --> 00:10:36,135
that's a big hurdle to get some kind
of AI cleared by the FDA for actual

189
00:10:36,135 --> 00:10:37,905
patient use in the clinical setting.

190
00:10:37,905 --> 00:10:40,815
So, you know, keeping those things
in mind, I think, is really critical.

191
00:10:42,135 --> 00:10:45,435
Now, acknowledging that healthcare
companies, uh, come in lots of different

192
00:10:45,435 --> 00:10:49,755
sizes and different flavors, but I think
about Fisher Price, my first AI project.

193
00:10:50,195 --> 00:10:55,295
Um, which C level officer would
you typically think would be like a

194
00:10:55,295 --> 00:10:57,245
good first customer for a user AI?

195
00:10:58,405 --> 00:10:59,915
That's a great question.

196
00:10:59,975 --> 00:11:04,365
I have, um, I've been talking to some
folks about this recently, and I was

197
00:11:04,365 --> 00:11:09,775
just invited to, like, a roundtable to
listen to chief information officers and

198
00:11:09,775 --> 00:11:11,435
how they're being approached about AI.

199
00:11:12,055 --> 00:11:15,195
And interestingly, I think what's
happening right now is I think

200
00:11:15,195 --> 00:11:19,630
it's the CFOs of most organizations
who are being approached about

201
00:11:19,650 --> 00:11:21,520
AI and the decision-makers

202
00:11:21,790 --> 00:11:25,710
and that comes back to that ROI question
and the fact that a lot of these are

203
00:11:25,710 --> 00:11:29,720
really being targeted for a lot of
these applications are being targeted

204
00:11:29,720 --> 00:11:33,630
for administrative tasks, internal
efficiency gains, things like that.

205
00:11:34,130 --> 00:11:40,060
So what you might think of as like
these CIO, CTO folks who are first

206
00:11:40,060 --> 00:11:43,570
approached decision-makers, I think we're
actually seeing that it's more people

207
00:11:43,570 --> 00:11:45,310
on the financial decision making side.

208
00:11:46,640 --> 00:11:48,020
Yeah, that, that resonates for sure.

209
00:11:48,050 --> 00:11:51,020
You know, part of the pur purpose
of this, uh, this podcast is really

210
00:11:51,020 --> 00:11:54,080
to give our listeners some ideas of
where they can get started, right?

211
00:11:54,380 --> 00:11:55,750
Pragmatic advice.

212
00:11:56,110 --> 00:11:59,380
In that vein, do you have any specific
examples from your recent history

213
00:11:59,380 --> 00:12:02,800
where AI has made an impact in terms
of patient outcomes or operational

214
00:12:02,800 --> 00:12:05,160
efficiency or just reduction of toil?

215
00:12:06,250 --> 00:12:07,060
Definitely.

216
00:12:07,090 --> 00:12:10,270
So some of the applications that I've
seen that I think have been really

217
00:12:10,270 --> 00:12:15,745
exciting in AI, that are, they, they
just are adjacent to clinical,  but they

218
00:12:15,745 --> 00:12:20,425
really are administrative applications,
are things that help reduce the

219
00:12:20,425 --> 00:12:22,755
burden of clinical documentation.

220
00:12:22,945 --> 00:12:27,300
So, really solving this problem of
providers are getting burned out,

221
00:12:27,360 --> 00:12:31,920
they're spending way too much time
having to write up patient notes, sift

222
00:12:31,930 --> 00:12:37,420
through patient notes and synthesize
past medical information about a patient.

223
00:12:37,730 --> 00:12:42,660
When you kind of boil that down, those
are information synthesis problems,

224
00:12:42,660 --> 00:12:43,850
which is what AI is really good at.

225
00:12:44,490 --> 00:12:48,570
The applications that I've seen that
do things like ambient documentation.

226
00:12:48,700 --> 00:12:53,120
So for example, a voice recording,
just like you and I are doing right

227
00:12:53,120 --> 00:12:58,265
now, of a patient-doctor conversation,
which gets translated to text.

228
00:12:58,335 --> 00:12:59,045
That's easy.

229
00:12:59,045 --> 00:13:04,085
That's been around for a while, but is
then with an LLM translated into medical

230
00:13:04,095 --> 00:13:06,295
jargon that can go into a patient's chart.

231
00:13:06,645 --> 00:13:08,055
That's incredibly powerful.

232
00:13:08,095 --> 00:13:11,665
That, you know, that saves
the doctor so, so much time.

233
00:13:12,065 --> 00:13:16,705
And I think I've seen amazing reviews
and feedback from clinicians themselves

234
00:13:16,705 --> 00:13:19,705
on how powerful that type of technology
is going to be in the clinic.

235
00:13:20,565 --> 00:13:24,065
I've also seen a lot of applications
of AI that I think are really

236
00:13:24,065 --> 00:13:28,840
promising from like an investment and
investability perspective of revenue

237
00:13:28,840 --> 00:13:31,360
cycle management, billing and coding.

238
00:13:31,650 --> 00:13:35,820
Those are heavily manual processes
right now that are prone to error

239
00:13:36,150 --> 00:13:37,950
that I think AI is excelling at.

240
00:13:38,710 --> 00:13:42,850
And then this is my favorite example, just
kind of throwing back to my imaging days.

241
00:13:42,880 --> 00:13:47,605
This is not AI the most recent example
that, I encountered this years ago,

242
00:13:48,045 --> 00:13:53,315
but I remember at an organization I was
at, we were working on really advanced,

243
00:13:53,315 --> 00:13:58,235
really cutting-edge neuroimaging where
we were developing technologies that

244
00:13:58,235 --> 00:14:02,925
could, like, automatically segment images
of the human brain into all of these

245
00:14:02,925 --> 00:14:05,425
amazing, very descriptive statistics.

246
00:14:05,925 --> 00:14:11,210
And it was hard to sell that because it
was hard to tell a clinician why this

247
00:14:11,230 --> 00:14:15,570
would help improve patient care or like
why this would save them money for many

248
00:14:15,570 --> 00:14:17,930
nuances related to like human brain stuff.

249
00:14:18,260 --> 00:14:21,440
But um, it was amazing technology
that couldn't find a home.

250
00:14:22,370 --> 00:14:28,420
Conversely, there was this technology
coming out at the same time that could

251
00:14:28,735 --> 00:14:35,185
take a cue of like a hundred images in an
emergency room and automatically identify

252
00:14:35,185 --> 00:14:37,345
if any one of them was a brain bleed.

253
00:14:37,945 --> 00:14:43,055
And just for those listening who don't
know, almost any of us could be trained to

254
00:14:43,055 --> 00:14:45,915
a spot a major brain bleed pretty easily.

255
00:14:45,915 --> 00:14:48,395
It's a very visually obvious thing.

256
00:14:49,015 --> 00:14:53,495
But so what this algorithm did, it
wasn't like finding some really hidden

257
00:14:53,705 --> 00:14:58,530
issue, but it would find something that
was incredibly emergent that might be

258
00:14:58,530 --> 00:15:03,330
really low in the radiologist's read
queue and pop it to the top and say,

259
00:15:03,380 --> 00:15:07,280
basically, if you don't read this image,
this patient might die in the next hour.

260
00:15:07,750 --> 00:15:11,210
And that's like an
orchestration task, right?

261
00:15:11,250 --> 00:15:15,240
That's like a clinical workflow
task, but it was such a better at

262
00:15:15,240 --> 00:15:19,340
the time application of AI, if you
really wanted to, like, help patients.

263
00:15:19,610 --> 00:15:21,480
So I just put those two side by side.

264
00:15:21,480 --> 00:15:24,800
It's one of my favorite examples
here of, you know, sometimes the

265
00:15:24,800 --> 00:15:26,460
simpler solution, the better one.

266
00:15:26,940 --> 00:15:27,420
I love that.

267
00:15:27,600 --> 00:15:28,000
I love that.

268
00:15:28,000 --> 00:15:30,920
I mean, it reminds me that we should be
always thinking about kind of the basic

269
00:15:30,960 --> 00:15:32,820
variables that drive outcomes, right?

270
00:15:32,880 --> 00:15:36,450
Time, timeliness, reduction in
effort, all these things, right?

271
00:15:36,960 --> 00:15:41,170
So, you know, uh, stealing from the
product manager book, so looking at

272
00:15:41,170 --> 00:15:45,880
your typical, let's say healthcare
provider, what person or role would

273
00:15:45,880 --> 00:15:50,320
you want to target to, uh, delight
with AI within that organization?

274
00:15:50,330 --> 00:15:51,770
Whose job do you make it easier?

275
00:15:52,040 --> 00:15:53,610
How do you help some be more successful?

276
00:15:54,690 --> 00:15:55,860
That's such a good question.

277
00:15:57,840 --> 00:16:01,740
So, personally, I think we should
be really making things for patients

278
00:16:01,740 --> 00:16:06,390
and physicians, like, the people
delivering and receiving care and

279
00:16:06,640 --> 00:16:12,280
physicians need so much tooling
like this to help with burnout.

280
00:16:12,550 --> 00:16:16,420
So I think delighting the physician,
the ambient documentation, question

281
00:16:16,450 --> 00:16:18,650
or example I gave is perfect there.

282
00:16:19,035 --> 00:16:22,185
However, usually physicians are
not the financial decision-makers,

283
00:16:22,205 --> 00:16:23,855
they're not purchasing your product.

284
00:16:24,175 --> 00:16:29,165
And so, I do think that you really have
to be able to always show delight to

285
00:16:29,165 --> 00:16:31,435
the customer that's paying in the end.

286
00:16:31,805 --> 00:16:37,485
And so, being able to say to them, this
is going to allow your physicians to

287
00:16:37,665 --> 00:16:43,875
spend X amount more time with patients,
or, show some real financial ROI there is

288
00:16:43,905 --> 00:16:46,075
really, really important to keep in mind.

289
00:16:46,675 --> 00:16:46,925
Yeah.

290
00:16:47,305 --> 00:16:49,145
Thank you for that.

291
00:16:49,475 --> 00:16:50,985
I, really, that really does
make a lot of sense to me.

292
00:16:50,995 --> 00:16:53,795
You know, I personally have always
loved the lens of like trying to delight

293
00:16:53,795 --> 00:16:55,365
somebody because we could understand that.

294
00:16:55,765 --> 00:16:58,625
But let's take a, let's get a little
speculative for a second, right?

295
00:16:58,635 --> 00:17:00,375
Think about the future AI in healthcare.

296
00:17:00,775 --> 00:17:03,575
What do you think, what
are you excited about?

297
00:17:04,475 --> 00:17:07,175
What do you think are the big
opportunities and challenges

298
00:17:07,175 --> 00:17:08,005
we might be able to tackle?

299
00:17:10,265 --> 00:17:14,875
So what I'm excited about, this
is, this is a hard question.

300
00:17:14,885 --> 00:17:17,945
I've actually been asked this a
few different times and I'm always

301
00:17:17,975 --> 00:17:21,865
kind of evolving my perspective,
I think, on a regular basis.

302
00:17:22,625 --> 00:17:26,025
Let me zoom out before I answer what
I'm most excited about and tell you what

303
00:17:26,025 --> 00:17:28,625
I think some of the biggest problems
are and we'll, we'll work backwards.

304
00:17:29,675 --> 00:17:30,775
Always lead with problems.

305
00:17:32,380 --> 00:17:37,405
Some of the things in healthcare
that are just, in my mind, so broken

306
00:17:37,425 --> 00:17:40,645
and in need of change really do
come down to what we started talking

307
00:17:40,645 --> 00:17:42,395
about was data infrastructure, right?

308
00:17:42,545 --> 00:17:44,885
Data fragmentation and
data infrastructure.

309
00:17:45,435 --> 00:17:48,065
This, like, shows up in
a lot of different ways.

310
00:17:48,115 --> 00:17:54,145
This shows up as technology not being
able to scale across organizations,

311
00:17:54,145 --> 00:17:56,045
technology not being able to generalize.

312
00:17:56,605 --> 00:18:00,155
And one of the things, I think,
that's becoming more and more

313
00:18:00,155 --> 00:18:04,245
apparent in the US is kind of health
disparities across the country.

314
00:18:04,455 --> 00:18:09,335
There are growing parts of
America that are losing doctors.

315
00:18:09,725 --> 00:18:11,465
What are those people doing?

316
00:18:11,465 --> 00:18:14,595
Like, that, that, that's going to cause
more chronic conditions and everything.

317
00:18:15,155 --> 00:18:19,455
And so I think when we think about like,
where do we really want to be, like,

318
00:18:19,465 --> 00:18:23,825
where's technology really, really going
to help us in the next 20, 30 years?

319
00:18:25,305 --> 00:18:30,000
Personally, I'm not sure that it's going
to be in these super futuristic things of

320
00:18:30,000 --> 00:18:35,060
like, doing a full body scan and finding
like the one cell in your body that is

321
00:18:35,330 --> 00:18:39,220
potentially cancerous that I think is like
a lot of the sci-fi things we think about.

322
00:18:39,530 --> 00:18:43,140
I think the promise is really,
how do we get equal care

323
00:18:43,505 --> 00:18:47,465
to people that live in every single
setting in America across the whole

324
00:18:47,515 --> 00:18:52,835
country when there's obvious resource
deficiencies and the supply and demand

325
00:18:52,835 --> 00:18:54,595
is just like not the same everywhere?

326
00:18:54,765 --> 00:18:59,495
So if someone can figure out how to fit
AI into that problem, like, that is what

327
00:18:59,495 --> 00:19:02,045
I think will really, really drive change.

328
00:19:02,365 --> 00:19:06,305
Really lofty goal, but it's probably the
thing I think about the most when I think

329
00:19:06,315 --> 00:19:08,985
about AI and tech making a real impact.

330
00:19:09,535 --> 00:19:12,515
So, to put words in your mouth, I
think you're reminding us that AI

331
00:19:12,515 --> 00:19:15,145
is a tool, not the end, the outcome.

332
00:19:16,470 --> 00:19:18,800
Yes, I think that's a great way to put it.

333
00:19:18,800 --> 00:19:22,900
I mean, there are so many healthtech
and healthcare companies out there

334
00:19:22,900 --> 00:19:27,400
right now that are using AI internally,
but they're not like AI companies.

335
00:19:27,440 --> 00:19:29,870
And I think that is the right
way to be thinking about it.

336
00:19:29,870 --> 00:19:32,460
You know, use AI to get done
the things that you need to

337
00:19:32,460 --> 00:19:33,880
get done and do them better.

338
00:19:34,335 --> 00:19:37,705
But sometimes when we have a hammer,
everything looks like a nail, and

339
00:19:37,725 --> 00:19:42,245
we, you don't need to become an AI
company to really change healthcare.

340
00:19:43,195 --> 00:19:46,365
Let me share with you some, I'm seeing
a little bit, I'd like to know if

341
00:19:46,365 --> 00:19:48,065
you're having similar experience.

342
00:19:48,625 --> 00:19:51,495
So, you know, in Newfire, we, we
do talk again, we talked to a lot

343
00:19:51,515 --> 00:19:52,775
of clients about implementing AI.

344
00:19:54,090 --> 00:19:57,540
I've, I've noticed a funny thing, it's
because you mentioned data foundations

345
00:19:57,870 --> 00:20:00,800
and fundamentals that made me think
about this, was that as we talk about

346
00:20:00,800 --> 00:20:04,090
AI, frequently the conversation starts
becoming like, oh what's the state of

347
00:20:04,090 --> 00:20:08,010
your data platform, your data warehouse,
your data quality, and it's driving

348
00:20:08,010 --> 00:20:11,880
more awareness of these systems that
have been languishing for a long time.

349
00:20:12,280 --> 00:20:13,250
Are you seeing the same thing?

350
00:20:14,440 --> 00:20:15,000
Yes.

351
00:20:15,020 --> 00:20:17,190
I think that that's a great observation.

352
00:20:17,190 --> 00:20:24,570
People are, to your point, like, have
kind of allowed their systems to deprecate

353
00:20:24,570 --> 00:20:26,480
It just not be as advanced anymore.

354
00:20:26,480 --> 00:20:30,180
And now they're like, Oh no, I can't, I
can't apply the cutting-edge technologies.

355
00:20:30,180 --> 00:20:31,250
And that's okay.

356
00:20:31,250 --> 00:20:34,900
I think that this is a great impetus
for us to all look and say, how

357
00:20:34,900 --> 00:20:36,630
do we build a better foundation?

358
00:20:36,920 --> 00:20:40,430
So we definitely are
seeing the same thing.

359
00:20:40,440 --> 00:20:44,090
And that's why the problems that
excite me are actually really these

360
00:20:44,130 --> 00:20:48,500
infrastructural problems and solutions
so that we can enable these things

361
00:20:48,640 --> 00:20:50,790
that are much farther down the road.

362
00:20:52,675 --> 00:20:55,905
So let's take some of the implementation
challenges off the table for a second.

363
00:20:56,435 --> 00:20:59,385
Where do you wish you
could build an AI solution?

364
00:20:59,805 --> 00:21:01,855
Well, what problem do you wish
you could address with AI?

365
00:21:03,965 --> 00:21:05,585
That's a really good question.

366
00:21:06,875 --> 00:21:10,435
I will, I'll go back to what I said
about like this really future vision.

367
00:21:10,895 --> 00:21:15,445
I wish that we could figure out, and I
mean plenty of people are working on this

368
00:21:15,445 --> 00:21:20,345
just to be clear, and I think plenty of
people like share this view, this is not

369
00:21:20,675 --> 00:21:26,645
something I just thought of, but this
idea of like a person who lives in rural

370
00:21:27,310 --> 00:21:34,720
Arkansas who has a stroke and cannot
get seen by a neuroradiologist in time.

371
00:21:34,990 --> 00:21:39,500
Yet neuroradiologists exist in this
country who could read their scan

372
00:21:39,500 --> 00:21:46,405
and advise, bringing expertise to
places where it currently does not

373
00:21:46,405 --> 00:21:51,355
exist is where I would really want to
focus my time in like an ideal world.

374
00:21:51,775 --> 00:21:53,385
And it's not just telehealth.

375
00:21:53,425 --> 00:21:57,355
It's not just like connecting,
uh, a radiologist who's remote.

376
00:21:57,685 --> 00:22:01,365
It's maybe it's not even connecting a
radiologist, maybe it's actually making

377
00:22:01,365 --> 00:22:04,605
an AI that can automatically read that
image, which people are working on.

378
00:22:04,935 --> 00:22:10,985
Maybe it's creating better surgical
techniques and tooling and suites that

379
00:22:11,285 --> 00:22:15,975
do more robotic tech surgeries, right,
that don't need as much human expertise.

380
00:22:16,015 --> 00:22:16,895
I don't know.

381
00:22:17,205 --> 00:22:20,305
But I do think that the thing
that's most exciting to me is like

382
00:22:20,765 --> 00:22:23,235
fixing this supply chain issue.

383
00:22:24,165 --> 00:22:27,355
Thanks for sharing that because it
makes me, it allows me to sort of

384
00:22:27,355 --> 00:22:30,305
categorize, um, you know, something
we, we see a lot at Newfire.

385
00:22:30,325 --> 00:22:32,465
So we have the privilege to work
with a lot of healthcare startups.

386
00:22:32,935 --> 00:22:35,275
And I realized what a lot of
them are trying to do is bring

387
00:22:35,315 --> 00:22:37,395
expertise to the problem, right?

388
00:22:37,395 --> 00:22:39,605
They're trying to solve for this expert
gap and they're trying to bring it to

389
00:22:39,605 --> 00:22:43,445
the patient, which to me, intuitively
feels like the right thing to do.

390
00:22:44,085 --> 00:22:45,215
I totally agree.

391
00:22:45,225 --> 00:22:49,135
Yes, I've seen some amazing healthtech
startups out there that are, this

392
00:22:49,135 --> 00:22:52,965
is exactly their model, is you know,
there's patients out there that need

393
00:22:52,965 --> 00:22:57,495
a certain type of care and either the
place that they live or the insurance

394
00:22:57,495 --> 00:23:02,725
coverage they have, or maybe both
prohibits them from getting those things.

395
00:23:02,725 --> 00:23:06,695
And I mean, and even at Morgan Health,
we have done research and we've learned

396
00:23:06,705 --> 00:23:14,605
that physical lack of access to care is
a key driver of poor outcomes and, you

397
00:23:14,605 --> 00:23:20,355
know, misdiagnoses more so than not being
able to afford care, which is interesting

398
00:23:20,365 --> 00:23:23,885
because sometimes we think of them as
being the same thing, but physical access

399
00:23:23,905 --> 00:23:25,235
is different than financial access.

400
00:23:26,175 --> 00:23:29,865
I have colleagues in Canada who are
in healthcare and they are trying

401
00:23:29,865 --> 00:23:32,895
to tackle this problem is very,
yeah, a front-of-mind for them

402
00:23:32,915 --> 00:23:36,400
because of course how disparate
Canada is in terms of geography.

403
00:23:37,210 --> 00:23:38,680
Oh, that's really interesting.

404
00:23:38,680 --> 00:23:38,850
Yeah.

405
00:23:38,850 --> 00:23:41,200
I'm sure it's an even bigger
challenge for them there.

406
00:23:41,300 --> 00:23:44,690
Particularly around things
like LGBTQ  care, just having a

407
00:23:44,690 --> 00:23:47,660
primary care physician, if you're
in northern British Columbia.

408
00:23:48,670 --> 00:23:49,210
Yeah.

409
00:23:49,420 --> 00:23:52,310
I mean, and you know where we
see this a lot actually is, is

410
00:23:52,310 --> 00:23:54,120
mental health in the US too.

411
00:23:54,500 --> 00:24:00,330
There are, there's kind of like a dearth
of local mental health providers who

412
00:24:00,730 --> 00:24:04,960
will take insurance, and then when
you talk about specific populations,

413
00:24:04,960 --> 00:24:08,830
like LGBTQ populations, it's like you
have an even harder time of finding

414
00:24:08,830 --> 00:24:10,430
somebody near you in your network.

415
00:24:10,710 --> 00:24:14,850
And so connecting expertise
across like geographies is huge.

416
00:24:15,750 --> 00:24:20,140
Emily, similar to me, you've been
both a provider of technology,

417
00:24:20,580 --> 00:24:22,190
also a consumer of technology.

418
00:24:22,460 --> 00:24:25,820
So let me give you an opportunity
as a customer and a consumer of

419
00:24:25,820 --> 00:24:29,900
technology, of solutions, what do you
want companies like Newfire to build?

420
00:24:30,010 --> 00:24:32,230
How can we make a bigger
difference in this industry?

421
00:24:33,080 --> 00:24:33,850
Scalable

422
00:24:33,850 --> 00:24:35,630
Scalable data platforms.

423
00:24:35,660 --> 00:24:41,740
I think something I have experienced, and
you alluded to this earlier, is so many

424
00:24:41,740 --> 00:24:45,735
health tech companies are out there, and
this is, completely not a knock on them.

425
00:24:45,785 --> 00:24:46,905
I I've lived it.

426
00:24:46,915 --> 00:24:51,545
I understand why these decisions are made,
but so many health check companies are out

427
00:24:51,545 --> 00:24:54,465
there building internally from scratch.

428
00:24:54,745 --> 00:24:57,255
They, they're coming in and they're
saying, how do I build this, like,

429
00:24:57,295 --> 00:25:02,115
bespoke something, data platform,
data warehouse for my needs.

430
00:25:02,565 --> 00:25:06,615
And, as I said, there's many reasons
why people choose to build versus

431
00:25:06,615 --> 00:25:12,975
buy, but it really, I think, is not
always a good long-term investment.

432
00:25:13,405 --> 00:25:18,045
And I, I wish that there were an entity
out there, and maybe it's Newfire, who

433
00:25:18,075 --> 00:25:23,355
could build more kind of scalable data
platforms so that these early healthtech

434
00:25:23,375 --> 00:25:28,155
start, healthtech startups that have
a lot of promise don't actually need

435
00:25:28,155 --> 00:25:33,960
to burn all that capital building what
they think they need to do so bespoke.

436
00:25:34,320 --> 00:25:37,240
Um, I could, you know, we could do a
whole podcast on build versus buy data

437
00:25:37,900 --> 00:25:40,610
warehouses, so I won't go further.

438
00:25:41,445 --> 00:25:44,710
I, I, I think that's that, you
know, and I do think the, the

439
00:25:44,710 --> 00:25:45,940
world is moving in that direction.

440
00:25:45,940 --> 00:25:47,650
That's one of the benefits
of having gone remote.

441
00:25:47,680 --> 00:25:50,690
Of course, there's a lot of costs
having gone remote, but there's an

442
00:25:50,710 --> 00:25:54,070
opportunity to pull in expertise of
all types from around the world and,

443
00:25:54,160 --> 00:25:56,825
you know, in different time zones, um,
you're starting to bear some fruit.

444
00:25:58,385 --> 00:25:59,825
Yeah, yeah, definitely.

445
00:25:59,865 --> 00:26:02,605
I mean, the world has changed quite a lot.

446
00:26:02,605 --> 00:26:06,205
I will say I do so many people
are still, you know, building

447
00:26:06,215 --> 00:26:08,235
the same thing in like silos.

448
00:26:11,435 --> 00:26:16,245
Well, so Emily, do you have any other,
uh, any, uh, final words you want to

449
00:26:16,245 --> 00:26:17,835
share as we wrap this conversation up?

450
00:26:18,215 --> 00:26:19,655
30 minutes goes by really fast.

451
00:26:20,415 --> 00:26:21,705
It goes by so fast.

452
00:26:21,735 --> 00:26:26,345
I would say I, I hope everybody like
stays excited, stays hungry about

453
00:26:26,565 --> 00:26:31,355
all of the amazing opportunities
there are right now for AI and just

454
00:26:31,365 --> 00:26:33,025
data in general and healthcare.

455
00:26:33,345 --> 00:26:36,735
I'm a, I'm a first
principles type of woman.

456
00:26:36,735 --> 00:26:39,905
I think a lot about like,
"what is my base case?"

457
00:26:39,945 --> 00:26:42,275
anytime I'm trying to
solve a complex problem.

458
00:26:42,755 --> 00:26:46,140
And I think approaching things that
way when we're thinking about these

459
00:26:46,140 --> 00:26:50,260
like lofty, complex technology
problems is really helpful and

460
00:26:50,280 --> 00:26:55,380
grounding and sort of actually driving
more effective solutions early on.

461
00:26:55,430 --> 00:26:58,330
And, often your base case is going
to come down to infrastructure,

462
00:26:58,340 --> 00:26:59,540
which we talked about a lot.

463
00:26:59,860 --> 00:27:04,030
So that's my, my parting wisdom is to
try to approach problems in that way.

464
00:27:04,640 --> 00:27:07,290
Yeah, this is not a, uh, this
is not a 30-day mission, right?

465
00:27:07,290 --> 00:27:09,670
This is a lifetime mission and
none of us can do it by ourselves.

466
00:27:10,580 --> 00:27:11,820
Yes, definitely.

467
00:27:11,860 --> 00:27:17,260
I, I think that state working in
healthcare data is, um, there will

468
00:27:17,270 --> 00:27:19,180
always be work to be done for sure.

469
00:27:19,710 --> 00:27:20,910
Well, Emily, thank you so much.

470
00:27:20,940 --> 00:27:22,640
This has been an awesome conversation.

471
00:27:22,980 --> 00:27:25,320
I really hope that our listeners
have gotten something out of this.

472
00:27:25,640 --> 00:27:28,245
I think some of the topics
I've heard were AI is here.

473
00:27:28,825 --> 00:27:30,995
It's making a big difference,
but you need to be careful

474
00:27:30,995 --> 00:27:32,335
about what problems you tackle.

475
00:27:32,625 --> 00:27:35,365
Don't be afraid of focusing on
efficiency as that may have the biggest

476
00:27:35,375 --> 00:27:36,725
impact on your patient outcomes.

477
00:27:37,045 --> 00:27:40,355
And finally, look for partners who have
the skill sets to help you be successful

478
00:27:40,425 --> 00:27:41,815
like our clients at Newfire have.

479
00:27:43,905 --> 00:27:44,775
Yes, that's great.

480
00:27:44,775 --> 00:27:45,245
Thanks Gordon.

481
00:27:45,245 --> 00:27:46,815
This was a great conversation.

482
00:27:47,285 --> 00:27:47,885
Thanks Emily.

483
00:27:47,895 --> 00:27:49,015
I hope you have a great day.

484
00:27:49,330 --> 00:27:49,670
Cheers.

485
00:27:49,780 --> 00:27:50,190
Thanks, Gordon.

486
00:27:50,470 --> 00:27:50,690
You too.

487
00:27:52,390 --> 00:27:53,170
Thanks for tuning in.

488
00:27:53,300 --> 00:27:55,860
We hope today's conversation
with Emily Lindemayer, Executive

489
00:27:55,860 --> 00:27:59,110
Director of Data Healthcare
Innovation at JPMorgan Chase &nCo.

490
00:27:59,380 --> 00:28:03,060
has given you a fresh perspective on
how AI is transforming health, starting

491
00:28:03,060 --> 00:28:06,440
with the foundational need to streamline
data and reduce administrative burdens.

492
00:28:06,890 --> 00:28:09,630
Stay tuned for more episodes where
we continue to explore the toughest

493
00:28:09,630 --> 00:28:12,000
challenges and smartest solutions
in business and technology.

494
00:28:12,390 --> 00:28:13,240
Like and subscribe.

495
00:28:13,730 --> 00:28:16,390
Until next time, keep innovating
and solving the hard problems.

496
00:28:16,790 --> 00:28:19,840
This is Hard Problems,
Smart Solutions, The Newfire

497
00:28:19,840 --> 00:28:20,137
Podcast.