Meeting Title: Sunstone x Brainforge Project Kickoff Date: 2026-05-12 Meeting participants: Robert Tseng, Awaish Kumar, Greg Stoutenburg, Jake’s Notetaker, Jake Sandler, eakta , Kim Maillie, Lindsay Ozyck, jodycrowley, Jess Riegel, celialeeks


WEBVTT

1 00:01:18.300 00:01:19.090 Robert Tseng: Hey guys.

2 00:01:19.090 00:01:20.080 Greg Stoutenburg: Hello again.

3 00:01:20.350 00:01:21.799 Robert Tseng: Oh, we all put on our glasses.

4 00:01:21.800 00:01:23.920 Greg Stoutenburg: Check it out. Robert, I didn’t know you had glasses.

5 00:01:24.930 00:01:27.409 Robert Tseng: They’re just blue light glasses.

6 00:01:27.590 00:01:28.419 Greg Stoutenburg: Okay.

7 00:01:29.320 00:01:32.330 Robert Tseng: when I put them on when my eyes are feeling kind of strained.

8 00:01:33.710 00:01:35.339 Greg Stoutenburg: That’s basically what these are, I can see fine.

9 00:01:35.500 00:01:38.940 Greg Stoutenburg: I mean, there’s, like, there’s, like, a little bit of a lift, but, like, only a little.

10 00:01:39.190 00:01:39.610 Robert Tseng: Yeah.

11 00:01:39.610 00:01:40.380 Greg Stoutenburg: Hey, Jake.

12 00:01:46.370 00:01:49.780 Greg Stoutenburg: I feel like I’m in a shadow in this office, let me… I don’t like that.

13 00:01:51.220 00:01:52.129 Jake Sandler: Can you hear me?

14 00:01:52.720 00:01:53.660 Robert Tseng: Now we can hear you.

15 00:01:53.830 00:01:55.890 Jake Sandler: Oh, okay. Hey, how’s it going, guys?

16 00:01:57.140 00:01:57.760 Robert Tseng: Good.

17 00:01:59.960 00:02:01.579 Jake Sandler: Awash, good to see you.

18 00:02:01.860 00:02:04.410 Awaish Kumar: Yeah, good to see you. How about you? How are you doing?

19 00:02:05.740 00:02:09.529 Jake Sandler: It’s another day in Sunstone Paradise.

20 00:02:13.040 00:02:13.430 Lindsay Ozyck: I am.

21 00:02:13.430 00:02:14.360 Jake Sandler: That’s good.

22 00:02:15.270 00:02:15.940 Lindsay Ozyck: Bye!

23 00:02:15.940 00:02:16.979 Jake Sandler: Hey, Lindsay!

24 00:02:16.980 00:02:18.100 Lindsay Ozyck: How’s it going?

25 00:02:18.500 00:02:19.710 Jake Sandler: Good, how are you?

26 00:02:19.710 00:02:20.470 Lindsay Ozyck: Good.

27 00:02:22.520 00:02:26.790 Jake Sandler: As the Brainforge team, I’ll do sort of formal introductions when everybody gets on.

28 00:02:27.140 00:02:27.690 Lindsay Ozyck: Great.

29 00:02:29.970 00:02:32.989 Jake Sandler: Thanks for shooting over the org chart earlier, Lindsay.

30 00:02:33.510 00:02:35.240 Jake Sandler: Very helpful. Hey, Jody.

31 00:02:35.390 00:02:36.310 Lindsay Ozyck: Happy to help.

32 00:02:41.300 00:02:46.809 Jake Sandler: This is the Brainforge team. I just tell them Lindsay, once everybody joins, we’ll do sort of formal introductions around…

33 00:02:47.700 00:02:48.469 jodycrowley: It’s good.

34 00:02:52.800 00:02:55.360 eakta : Me, that’s green. Oh, that’s okay.

35 00:02:55.740 00:02:56.550 Jake Sandler: Hector.

36 00:02:57.050 00:02:58.290 eakta : Hello!

37 00:02:58.460 00:02:59.210 eakta : I’m coming.

38 00:02:59.210 00:02:59.870 Jake Sandler: don’t know.

39 00:03:00.010 00:03:01.510 eakta : As soon as logging in.

40 00:03:02.020 00:03:02.810 Jake Sandler: Perfect.

41 00:03:05.140 00:03:08.989 Jake Sandler: And I think the last person we’re waiting for is Kim Malley.

42 00:03:09.130 00:03:10.439 eakta : She’s here. She’s here.

43 00:03:10.610 00:03:11.830 Jake Sandler: Oh, hey Kim.

44 00:03:12.710 00:03:14.349 eakta : We’re just all in the conference room.

45 00:03:15.510 00:03:21.419 Lindsay Ozyck: I’ll actually join you guys. I didn’t realize how many people were gonna be on this meeting that are in admin.

46 00:03:21.900 00:03:23.720 Lindsay Ozyck: So, I’ll be right over.

47 00:03:23.720 00:03:25.129 Robert Tseng: Where is the office?

48 00:03:26.420 00:03:27.410 Jake Sandler: It’s unlimited.

49 00:03:27.560 00:03:28.510 Jake Sandler: Rockville.

50 00:03:28.510 00:03:30.620 Robert Tseng: Oh, it’s the Rockville, okay, it’s the Rocksville one, okay.

51 00:03:30.620 00:03:36.640 Jake Sandler: Yeah, so the clinic… the clinic and the admin offices are in different locations, but they’re, like, 5 minutes apart in Rockville.

52 00:03:36.640 00:03:38.840 Robert Tseng: Oh, okay. Got it.

53 00:03:38.910 00:03:39.880 Jake Sandler: Yeah.

54 00:03:41.050 00:03:55.950 Jake Sandler: Well, thank you, everybody. I know there’s a lot going on for joining, for this kickoff, and hopefully this is fun for everybody, and will relieve some of the anxiety that we’re all feeling about the systems at Sunstone.

55 00:03:56.090 00:04:09.590 Jake Sandler: So, you guys have each heard… I’ve talked to each of you about this project, and so this kickoff is really to bring in folks who have intimate knowledge of where patient data lives in the various places across the Sunstone ecosystem.

56 00:04:09.590 00:04:17.809 Jake Sandler: It’s for you all to meet the BrainForge team, and we’ll do introductions in a moment, and then really be able to ask questions about what this project is.

57 00:04:17.850 00:04:22.229 Jake Sandler: And just to set the tone at the top, for the next 6 to 8 weeks, we really anticipate

58 00:04:23.030 00:04:40.289 Jake Sandler: disrupting your guys’ workflow very… as little as possible, so it’s really just going to be asking questions, poking around, but knowing there’s a real-time migration, knowing there’s other key projects. We’ve talked a lot as a team with Brainforge about making sure this is in parallel to that, and isn’t going to…

59 00:04:40.290 00:04:50.550 Jake Sandler: disrupt that work. Hopefully, I think, in the medium and long term, this could help a lot of that work, but that’s to come later, and this is not going to be getting involved in that yet.

60 00:04:50.810 00:05:08.970 Jake Sandler: So without further ado, I will introduce… I’ll let Robert, who leads Brainforge, sort of introduce himself and the rest of the team, and then it would be great if everybody could give the Brainforge team a quick overview of your role and tenure at Sunstone as well.

61 00:05:09.890 00:05:11.549 Jake Sandler: So, Robert, take it away.

62 00:05:12.060 00:05:26.079 Robert Tseng: Great. Thanks, Jake. And, good to see you, Ekta. And I think Kim’s there, and I’ve met Kim before, too. So, yeah, I think, I guess I’ll start off with, yeah, I lead Brainforge. We are, helping

63 00:05:26.080 00:05:37.149 Robert Tseng: we do data and AI implementations for clients, and we have a lot of experience in the healthcare space. So, as I kind of maybe have mentioned to a few of you before.

64 00:05:37.150 00:05:52.349 Robert Tseng: We’ve worked extensively with… with Salesforce, and other… other ERP CRM systems, and then, kind of, we have worked, kind of, the whole range from, kind of, telehealth-only, kind of, practices, like Talkspace, like MediHealth.

65 00:05:52.350 00:05:57.290 Robert Tseng: That had pretty comp… you know, maybe complicated digital

66 00:05:57.290 00:06:12.959 Robert Tseng: kind of pre-screening journeys, that, you know, everything is just kind of managed, like, online, to also working with, like, specialty health clinics with multiple sites in the Northeast, helping them kind of set up their entire technology stack from scratch.

67 00:06:12.960 00:06:29.580 Robert Tseng: Everything from connecting diagnostics that they use in clinics, to also kind of a CRM that they use at the back end, as well as other back office functions as well. So, kind of have a flavor for maybe the situation that Sunstone is in, where we’ve… we’ve spent the past few days kind of just

68 00:06:29.580 00:06:41.719 Robert Tseng: digging into some of the materials that were sent over. We’re super excited to get started with the team, and yeah, I’ll just kind of let the other folks from my side kind of introduce themselves briefly and their roles on this project.

69 00:06:43.990 00:06:49.000 Awaish Kumar: Hello, my name is Avish Kumar, and… I’m a data engineer.

70 00:06:49.240 00:06:59.979 Awaish Kumar: So I’m… I will be kind of building… setting up the data systems here, starting from setting up the data infrastructure, building the data pipelines.

71 00:07:00.080 00:07:10.089 Awaish Kumar: And finally, like, building the data maps layer, which is kind of business layer, where you have the unified, decentralized master patient data set, or

72 00:07:10.260 00:07:15.929 Awaish Kumar: Yeah, things like that. So I will be heavily involved in the implementation part of that.

73 00:07:17.490 00:07:41.629 Greg Stoutenburg: Hi everyone, I’m Greg. My background is in product management, and especially product analytics. I’ve done projects for providers of, well, various companies, including some in the healthcare industry, where we’ve had to do things like careful journey mapping of patients and customers, and identity resolution, and making sure that data that’s kept in disparate systems is able to be reconciled and united and usable.

74 00:07:41.630 00:07:42.689 Greg Stoutenburg: by stakeholders.

75 00:07:44.490 00:07:49.670 Jake Sandler: Amazing. Thank you, Brainforge team, and maybe, Jody, why don’t we start with you?

76 00:07:50.400 00:07:51.359 Jake Sandler: For the sun’s.

77 00:07:51.360 00:08:10.129 jodycrowley: Sure. Yeah, thank you. Hi, everybody. Good to meet everybody. I’m actually over at the clinic, so like Jake said, about 5 minutes or so down the road from our admin offices here in Rockville. I am new to the Sunstone team. I’ve been hired as their Vice President of Operations, which is a new

78 00:08:10.130 00:08:13.999 jodycrowley: position for the organization in preparation for.

79 00:08:14.000 00:08:35.109 jodycrowley: scaling and getting all of our systems, in play for the next phase and stage of our organization. So, appreciate the invite to this. I mostly will be listening and hearing what’s going on, because my new friends, here at Sunstone know way more about the systems than I do, as I’m…

80 00:08:35.110 00:08:42.859 jodycrowley: I don’t know, guys, what, a whole 7 days or so into this? So, but I’m a quick learner and really appreciate your support of the team.

81 00:08:45.230 00:08:46.439 Robert Tseng: Great to meet you, Jody.

82 00:08:46.610 00:08:47.470 jodycrowley: Yeah, thanks.

83 00:08:48.520 00:08:49.880 eakta : Using it.

84 00:08:50.520 00:08:52.920 Jake Sandler: Thanks, Jody. Ekta, you wanna go next?

85 00:08:52.920 00:08:59.770 eakta : Oh, I’m Ekta Agarwal, and I’m the Clinical Systems Manager. I’ve been here

86 00:09:00.190 00:09:10.210 eakta : About 3 years, right? Yeah, 3 years. Yep, I just had my anniversary. And… As we’ve had systems…

87 00:09:11.900 00:09:22.220 eakta : come in. I’ve kind of managed them. So… but mostly the Health Cloud, Salesforce piece, and the ECW piece are the…

88 00:09:22.920 00:09:24.730 eakta : ones I’m the most familiar with.

89 00:09:26.360 00:09:34.240 Jake Sandler: Very humble. Ekta is the glue that holds Sunstone together in so many ways.

90 00:09:34.570 00:09:36.209 Jake Sandler: How about Kim?

91 00:09:37.970 00:09:41.219 Kim Maillie: I apologize, my camera isn’t working for some reason.

92 00:09:41.450 00:09:42.240 eakta : Yours.

93 00:09:42.660 00:09:43.060 Jake Sandler: That word.

94 00:09:43.060 00:09:44.380 Kim Maillie: Someone else picked it.

95 00:09:44.540 00:09:52.880 Kim Maillie: Research data manager. I’ve been here for, I guess, a year and a half? So, I work with all of the sponsor EDC systems.

96 00:09:53.020 00:09:56.209 Kim Maillie: And I also work in ECW,

97 00:09:56.470 00:10:04.869 Kim Maillie: for, mostly I get information out of ECW to put into the research systems, and then I also work with,

98 00:10:05.200 00:10:06.879 eakta : research annoys in…

99 00:10:06.900 00:10:14.229 Kim Maillie: DDOTs in real time, and I also work with, Talk therapy, insurance billing.

100 00:10:16.160 00:10:25.089 Robert Tseng: Great, good to meet you, Kim. I think that’ll be super important for us at the end, when we’re gonna need to know what this data’s gonna have to look like, going back to the sponsors.

101 00:10:29.210 00:10:30.490 Jake Sandler: Amazing. Celia?

102 00:10:31.200 00:10:50.780 celialeeks: All right, hi, everybody. I am Celia. Can you please mute? Sorry, the echo is killing me. And I am the practice administrator here at Sunstone. I, have been with Sunstone since inception, so I have,

103 00:10:51.120 00:11:00.959 celialeeks: I had the opportunity to sort of see. I think I know where a lot of the old data is kept, so I’ll be happy to give you

104 00:11:01.230 00:11:07.680 celialeeks: what I, you know, access to what I remember. But, yeah, I was employee number two, so…

105 00:11:09.430 00:11:12.219 Jake Sandler: Absolutely, how long have you been with Kim and Manish, beyond Sunstone?

106 00:11:12.220 00:11:16.330 celialeeks: So, Manish and I have worked together on and off for…

107 00:11:17.480 00:11:19.419 celialeeks: It’s gonna date me. 22 years.

108 00:11:19.420 00:11:20.640 Jake Sandler: Sorry.

109 00:11:21.280 00:11:33.610 celialeeks: I worked… I… my background is in oncology, so I worked with him when he was practicing, in his private practice, and then I’ve worked with Kim since 2015, so… and…

110 00:11:33.610 00:11:44.780 celialeeks: Kim Malley, actually, and I… Kim Malley is also background in oncology, so she and I… I’ve known her just as long. So, 22 years.

111 00:11:44.880 00:11:45.590 celialeeks: Amazing.

112 00:11:46.720 00:11:47.370 Jake Sandler: Yeah.

113 00:11:47.370 00:11:47.850 Robert Tseng: Good to meet you.

114 00:11:48.450 00:11:54.850 celialeeks: Yes, but Sunstone… since inception, so I can get you the info from here. Fingers crossed.

115 00:11:54.850 00:11:56.879 Robert Tseng: Yeah, you’re our librarian of the group.

116 00:11:59.550 00:12:01.149 celialeeks: And then Lindsay.

117 00:12:01.610 00:12:18.230 Lindsay Ozyck: Hi, I’m Lindsay. I’m the People Ops Manager here at Sunstone, but before that, I was Chief of Staff, so I had a lot of sightline into how we use various systems. You know, if Kim Malley is sort of, like.

118 00:12:18.230 00:12:32.289 Lindsay Ozyck: our research data systems and act as our clinical systems, health cloud, and ECW. I’m sysadmin on most of the other systems,

119 00:12:32.530 00:12:52.310 Lindsay Ozyck: we have a IT provider, DP Solutions, that is also sysadmin on those systems, but they don’t, have as much of the historical context of, like, how we use them and why they were set up the way that they were set up. So, I’ve been here with the company for about 3 years.

120 00:12:52.310 00:12:56.559 Lindsay Ozyck: And, yeah, if there are questions about,

121 00:12:57.140 00:13:04.940 Lindsay Ozyck: sort of how we’ve set up various systems and how we currently use them. I have a pretty decent sense of that.

122 00:13:07.230 00:13:09.450 Robert Tseng: Great. Good to meet you, Lindsay.

123 00:13:11.290 00:13:34.559 Jake Sandler: Amazing. Well, thanks, everybody. Very excited to get this kicked off, and honestly, Robert, I’ll turn it over to you. I’ve just asked Robert to sort of walk through the overall project, so you guys have the 30,000-foot view, and then I think we’re gonna go into more tactics, and really, I want the Sunstone team, if you have questions, if something doesn’t make sense, if you’re unsure what we mean, or what the team means by certain things, this is…

124 00:13:34.560 00:13:41.140 Jake Sandler: meant to be interactive, not us just… and this one’s not the brain for you, just talking at you, so please raise your hand.

125 00:13:41.140 00:13:50.039 Jake Sandler: jump in, let’s make this collaborative, you guys are going to be key champions inside the organization as we continue this work, and so want to make sure that

126 00:13:50.060 00:13:57.160 Jake Sandler: it makes sense to you all, and you will have much more institutional knowledge than I do, and then Brainforge does, so,

127 00:13:57.260 00:14:03.639 Jake Sandler: yeah, just please feel invited to jump in at any point. And with that, Robert, I’ll kick it over to you.

128 00:14:04.130 00:14:21.329 Robert Tseng: Okay, thanks, Jake. And yeah, recognize that it’s very rare to have all of us in a room together, so I really appreciate taking the time. Like Jake said, our goal is to try to collaborate in a way that’s not disruptive to your day-to-day, and so we do that by trying to over-communicate as much as we can.

129 00:14:21.330 00:14:46.320 Robert Tseng: In a, you know, obviously in a… in a way that’s actually helpful, you know, with… by sending memos, by kind of sharing the documentation we’re doing, so just kind of prefencing that may be overwhelmed by all these things that we’ve put together that you’ve not seen before, but just want you to just kind of feel grounded in, like, what you’re familiar with, and just to speak to what you do know. I think that’s what’s… then we can kind of sort… sort through, kind of all the details of where that information needs to flow to.

130 00:14:46.320 00:14:46.880 Robert Tseng: True.

131 00:14:46.880 00:14:57.720 Robert Tseng: So, I’ll just kind of share my screen for now. I’m assuming that most people have… are looking in front of a screen. So I’ll just kind of reference a couple things here.

132 00:14:57.720 00:15:12.709 Robert Tseng: On the left side of my screen, I… this is kind of the memo that I shared out earlier. Totally fine if you didn’t do… didn’t get a chance to do the pre-read, so I just wanted to kind of talk at a high level, kind of what we are planning to do over these next 6 to 8 weeks.

133 00:15:12.710 00:15:18.990 Robert Tseng: And so, yeah, we’ve kind of defined it as, like, four phases, which I’ll kind of get to more in this, like.

134 00:15:18.990 00:15:30.320 Robert Tseng: in the spreadsheet, but really there’s kind of two tracks going on. One is kind of a discovery track, where we are really trying to understand all the definitions that we need to

135 00:15:30.320 00:15:53.330 Robert Tseng: That we need to know, specifically around, kind of, patient identity rules. You know, we’ve… we’ve spent some time looking at old notes and video run-throughs of, specifically, ECTA, kind of talking through some of the nuances, especially ECW and Salesforce Health Cloud, kind of where the ID… where patient IDs, kind of originate, and when additional pieces of information kind of get stitched in, so…

136 00:15:53.330 00:16:16.169 Robert Tseng: we kind of have a baseline understanding of that, but I’m sure there’s… there’s more edge cases. And, you know, just from my personal experience dealing with healthcare data sets, the volume of, kind of, the actual patients may not be that high all the time, but it’s… it’s very intensive, trying to figuring out all the different, kind of, like, flow charts and the way that, kind of,

137 00:16:16.220 00:16:27.820 Robert Tseng: Data gets captured when it branches out, because you’re… especially, you know, you have… patients may have many, many touchpoints, and we’re pulling in data from all types of sources, both

138 00:16:27.820 00:16:52.319 Robert Tseng: you know, unstructured and unstructured. And so, yeah, kind of building out this semantic layer, or kind of a standardized way of being able to describe kind of where data is coming from and what it actually means, I think we’ll probably need your help in kind of level setting on that. And then, yeah, just really just making sure that the assumptions that we’re making to fill in all the gaps are, kind of make sense with the way that

139 00:16:52.320 00:16:56.279 Robert Tseng: Some of the… some of the folks on this call actually think about the business.

140 00:16:57.140 00:17:20.570 Robert Tseng: And so, on the other side, there’s kind of the track where we’re actually going to be doing some of the building in this phase, too, because this is not just a pure documentation kind of engagement. We want to roll up our sleeves and actually, you know, land the data in an environment where we can actually build kind of a comprehensive model that may ultimately be a better solution for moving data out of

141 00:17:20.569 00:17:44.870 Robert Tseng: like a… like a real-time or an ECW to, to the… to the, to a spon… to the sponsor and the, and the output that they want. So, you know, our… our goal, you know, maybe 8, 8 weeks out from now is… is really to be able to, to, to demonstrate that capability and hopefully see… see some, and get some, get, get your sign-off on… on that, that that’s actually helpful.

142 00:17:45.730 00:17:48.800 Robert Tseng: I guess, you know.

143 00:17:48.960 00:18:02.890 Robert Tseng: I guess I’ll just pause there. I realize that I can’t actually see anybody’s face anymore while I’m doing a full screen share, so please just jump in and just cut me off, otherwise I won’t be able to tell. But if no questions, I’ll just kind of keep going through this.

144 00:18:04.430 00:18:12.779 Jake Sandler: Robert, I’ll just jump in for a second. Yeah. So, Sunstone team, does that sort of make sense, how Robert’s laying it out? Is there any, you know…

145 00:18:13.020 00:18:24.410 Jake Sandler: Anything he said that you have questions about, or just want to jump in and get clarity, and just want to make sure, like, it’s clear to you guys what the teams… the Brandforge team is going to be undertaking.

146 00:18:26.550 00:18:31.289 Jake Sandler: Lindsay, I see you nodding. Ecto, Celia, Cam Valley, makes sense to you guys.

147 00:18:31.940 00:18:33.339 Jake Sandler: Cody, cool.

148 00:18:34.040 00:18:34.979 Jake Sandler: Back to you.

149 00:18:35.260 00:18:48.749 Robert Tseng: Okay, great, thanks, appreciate it. Yeah, so I’m not going to kind of go into the rest of this doc. I still think it’s a good, good read if you have a chance to kind of look through it on kind of what, what, you know, just where our starting point, and so I’ll just kind of lightly touch on

150 00:18:48.750 00:19:00.719 Robert Tseng: This… the spreadsheet we put together, the idea is to really kind of pack things up nicely, you know, 6 to 8 weeks from now. You’re gonna… your team’s gonna have this, it’s gonna be that one place, one central place where we’ve… where we’ve built out everything.

151 00:19:00.820 00:19:10.990 Robert Tseng: And so, yeah, I think from, like, a, you know, what you can expect from a roadmapping perspective, this is really just kind of high level. On our side, we kind of run

152 00:19:10.990 00:19:17.320 Robert Tseng: We run engineering sprints, we do… we do weekly sprints, and so we’ve kind of broken this out into an 8-week kind of,

153 00:19:17.320 00:19:39.980 Robert Tseng: engagement, and there’s going… going through different phases. So, really, for this first week of what we’re trying to get to, you know, there’s a few things. One is, like, yeah, kind of kicking off all these docs, making sure that the team understands, kind of, both the how and the why for why we’re… for when we’re starting to poke around, getting access to the systems that we need, that we need to get to.

154 00:19:40.160 00:19:44.419 Robert Tseng: I see a comment that jumped in. I’m zooming in on the spreadsheet, totally.

155 00:19:44.770 00:19:46.049 Robert Tseng: Yeah.

156 00:19:46.370 00:19:49.689 Robert Tseng: I’ll just kind of drag this out.

157 00:19:50.380 00:20:03.410 Robert Tseng: Yeah, so I’m just kind of hovering here. And then, yeah, understanding that there is this kind of real-time system right here, so I want to make sure that we’re… yeah, I kind of wrote, like, risk mitigation here, understanding kind of what

158 00:20:03.680 00:20:09.790 Robert Tseng: you know, where we need to, not, like, I guess, like.

159 00:20:10.030 00:20:33.120 Robert Tseng: understanding that that’s already in progress, there’s going to be some gap as systems are being migrated from one to the other. Like, when do… when do we touch real time, when do we not? Like, kind of understanding that sequencing. That’s not something we have to kind of figure out on this call, but probably something that I’d like to talk, at least with Jake and Ekta, about, whether it’s async or on a different call, just to make sure that, you know, when we’re kind of

160 00:20:33.220 00:20:44.340 Robert Tseng: making our rounds in phase one that we’re not gonna kind of get blocked… blocked on that piece. But yeah, that’s kind of how we’re thinking about it. Yeah, internally, we’ll break all… we break all this into tickets. I know

161 00:20:44.340 00:20:54.289 Robert Tseng: Kim… Kim Rohde wanted to kind of see, you know, the level of granularity on, like, what we’re working at at the ticket level as well, but I’m not gonna kind of share that with this audience.

162 00:20:54.290 00:20:59.100 Robert Tseng: And then on the second… on the second tab here, we have, kind of…

163 00:20:59.100 00:21:18.530 Robert Tseng: this… what I would consider this this master question list that we have. I think all the questions that we’ve laid out that we think we need to answer, so that we can be set up to do, what we’ve always set out to do. A lot of these kind of questions, we already kind of have some context, I’ve already pre-filled it in, and you know, I might pick a few questions from here to just ask to the group.

164 00:21:18.530 00:21:23.510 Robert Tseng: Towards the end of the call, if we wanna… if we have time to just kind of get the conversation going.

165 00:21:23.510 00:21:33.100 Robert Tseng: But otherwise, you know, I think this is something that I could probably share async, start tagging owners now that I kind of understand people’s roles, and it would be great just to kind of

166 00:21:33.100 00:21:49.330 Robert Tseng: even if you just drop in a doc or, you know, point us in some direction, I think that would help us to kind of get some more clarity on the guardrails of the project and kind of understanding where critical pieces of data are.

167 00:21:49.850 00:22:00.039 Robert Tseng: Yeah, and then I, you know, the rest of the stuff is kind of more kind of detailed, kind of just what we’re gonna actually be doing, kind of what does, what does success look like.

168 00:22:00.040 00:22:10.339 Robert Tseng: We have a resource library of all the different things that we’re going to be referencing, externally, but also a kind of internal documents log that we will also be maintaining. We have a master Google Drive folder with

169 00:22:10.340 00:22:33.269 Robert Tseng: With, with Jake, so I think that’ll probably be distributed to the team, so you’ll all have it in one place. Just want to give your team the reassurance that we kind of have, yeah, we have this project pretty, pretty tightly wrapped, and, yeah, we want to just make sure that we’re building, we’re co-building with you in a transparent way, and, yeah, make it easy for people to kind of jump in and contribute and not feel like.

170 00:22:33.270 00:22:57.829 Robert Tseng: There’s just, like, this, you know, typical data black box that, like, nobody really understands what’s going on. We want to kind of, like, break the… that paradigm of being kind of, like, the IT partner in this project that is very much integrated and, with, with your team and, yeah, can communicate what we’re doing. So, yeah, that’s kind of the spiel that I wanted to give at a high level for

171 00:22:57.830 00:23:12.090 Robert Tseng: kind of what we put together, just to kick off the project. I’ll kind of stop sharing screen so I can go back to kind of seeing everybody, and then, pause for questions, and if not, I can kind of keep directing it with more questions from my side.

172 00:23:13.810 00:23:16.329 Lindsay Ozyck: I have a quick question. Yeah.

173 00:23:17.030 00:23:25.560 Lindsay Ozyck: Just waiting… oh, perfect. So, wondering, what…

174 00:23:26.280 00:23:38.460 Lindsay Ozyck: like, inputs, if scoping has already been done around, like, what’s needed from a research regulatory perspective?

175 00:23:38.550 00:23:53.690 Lindsay Ozyck: around the handling of patient data, because there’s obviously, like, HIPAA and all of that stuff that we’re bound by, but then on top of that, there’s also typically criteria that, the… the sponsors

176 00:23:54.230 00:24:07.319 Lindsay Ozyck: are looking at when they audit us. So just wondering if you guys have all the inputs that you need for that, or if that’s something that we would need to, flesh out and hand over.

177 00:24:07.770 00:24:27.189 Robert Tseng: Yeah, I guess, Jake, I don’t know if this is related to what you shared with me earlier today, just different kind of, trial… trial data sets. I didn’t actually kind of look too deeply into what the differences were, but since they’re coming from different… different sponsors, like, is… is that more kind of what, what Lindsey’s referring to? Yeah.

178 00:24:27.190 00:24:33.869 Jake Sandler: Yeah, Lindsay, are you… are you talking about, like, hey, sponsors have certain requirements on top of HIPAA that we need to follow, and that… that might…

179 00:24:34.960 00:24:39.479 Jake Sandler: inhibit or impede data access, or am I misinterpreting?

180 00:24:39.870 00:24:44.360 Lindsay Ozyck: I think, like, where the… my question is,

181 00:24:45.240 00:24:53.849 Lindsay Ozyck: We’ve had questions from sponsors around how our instances are set up,

182 00:24:53.930 00:25:05.600 Lindsay Ozyck: they are looking at, like, CFR compliance. Like, I know CFR Part 11 is one of them. There are other, regulations that relate to

183 00:25:05.630 00:25:14.920 Lindsay Ozyck: data that we have to follow because we’re doing… Gotcha. So yeah, just those kinds of things where,

184 00:25:15.110 00:25:28.209 Lindsay Ozyck: Just making sure that you guys have all the inputs that you need, so that as you recommend changes and tweaks to systems, they’re still lining up with the criteria that our sponsors are going to be evaluating our systems against.

185 00:25:28.850 00:25:29.210 Jake Sandler: Yeah, definitely.

186 00:25:29.210 00:25:45.730 Robert Tseng: We definitely would want those requirements, and I think just to clarify on our end, like, we won’t be pushing changes to your systems, at least in the sprint. Like, we’re going to be very, kind of, like, trek very cautiously, even if we’re… yeah, we’ve been sent over some data sets, anything that we do is

187 00:25:45.730 00:25:49.499 Robert Tseng: It’s gonna be completely isolated from the systems that you’re doing.

188 00:25:49.500 00:26:13.520 Robert Tseng: We’ve talked about internally already about, like, landing the data in, like, a data warehouse, like, that’s completely privatized. There’s a couple different solutions that we’ve been looking at, just so that we can at least have a working environment, like a sandbox, for us to, like, experiment on the modeling side. But we would strip out anything that’s PII there and only replace it with hypothetical patient data, so…

189 00:26:13.520 00:26:24.970 Robert Tseng: that you can be assured that, like, we follow, you know, HIPAA procedure. We’re also… we are inside of BAA with your team, so we’re happy to kind of adhere to whatever the sponsors need.

190 00:26:25.380 00:26:25.890 Lindsay Ozyck: Okay.

191 00:26:26.820 00:26:33.160 Jake Sandler: But yeah, Lindsey, I think when we get, like, at the end of the project, when we get this master schema, we’ll definitely want to make sure it’s…

192 00:26:33.160 00:26:53.739 Jake Sandler: if there’s CFR or similar kind of requirements that would affect, like, how the data is structured, which is… I have very limited understanding of the trial regulatory stuff, but I understand, like, there is, like, certain random requirements that we need to make sure we’re following, so getting those specifics would be great.

193 00:26:53.740 00:26:56.740 Jake Sandler: Through, you know, by the end of the project, for sure.

194 00:26:57.030 00:26:58.300 Lindsay Ozyck: Okay, awesome.

195 00:27:01.910 00:27:04.319 Robert Tseng: Great. Any other questions for now?

196 00:27:11.110 00:27:23.050 Robert Tseng: Okay, well then, I’ll kind of maybe start from… kind of start to drill into things a little bit more. So I think, Jake, you’ve kind of got us sorted on, like, where… who…

197 00:27:23.050 00:27:33.409 Robert Tseng: you know, we need to get some sense on emails, and, like, kind of roles, access, permissions, I’m assuming we can just kind of handle off this call. But I guess, are there any kind of other source systems that

198 00:27:33.420 00:27:42.279 Robert Tseng: we discussed last time, which… that we’re… that, maybe we’re still missing something, just to kind of recall, you know, it’s ECW, Health Cloud.

199 00:27:42.280 00:28:02.460 Robert Tseng: I don’t think we said we would do access to real time right away, but I mean, if that’s changed, we can do that. And then I think you have a lot of files in Google Workspace, and so really kind of knowing how we wanted to be involved there, if it’s just going to be having access to that through our… the Sunstone emails that you’ll be giving us.

200 00:28:02.460 00:28:06.850 Robert Tseng: Just wanted to kind of get that housekeeping out of the way for now.

201 00:28:07.220 00:28:16.569 Jake Sandler: Yes, and I think Lindsey will be helping on some of those, but I think it’ll be shared across this team. But yeah, I think, Lindsay, we’re getting those emails set up.

202 00:28:16.570 00:28:35.059 Jake Sandler: I think Jess was taking the point on responding to the job form. I don’t know if that’s come through yet, for the BrainForge team. But then… so, like, I think the process is, Lindsay, correct me if I’m wrong, we, like, fill out this form that, like, sort of identifies, here’s the emails we need and the access, you know, that these emails need, and then there’s an onboarding process.

203 00:28:35.060 00:28:36.009 eakta : Oh, mute.

204 00:28:37.950 00:28:38.840 Lindsay Ozyck: Sorry.

205 00:28:40.730 00:28:41.950 Lindsay Ozyck: After you don’t know.

206 00:28:43.150 00:28:45.719 Lindsay Ozyck: Oh, it’s gonna be good now.

207 00:28:46.290 00:28:51.460 Lindsay Ozyck: I was just wondering if, so we have the one…

208 00:28:51.570 00:29:06.610 Lindsay Ozyck: the tech at Sensor Therapies that we created last time, do we… does everyone from the Brainforge team need access to all three main systems, the Google Workspace, Health Cloud, and ECW?

209 00:29:08.320 00:29:11.330 Robert Tseng: I… I think we can… Go ahead.

210 00:29:11.770 00:29:16.490 Jake Sandler: Yeah, I think it would be… I mean, ideal, like, I think in an ideal world, if there’s not… if it’s not a huge…

211 00:29:17.100 00:29:30.059 Jake Sandler: burden would be what we said, talk about 3 emails, Robert, with access to the core systems, so, like, a tech 1, 2, and 3, or we can use the tech, plus, like, a tech 2 and 3, however you want to do it. Yeah. However, if that becomes…

212 00:29:30.290 00:29:40.979 Jake Sandler: there’s, like, a big cost, or that’s, you know, a huge pain for some reason, I think the Brainforce team is flexible and is open to having one that they can, like, use, share securely amongst them.

213 00:29:41.400 00:29:42.320 Jake Sandler: Is that right?

214 00:29:42.320 00:29:53.279 Robert Tseng: I would say, at minimum, two would be ideal, just because, like, I feel like a lot of these systems, if you log in at the same time, like, you just get kicked out, and so sometimes it’s just…

215 00:29:53.280 00:30:03.459 Robert Tseng: Easier to have at least one. One can basically be the service account, if we need to run anything kind of programmatically, like, that account gets used, and then the other one the human can actually be using at the same time.

216 00:30:04.430 00:30:05.160 Robert Tseng: Yeah.

217 00:30:05.510 00:30:07.769 Lindsay Ozyck: Okay, we don’t have any restraints.

218 00:30:08.300 00:30:12.799 Lindsay Ozyck: an ECW, it’s just Health Cloud licensing that I need to check,

219 00:30:12.980 00:30:13.590 Robert Tseng: Okay.

220 00:30:14.010 00:30:16.230 Lindsay Ozyck: But I should be able to do two, should be fine.

221 00:30:16.600 00:30:18.320 Robert Tseng: Okay Great.

222 00:30:18.570 00:30:22.970 Robert Tseng: And yeah, we can reuse the one that you already set up, so I’m not… yeah, we don’t need, like, two net new ones.

223 00:30:22.970 00:30:33.389 Lindsay Ozyck: But, I was just… I don’t think there’s any data in real time right now. Okay. Any patient data?

224 00:30:36.050 00:30:39.250 Lindsay Ozyck: I mean, I don’t mention that into statuses.

225 00:30:40.110 00:30:42.660 Lindsay Ozyck: But the kind of thing we see that we do.

226 00:30:42.950 00:30:51.610 Lindsay Ozyck: Oh, yeah, and Health Club, you know. Well, I’m sad, so they’re not that big. Okay, that’s fine. I mean, we’ll get you real-time

227 00:30:52.780 00:30:58.399 Lindsay Ozyck: let’s start with these three. It’s gonna be more data-heavy with the Google Drive, ECW, and

228 00:31:00.190 00:31:04.209 Lindsay Ozyck: And Salesforce, and then we can… if you need real-time, that’s no problem.

229 00:31:05.360 00:31:09.019 Robert Tseng: Okay, yeah, I think that sounds good. That’s kind of what we were expecting.

230 00:31:10.160 00:31:14.050 Jake Sandler: And then, Robert, for audio-visual, It’s gonna be less important.

231 00:31:14.610 00:31:22.439 Jake Sandler: but, like, Vault will have those files, and I think Dropbox has some old ones, and so in terms of just being able to stitch, like.

232 00:31:22.870 00:31:39.130 Jake Sandler: files from Volt to the proper patient ID with the proper, like, visit date. That’s been a challenge for the team. There’s an exercise happening internally where they’re trying to, like, basically match which video files go to which patient for which visit. And so just understanding

233 00:31:39.780 00:31:43.820 Jake Sandler: the pain points there, I think, will be helpful as you create the schema.

234 00:31:44.820 00:31:45.430 Robert Tseng: Okay.

235 00:31:45.620 00:31:51.249 Robert Tseng: Yeah, I think that’s… that’s definitely a good call-out.

236 00:31:52.460 00:31:57.049 Robert Tseng: I have… if I can kind of keep moving along Ben, so…

237 00:31:57.630 00:32:12.719 Robert Tseng: Yeah, I think one of the biggest complications that I’ve observed from some of the kind of walkthroughs I’ve already seen is that, yeah, you know, cross-system linkage is kind of the biggest complexity, and so, there… I don’t know if…

238 00:32:12.830 00:32:19.079 Robert Tseng: If there are known areas where, like, duplicate… duplication happens, where, like.

239 00:32:19.260 00:32:33.279 Robert Tseng: multiple patients are long… like, for one example, I think in the… in the Health Cloud kind of workflow, where, there could be multiple patients on the same day coming for multiple visits, or, like, the same patient

240 00:32:33.280 00:32:43.350 Robert Tseng: on the same day, coming from multiple visits, or there’s, like, the kind of… it shows up multiple times. I think those types of situations where, I… I mean, I’m sure we could kind of…

241 00:32:43.350 00:32:53.220 Robert Tseng: figure out more of that in a deeper dive, situation, but if… I think anything that’s… that’s around duplication or, I think…

242 00:32:55.910 00:32:56.590 Robert Tseng: Did it?

243 00:32:57.240 00:33:21.480 Robert Tseng: patient kind of originating in a different system, and then eventually showing up in a downstream system. I think that those linkages, like, I feel like, are probably where we’re gonna get hung up on the most. But anyway, I think that’s more kind of away from your lane, so I’ll let you kind of, like, once… once we really get access to things, to kind of pick… pick through it and ask more specifically once we get to those, get to those points.

244 00:33:21.620 00:33:33.579 Robert Tseng: So, I guess, I guess it’s not really… not really a question. That’s just a call-out of where… where I… I think we’re… we’re gonna… we’re gonna probably need… need more support from the team, as we’re moving along here.

245 00:33:33.580 00:33:43.359 Jake Sandler: Robert, would it be helpful for ECTA or others to sort of share, like, the biggest pain points of that nature now, or do you want to wait until deeper follow-up?

246 00:33:43.870 00:33:47.000 Robert Tseng: I mean, I’m… I’m happy to hear…

247 00:33:47.260 00:34:00.849 Robert Tseng: kind of some of the… if… if you’re ready to share, like, we… yeah, it would be great to kind of get some… any of that… of those, leading thoughts now, and then, we can… that can help steer us as we’re… as we’re kind of scheduling a follow-up.

248 00:34:01.770 00:34:20.280 Jake Sandler: So maybe I’ll frame it up this way, Ekta and Celia and team, like, if there’s… there’s a lot of different data struggles, right, in various ways, but if you had to sort of highlight maybe two or three of the… the things that create the most difficulty for your… for the teams to do their job, like, how, like, what are… what is the nature of those issues that occur today?

249 00:34:20.550 00:34:21.530 celialeeks: So…

250 00:34:25.550 00:34:38.609 celialeeks: All right, I would say from the top would be the various stages of data that exist in various platforms. So, Google Drive holds a lot of

251 00:34:39.350 00:34:43.250 celialeeks: The initial data and the initial,

252 00:34:43.370 00:35:00.609 celialeeks: information, and in different formats. Then there’s also the nomenclature that you’ve already brought up around the consistent naming and being able to locate information, because it’s not named the same throughout each system, and sometimes it’s not even named the same within

253 00:35:01.020 00:35:04.690 celialeeks: this… the same system.

254 00:35:05.560 00:35:15.170 celialeeks: And then also identifying… I think we’ve… we’ve come across is what exactly are people looking for? Because I think that that…

255 00:35:15.520 00:35:18.229 celialeeks: Gets miscommunicated as well.

256 00:35:18.700 00:35:30.939 celialeeks: Acto, what are your thoughts? Am I… I feel like those are the biggest struggles we have, is… and not everybody has access to every system, and every folder. I don’t know, Acto.

257 00:35:30.940 00:35:33.150 Jake Sandler: And Celia went into saying…

258 00:35:33.370 00:35:34.310 celialeeks: Yeah, like, it is…

259 00:35:34.310 00:35:35.180 Jake Sandler: Go ahead, Ekta.

260 00:35:35.680 00:35:44.319 celialeeks: Inconsistency, like, the patient identifier, we don’t have a consistent patient identifier across any system.

261 00:35:46.760 00:35:52.939 celialeeks: And I think, like, we… And we’re not still with our data definitions, probably.

262 00:35:54.980 00:35:58.569 celialeeks: But we do have a lot of duplications, because across systems.

263 00:36:00.440 00:36:10.230 celialeeks: you, like, you know, if a file originates in the Google Drive, we also put it in HealthCloud so you have an easy link, and then it’s also uploaded in ECW, so you’ll have that same piece.

264 00:36:10.530 00:36:11.580 celialeeks: And 3.

265 00:36:11.780 00:36:12.920 celialeeks: three systems.

266 00:36:12.920 00:36:13.580 Robert Tseng: Yeah.

267 00:36:13.580 00:36:20.790 celialeeks: When you’re looking at, you know, sponsor-based systems, it’s then duplicated into that. So.

268 00:36:21.440 00:36:24.320 Awaish Kumar: So, like, these clinical systems, are they, like.

269 00:36:24.590 00:36:32.149 Awaish Kumar: Independent systems where the full journey of a patient goes, or the patient moves.

270 00:36:32.150 00:36:32.470 celialeeks: They’re…

271 00:36:32.470 00:36:34.670 Awaish Kumar: Through different… for the different stages.

272 00:36:35.000 00:36:38.569 celialeeks: So, no, well, so patients generally originate in Health Cloud.

273 00:36:38.770 00:36:42.729 celialeeks: And then they’re synced with… Yes, the NCW.

274 00:36:42.950 00:36:53.610 celialeeks: that… but that’s not historic data. So, like, historically, I think the other challenge will be from the inception time, where and how things were held and documented.

275 00:36:53.610 00:37:06.210 celialeeks: like, we don’t even have a clear cutoff of when every… or do you have, like, when everything was actually updated to Help Cloud? And, you know, there’s… So, yes, so what we went live with Help… so, if Sunstone started

276 00:37:07.690 00:37:14.710 celialeeks: In the summer of 2022, Salesforce went live in March of 23.

277 00:37:14.840 00:37:18.860 celialeeks: ECW went live in April of 24.

278 00:37:19.290 00:37:24.559 celialeeks: And then, and we started using real-time

279 00:37:24.750 00:37:30.230 celialeeks: In 2025, sometime. Yeah, 2025, sorry. At the end of 2025. So,

280 00:37:30.670 00:37:40.199 celialeeks: prior to ECW, actually, we used AdvancedMD, but I don’t know if we still have access to that either. So that data is really lost, but

281 00:37:40.800 00:37:47.760 celialeeks: So, a lot of, so those are kind of our cutoff points. So before Health Cloud

282 00:37:48.050 00:37:54.810 celialeeks: we have a spreadsheet of where everyone joined, and then we, we data-loaded that into HealthCloud.

283 00:37:55.530 00:38:00.240 celialeeks: But… That is not super accurate.

284 00:38:00.540 00:38:01.750 celialeeks: Probably.

285 00:38:02.180 00:38:05.020 celialeeks: Jake.

286 00:38:05.020 00:38:22.489 Jake Sandler: So, this is the first time I’m hearing that… so the AdvancedMD, there’s a bunch of data, entered into AdvancedMD. Was there… before that system went away, was there, like, a full export, so you have it at least in, like, a raw file, or is there actually just, like, some patient data that was in AdvancedMD that we think, like, is just totally lost to us?

287 00:38:22.490 00:38:38.179 celialeeks: It is important to note that I think AdvancedMD only held demographic data. I don’t think there were ever notes. There were no notes generated. Oh, no, we didn’t have any notes. That’s why we got rid of Advanced MD. Okay, I was like, we did not generate source data, but it did have… I thought it had some…

288 00:38:38.820 00:38:50.470 celialeeks: because the monitors were going in, I thought it had EKGs and stuff that Jen was doing in there. Their lab… medical records went in there. But I think we still… we may have those on Google Drive. We may have those duplicated, because, like, that’s the

289 00:38:51.040 00:38:57.080 celialeeks: And so… and we did have an intern that went and recreated the charts in ECW.

290 00:38:57.500 00:39:05.529 celialeeks: And so… Potentially, we have most of that data. It did not hold much, and no one really.

291 00:39:05.530 00:39:06.160 Jake Sandler: Yeah.

292 00:39:06.160 00:39:09.029 celialeeks: And I don’t think it was ever the source of truth.

293 00:39:09.030 00:39:09.370 Jake Sandler: Yep.

294 00:39:09.370 00:39:12.309 celialeeks: I mean, it was… I think it was…

295 00:39:14.090 00:39:23.299 celialeeks: It was intended to be the EMR, and it was intended to be the source of truth, but it was not actually used. I think that two people could use it.

296 00:39:24.570 00:39:25.650 Jake Sandler: Cool.

297 00:39:27.030 00:39:33.409 Awaish Kumar: Okay, so the data for patients, which is in Google Drive, it’s just for historical ones.

298 00:39:34.750 00:39:38.629 celialeeks: Hmm… well, it has all the vault data, right?

299 00:39:38.630 00:39:46.140 Robert Tseng: Yeah, there’s a lot of, like, non-structured data awaits that’s in there, so, like, multimedia is in there,

300 00:39:46.250 00:39:59.150 Robert Tseng: like, transcripts, videos, etc. are in there. So, but all, like, the structured form-based data, like, it seems like it’s been a shifting source of truth over the timeline of Sunstone.

301 00:39:59.150 00:40:22.909 Robert Tseng: I guess we’ll just have… I don’t think the goal of this engagement, at least, I mean, it’s maybe it’s good that we’re talking about it, is to, like, accurately backfill everything. I think we are… I mean, we want to… I think we had previously said, within 12 months window, we try to get as much as we can, and we use… just use that as a starting point, and then maybe, you know, if we… I’m sure we’ll find other skeletons in the closet and learn a lot about how to

302 00:40:22.910 00:40:26.230 Robert Tseng: How to backfill, and maybe that’s, like, something we could take on, like, later on.

303 00:40:27.380 00:40:39.899 celialeeks: But I think those are all of the places where data is held that we’ll need to do some digging, and everybody will sort of have to say, oh, well, I have this in this file here. Right.

304 00:40:40.210 00:40:42.570 celialeeks: If we need to match it…

305 00:40:42.820 00:40:57.520 celialeeks: I know one of the things that you guys mentioned, for, like, retroactive, like, correction of data and, like, organization of, like, data that already exists is the… some of the video files within Dropbox.

306 00:40:58.550 00:41:09.849 celialeeks: Some of that, potentially, is… could be cross-referenced with the export that I did of the Master Clinic schedule.

307 00:41:09.970 00:41:24.710 celialeeks: So I have a sheet that I can share with people. Oh, right, so we have all the visits that data a long time ago. We do have all the visits from 20… that were… we used to use… before we used to help prep the scheduler, we were using the Google Drive. Or Google Calendar, so… Google Calendar, yeah.

308 00:41:25.730 00:41:35.600 celialeeks: And then I exported that into a spreadsheet, and then cleaned it up so that the naming convention’s, like, a bit more, consistent, and you can find things.

309 00:41:38.700 00:41:39.570 Robert Tseng: Great.

310 00:41:40.460 00:41:48.199 Jake Sandler: That sounds like a great sheet to share, Lindsay, that would be awesome. So we’ll… I’ll give you a follow-up, and we’ll create, like, a shared Google Drive, and start to…

311 00:41:48.470 00:41:49.610 celialeeks: So if you…

312 00:41:50.310 00:41:57.289 celialeeks: The visit data combined with the patient data is going to be the most accurate to know that that’s the same person.

313 00:41:57.930 00:41:58.490 Robert Tseng: Yeah.

314 00:41:58.760 00:42:02.469 celialeeks: Right, to, like… Because in Health Cloud.

315 00:42:02.920 00:42:06.899 celialeeks: You can create multiple entities, different email addresses.

316 00:42:07.400 00:42:14.630 celialeeks: And so, you kind of have to cross-reference their birthday Email and phone number. And…

317 00:42:15.420 00:42:21.080 celialeeks: hopefully they haven’t moved. So that’s kind of how I run the duplication and help out.

318 00:42:21.710 00:42:23.210 Robert Tseng: Yeah, makes sense.

319 00:42:28.220 00:42:40.829 Robert Tseng: Okay, well, yeah, then I think, like, this is a great detail to kind of know, kind of, our starting point and, like, how we do the digging. I think maybe, like, Greg, just to maybe let’s kind of

320 00:42:41.370 00:42:54.419 Robert Tseng: point-pointed that deal a little bit. So, I mean, we have all the kind of pre-screening kind of patient flows from Jake. Obviously, like, there’s gonna be some of this kind of system-to-system kind of,

321 00:42:55.620 00:42:58.570 Robert Tseng: Movement that we have to, have to, have to, like.

322 00:42:58.790 00:43:16.370 Robert Tseng: document as well. I didn’t show this in the spreadsheet, but one of the things that we’re trying to put together is just, like, a comprehensive tracking plan of, like, all the different milestones for the patient. Maybe from your perspective, it’s not going to be so linear, and there’s actually quite, like, many paths to get to, you know, many paths to get to Rome.

323 00:43:16.370 00:43:33.750 Robert Tseng: But, like, I think, you know, I think that’s one way, you know, if we can get that pretty clean, that makes the data modeling very, very easy. But I feel like that’s probably, like, you know, one direction that Greg will probably want to take things in. But I’ll let you kind of jump in if you have anything else to say about that.

324 00:43:33.750 00:43:53.609 Greg Stoutenburg: Yeah, I think I’ll just jump in and say here, you know, as… so clearly there are lots of different data sources that have kind of come and gone. Some pieces of information for patients were saved over here, some were over there. Something that’s going to help really provide guidance for us as we try to put it all together is understanding the patient journey overall.

325 00:43:53.610 00:44:07.450 Greg Stoutenburg: And so, viewed this way, if we know what a pre-screen flow looks like, or we know what, like, regular engagement for a patient looks like, we can then use that as guidance to then go looking for the data that’s gonna help us, sort of.

326 00:44:07.450 00:44:16.839 Greg Stoutenburg: reconcile and connect what the, the sort of patient’s life cycle is with Sunstone overall. So, that,

327 00:44:16.840 00:44:30.990 Greg Stoutenburg: it’ll be very helpful for me then, you know, I’ll be doing this deep dive to understand what those workflows for the patient look like, and, you know, leaning on the team heavily to understand that, and with that information, that’ll just be an essential component of how we put together the patient identities overall.

328 00:44:31.380 00:44:35.659 celialeeks: Alright, so this is gonna be where I think we need to…

329 00:44:35.960 00:44:47.249 celialeeks: take a look at how we share that, because, each patient’s life cycle and journey will be dependent on the study that they’re on. So there’s going to be a lot of variables there. And…

330 00:44:47.250 00:45:02.120 celialeeks: there are some patients that are on multiple studies, you know, so I just… I want to be clear on exactly what it is that you’re looking for, because a patient that comes in to be pre-screened and, you know, they show interest where you’re starting from that.

331 00:45:02.130 00:45:13.620 celialeeks: You know, workflow, and then sort of following it through until we discharge a patient, but then perhaps they come back for some other

332 00:45:13.660 00:45:21.559 celialeeks: trial, or they come back to be part of our integration groups, and so there might be some data there.

333 00:45:22.960 00:45:24.620 celialeeks: I… I just wanna…

334 00:45:24.770 00:45:38.369 celialeeks: ask if there’s a good idea… if anybody has a good idea on how to track that out for them, other than, you know, sharing a schedule of assessments. Like, I just… I don’t know, maybe I don’t see an easier way.

335 00:45:38.800 00:45:44.349 celialeeks: Or do we look at a schedule and sort of rate it that way? I think there’s…

336 00:45:44.770 00:45:48.039 celialeeks: You know, there’s some variability in terms of, like.

337 00:45:49.290 00:45:57.110 celialeeks: what the EDC is that it goes into, what the, like, the sponsors have their different systems.

338 00:45:57.310 00:46:10.230 celialeeks: That, I’m assuming you guys are not touching automations between, like, our systems and sponsors’ systems? Or is that something that was within scope that you guys were gonna take a look at?

339 00:46:11.080 00:46:23.260 Robert Tseng: I don’t think we were going to build the automation, but we do want to know what you’re delivering to the sponsors. At least kind of, like, what I have structured for, like, what the final… one of the final deliverables would be, would be, like.

340 00:46:23.360 00:46:26.310 Robert Tseng: Kind of a per-sponsor, like…

341 00:46:26.390 00:46:41.650 Robert Tseng: package, so maybe sponsor A has, like, multiple trials, each trial kind of has something slightly different, but it’s like, at least we could still give, like, a package to each sponsor. I don’t know, maybe there are too many threads, and it’s not as simple as, like.

342 00:46:41.650 00:46:52.829 Robert Tseng: like, three trials per sponsor, because that would be pretty simple to do. But, like, yeah, I guess that was, you know, how I envisioned it at first.

343 00:46:53.190 00:47:11.089 Robert Tseng: Yeah, I guess that… I would like to kind of figure out if that’s actually the right assumption, or, you know, how many variations are we actually pushing to sponsors, and are there, like, similar form… is it a similar form across different sponsors, or is it really each one so different? Yeah.

344 00:47:11.900 00:47:15.699 Jake Sandler: Minay, maybe we start, Celia, like, how many trials have been run?

345 00:47:16.040 00:47:20.140 Jake Sandler: total in the last couple years. Like, there’s been, what, 14…

346 00:47:22.850 00:47:25.920 celialeeks: Mentally do the math, hold on.

347 00:47:26.990 00:47:29.630 Jake Sandler: And It’s been… as you’re…

348 00:47:30.570 00:47:42.259 Jake Sandler: in that math, I think there’s been multi-trials per sponsor, but there’s… there’s fewer sponsors than there are trials, right? So we’re talking about, like, maybe on average 2 to 3 trials per sponsor.

349 00:47:42.630 00:47:43.270 Robert Tseng: Okay.

350 00:47:43.500 00:47:50.230 celialeeks: Yeah, I think that the sponsors are all pretty individualized. There’s not, like, a lot…

351 00:47:50.460 00:48:14.530 celialeeks: I will say, yeah, and I will even say that, you know, say we did about… say we’ve closed out about 14 trials, I’m… I’m just guesstimating here, and we’ve had multiple trials within the same sponsor, but a really good example is we have one sponsor where we had two vastly different trials with them, and so therefore, the schedule, the participant’s journey on these trials were

352 00:48:14.630 00:48:22.000 celialeeks: completely different. It was a different indication, a different model of treating, because they were treating for different

353 00:48:22.250 00:48:29.549 celialeeks: you know, indication. One was depression, one was PTSD. And so, you know, it… It wasn’t as,

354 00:48:31.360 00:48:45.069 celialeeks: comparable as you would think, because it’s the same sponsor. Like, each trial is very individual. Now, the data and the capture and the pieces that you want to push out, like, I think that’s a broader conversation that is

355 00:48:45.250 00:49:03.280 celialeeks: above what… where, like, my thinking is happening. Like, you want to push this data set back to them to say, like, this is what we collect, this is what we can show we can do, versus what the data that we’re actually sharing with the sponsor, that we’re collecting per the protocol that goes through the electronic data capture system.

356 00:49:03.280 00:49:07.419 celialeeks: the EDC, that they’re collecting, they’re monitoring, they’re…

357 00:49:07.420 00:49:16.279 celialeeks: compiling, and we’re really looking at it from a sunstone perspective of what is the work that we do, what is the data that we’re collecting.

358 00:49:16.300 00:49:19.110 celialeeks: In addition to what the sponsor is…

359 00:49:19.380 00:49:25.129 celialeeks: also collecting. Is that sort of… Like, the… Yeah. Okay.

360 00:49:25.130 00:49:41.220 Robert Tseng: That makes sense. Yeah, so I think the Sunstone, like, if we have a golden patient data set, it will be bigger than anything that we give to, like, the sponsors. And then you just take cuts of that, and you pass that to the sponsors. That makes sense. Like, I think when we’ve worked with, different providers that have

361 00:49:41.340 00:49:42.610 Robert Tseng: federal…

362 00:49:43.470 00:49:50.990 Robert Tseng: That are working with, multiple vendors, like, you know, third-party systems, like, that’s typically how we would approach it.

363 00:49:51.190 00:50:08.490 celialeeks: Okay, so… Sorry, go ahead, Lindsay. No, I was just gonna say then, I mean, you… you’re really looking at… it’s specific to the protocol that the participant is on will define their journey, for the time that they’re engaged in a protocol, but Sunstone may have

364 00:50:08.490 00:50:21.159 celialeeks: these other factors, you know, we’re offering talk therapy, we’re offering these, you know, potential follow-ups if they’re engaging back with us, wanting to do testimonials. Like, is that all part of the information you may want to

365 00:50:21.160 00:50:23.320 celialeeks: gather in the data. Yeah.

366 00:50:23.390 00:50:24.140 celialeeks: So…

367 00:50:24.140 00:50:40.260 Robert Tseng: Yeah, I think we want to capture all as much of that as we can. I would say that we would probably prioritize getting what the sponsors need, because that seems like the most urgent thing to get to, but, like, we want to build up that backlog of… I’m sure we’re not going to capture everything in the first six to eight weeks, but hopefully we get

368 00:50:40.260 00:50:46.770 Robert Tseng: Get pretty close, and then, we can keep, like, tacking on… tacking on more data from there.

369 00:50:46.770 00:50:48.370 Robert Tseng: Yeah.

370 00:50:49.560 00:51:05.769 celialeeks: Yeah, so one thing that I want to flag around that, since we’re opening it up to sponsor, like, systems, is the sponsor for a number of these systems, they control access, so…

371 00:51:06.310 00:51:22.649 celialeeks: what that requires from anyone that is given access to those systems is for them to be trained and delegated as data entry on that study. So, there’s, like, protocol training, there’s systems training.

372 00:51:22.660 00:51:35.629 celialeeks: Kim Malley can expand on what else is required, but it’s… it’s per study. Every single study, you would have to train up on those things, and every single study, you would have to, like, sign the delegation of authority log, like.

373 00:51:35.630 00:51:41.770 Jake Sandler: Well, Lindsay, before we go down that… that trail, I just want to confirm, but we… Sunstone is first…

374 00:51:42.150 00:51:44.160 Jake Sandler: Documenting a version.

375 00:51:44.290 00:51:53.089 Jake Sandler: that then you submit through the sponsor system, right? So there’s, like, a copy, the Sunstone copy, that then gets entered into the sponsor system.

376 00:51:53.940 00:52:04.150 celialeeks: Yeah, you read the source document, and you manually enter the data into each sponsor system. And that should be in ECW, or on paper, or in the ECW.

377 00:52:04.150 00:52:05.100 Jake Sandler: Just to…

378 00:52:05.100 00:52:16.949 celialeeks: source of the information, not necessarily… and there is more in the source than is captured, you know, in the ADC, so I think it’s just important, as long as we’re paying attention to

379 00:52:17.780 00:52:19.819 celialeeks: to that. Well, they do…

380 00:52:19.820 00:52:24.210 Jake Sandler: Okay, so I just want to be clear, like, we’re not… we’re not… the brain forces, I don’t think, need access to any of the sponsor systems.

381 00:52:24.210 00:52:25.010 celialeeks: Because the…

382 00:52:25.010 00:52:35.219 Jake Sandler: data that would be entered, as long as that’s captured elsewhere, that’s what’s important, and then we can sort of… so I just want to, like, make sure you don’t feel like we have to give access.

383 00:52:35.620 00:52:37.200 celialeeks: If you do, like.

384 00:52:37.450 00:52:48.250 celialeeks: we do have our own, like, you know, for our IITs, for our Sunstone-sponsored studies, we have our own EDC that we can always give access to to give,

385 00:52:49.140 00:53:07.089 celialeeks: So, in the context of what that looks like and what data, how it’s entered, what kind of data is captured there. We also have guidelines from each sponsor that details exactly what they’re asking for. So, between the schedule of events and the protocol and the ELCRF guidelines, you can see clearly what they’re collecting.

386 00:53:08.250 00:53:20.520 Jake Sandler: Yeah, I mean, Kim, that’s a great point. I think, honestly, the guidelines and the schedule of events for each trial would be fantastic information, because that would lay out the flows for Brainforge.

387 00:53:20.810 00:53:31.740 celialeeks: One thing to flag around the studies is the nomenclature around the studies will be inconsistently listed, and that’s kind of one of the…

388 00:53:32.480 00:53:39.419 celialeeks: The hard parts about trying to find some of the data is that You know, like,

389 00:53:40.430 00:53:54.289 celialeeks: For example, our original study is 49… 347, yeah. 347, but we also call it COMP. No, no. 360, yeah. Yeah, it’s just… I don’t know. Whatever. There’s, like, several…

390 00:53:54.470 00:54:01.580 celialeeks: names for it, so sometimes… It’s hard to tell what patients are attached to which trial.

391 00:54:03.190 00:54:22.149 Robert Tseng: Yeah, I guess, quick question on that, since I understand that, like, medical code, we’ve worked with kind of clients that deal with legal code as well, like, it, like, the classification is not so black and white sometimes, you know, like, and especially since we’re relying on language that’s coming from docs and not, like, programmatic filters that we’re used to doing, and

392 00:54:22.150 00:54:31.700 Robert Tseng: maybe, like, more industry-agnostic data work. I wonder if that’s, like, a problem where, like, who’s, who’s, like, gonna be the… who’s gonna make the call, like, what…

393 00:54:31.730 00:54:33.190 Robert Tseng: Like, what gets…

394 00:54:33.360 00:54:49.190 Robert Tseng: how we do this classification? Or do… if we run into those kind of gray area, do we just, like, kind of surface that to the team and just kind of ask? I guess… yeah, we don’t want to make… I think there’s so much more cost… costly to make errors here for Unsure.

395 00:54:49.590 00:54:51.840 celialeeks: Between Celia and I, we’ve gone…

396 00:54:52.240 00:54:53.779 Robert Tseng: You probably know. Okay, got it.

397 00:54:55.720 00:55:12.469 Jake Sandler: Then I think in terms of coming up with proposed, like, universal identifiers, I think that’s an exercise we’ll want to go through. It’s like, okay, here are the six ways this one trial is called. Now, let’s, like, come up with a structure that makes sense going forward, make sure everybody

398 00:55:12.520 00:55:18.540 Jake Sandler: Feels good about that structure, and then we can get sign-off, and then going forward, we can start to, like, normalize.

399 00:55:20.080 00:55:20.660 Robert Tseng: Yeah.

400 00:55:21.220 00:55:26.760 Robert Tseng: I’m not worried about the patient stitching part. I think, like, to Ekta’s point, there seem to be enough data points that, like.

401 00:55:26.760 00:55:41.259 Robert Tseng: I think there’s even, like, a… there’s, like, a principle where, like, you only need, like, 3 to 5, like, kind of distinct, like, fields to really match, like, 95% of, like, data. That’s kind of what people do in the industry when they’re trying to, like.

402 00:55:41.260 00:55:50.709 Robert Tseng: kind of identify anonymous users, like, online. So, anyway, I think just applying that principle here, I feel like we should be able to get very close on that.

403 00:55:59.330 00:56:10.430 Robert Tseng: Okay, yeah, I mean, I think that’s… I mean, this is… this was great, really appreciate everyone’s participation here. Yeah, I think what we need to do is kind of harden a bit, like.

404 00:56:10.550 00:56:15.959 Robert Tseng: What does the ultimate sponsor kind of delivery… deliverable look like?

405 00:56:16.090 00:56:30.480 Robert Tseng: So maybe, like, I mean, I think that’s one… one learning that kind of got surfaced from this call, that, I mean, obviously, we just had… we kind of made some assumptions here, but I think really kind of clarifying that earlier on will… will help us better

406 00:56:30.480 00:56:39.660 Robert Tseng: Kind of, like, have a line of sight towards the end here, and not kind of get lost in the weeds of trying to, like, address everything in the backlog that comes up.

407 00:56:39.690 00:56:41.990 Robert Tseng: And then I think…

408 00:56:42.330 00:56:53.990 Robert Tseng: yeah, obviously, to Greg’s point, like, even if there isn’t, like, a consistent, like, kind of patient journey to fall back on, you know, really understanding, kind of, like.

409 00:56:54.290 00:57:13.050 Robert Tseng: Well, I think there’s probably consistency, you know, in different phases, so maybe, like, everything in the pre-screening is, you know, it’s probably a consistent experience. Where do the paths sort of diverge? And, like, I still think his exercise is super helpful for us, because, you know, otherwise, like, you know, on the engineering side, we typically will just, like.

410 00:57:13.100 00:57:30.730 Robert Tseng: we, you know, we would make… we may cover too much ground, and it’s, like, not as focused, and we don’t get to the goal as quickly. So, I think that’s kind of, like, the quantitative and the qualitative working together here. So, hopefully, even just from the questions that we asked today, the team could kind of see, like, how we’re poking at this from a few different angles.

411 00:57:33.580 00:57:34.280 Jake Sandler: Hey, Zane.

412 00:57:34.550 00:57:52.289 Jake Sandler: Thanks, Robert and team. In terms of next steps, I think, Lindsey, can I follow up with you just, like, on the access, making sure that we’re getting those two emails, and then once the two emails are good to go, like, access to the systems that we talked about today, are you the right person to coordinate with there?

413 00:57:53.080 00:58:09.580 celialeeks: Yeah, so, what we’re talking about with the job form, that’s accurate. So as soon as, Julie submits that job form, it will be routed to our IT team so that they can get that access set up.

414 00:58:09.690 00:58:27.880 celialeeks: But yeah, if there’s any questions, I know it’s, like, a lot of buttons to click and information to fill out, feel free to reach out to me. I will be out of office, from tomorrow until, next Monday. I’ll be back on Tuesday, so…

415 00:58:28.250 00:58:39.169 celialeeks: yeah, if that’s something that you want to connect with tonight, happy to. If it’s something that, you know, comes up later than that, then I can be available as soon as next Tuesday.

416 00:58:39.980 00:58:56.819 Jake Sandler: Great, appreciate it. And then in terms of creating a shared Google Drive folder where this… the Sunstone team can drop files in, that would be good for Brainforge to access, do you recommend we do that sort of in the IT sub-shared folder, and just create access to this team there, and then that way we can sort of dump everything?

417 00:58:56.920 00:59:02.450 Jake Sandler: In there, that, like, the protocols and guidelines you guys mentioned, stuff that would be useful to give them the context.

418 00:59:03.150 00:59:04.819 celialeeks: Yeah, I think that makes sense.

419 00:59:05.360 00:59:15.910 Jake Sandler: work for Celia ActaChem, if we create that shared folder, and then we could sort of send specific requests for some of the files mentioned today to sort of add a copy into that file, into that folder.

420 00:59:19.380 00:59:19.970 Jake Sandler: Right?

421 00:59:20.270 00:59:21.220 Jake Sandler: Amazing.

422 00:59:22.170 00:59:23.980 Jake Sandler: I’ve heard anything else on your side?

423 00:59:23.980 00:59:31.600 Robert Tseng: Yeah, I guess, we… so we’re gonna have, like, a weekly touchpoint with you, Jake. I don’t know what this group, like, I guess…

424 00:59:32.160 00:59:47.969 Robert Tseng: I guess, you know, we didn’t share it today, but, you know, Greg put together, like, a weekly deck template that we basically were going to share updates with, so just wanted to understand, like, who the audience would be if we were to kind of… I guess it seems like it’s a lot of just, kind of.

425 00:59:47.970 01:00:00.379 Robert Tseng: meeting, kind of talking to people individually, or do we want to just kind of keep… keep things… do we just, like, have a… is this the group to just, like, fire all the questions into? I’m not really sure how to route all this, yeah.

426 01:00:00.380 01:00:15.829 Jake Sandler: Yeah, so I think for now, why don’t you guys, like, let’s debrief Robert and your team and me after this, and then as we collect, sort of, the next level specific questions, what I want to do is, sort of, pull folks from this group in as needed. If it turns out we need

427 01:00:15.990 01:00:35.039 Jake Sandler: everyone here weekly to… this can be most expeditious. I think that’ll be my backup option, but my preference is… wanted everybody here to see, like, the full kickoff, and then if you need, like, more Health Cloud stuff, we’ll act into a meeting. If it’s, you know, so we need Celia’s background knowledge, we’ll pull Celia in for a meeting, and Lindsey, and so we’ll kind of…

428 01:00:35.220 01:00:42.079 Jake Sandler: Do as needed, but if the questions sort of end up being across everybody every time, then we can reassess that approach.

429 01:00:43.060 01:00:44.840 Robert Tseng: Okay, yeah, that sounds good to us.

430 01:00:46.800 01:00:55.689 Jake Sandler: Awesome. Well, thank you, everybody, I really appreciate it, and if you have any questions at all, don’t hesitate to reach out from the substance side, and I’m happy to do some one-on-ones and give more context.

431 01:00:56.280 01:00:57.669 celialeeks: Great. Thanks, thank you.

432 01:00:57.670 01:00:58.329 Robert Tseng: Great meeting rewind.

433 01:00:58.330 01:00:59.180 Greg Stoutenburg: Great to meet you.