Sensors and Predictive Technologies for Environmental Monitoring Panel | Edge in Tech Symposium 2024
- So next I'd like to introduce our next group of panelists who will be discussing sensors and predictive technologies for environmental monitoring and population health. I should say this panel, the people you see in front of you are just the tip of the iceberg of all the work that went into getting them in front of you. And so I really wanna recognize the people who helped from the program committee from each of the campuses to identify and conceptualize the panels themselves. And so in this case, the panel was organized and convened by Michele Barbato and Tod Stoltz, here at UC Davis.
And I wanted to say that actually about the previous one as well, that Katia Obraczka headed our colleague Michael Mackin at UC Santa Cruz also did a tremendous amount of work making sure that that panel came off well. Lorena Garcia, a professor of epidemiology and director of undergraduate education at UC Davis will be the moderator. So I'll welcome the panel to the stage. (audience applauding) - Good afternoon, thank you so much for joining us today.
It's a pleasure to be here. My name is Lorena Garcia, I'm a professor of epidemiology at the UC Data School of Medicine. I'm also a health equity advisor for the Office of Health Equity and Diversity Inclusion for UCD Health. But let's begin by presenting our guest speakers, our panelists, I'll give you a brief background, I have to put my glasses on now. So this is non-technology here. So first I would like to begin by presenting Dr. Colleen Naughton.
She's an associate professor in civil and environmental engineering at UC Merced. Her lab designs sustainable and culturally sensitive food, energy, water systems, through lifecycle assessment, geographic information systems, integration of anthropology and engineering and effective science policy. Before coming to UC Merced, she served as a science and technology policy fellowship with the Millennium Challenge Corporation in Washington DC through the American Association for the Advancement of Science. And certainly, Dr. Naughton has a few more accomplishments,
so her bio is available for you. Next, Dr. Katia Vega is an associate professor of the Department of Design at the University of California Davis, where she founded and directs the interactive organisms lab. By blurring the lines between human computer interaction, the skin and wearable technologies, her lab leads interdisciplinary explorations of the next generation of organisms device symbiosis. (Dr Garcia sighing) (audience laughing) And there's a lot more.
And next I would like to introduce Dr. Julie Fruetel. She currently serves as the Sandia National Laboratories Campus Partnership manager for UC Berkeley, UC Davis and Stanford. In this role, she is responsible for developing strategic partnerships in strengthening the talent pipeline for the laboratories. Dr. Fruetel joined the academic programs for the global security programs office where she led the global security laboratory directed research and development investment area portfolio, supported division level strategic planning and held the program deputy role for the Homeland Infrastructure Security and Resilience program within the energy and homeland security portfolio.
So first let's begin with Dr. Colleen Naughton. - Yeah, so I'll be presenting about environmental monitoring for public health and equity. So this is a large project, Healthy Central Valley Together, we have our project manager in the audience, Mira Souza and Tod Stoltz, also helps us overall with the project. And so y'all had lunch, maybe you visited the bathroom afterwards before here, so apologies, I'll be talking about poop or wastewater. And so this project is a huge collaboration to bring more wastewater monitoring for public health to the Central Valley, that we're more disadvantaged communities as you can see in like the Valley of California in kind of a healthcare desert.
So wastewater monitoring can provide data about the different infectious diseases. So we've notably monitored for SARS-CoV-2, but we have nine other pathogens now, flu, Mpox, Norovirus and adenovirus and some others that you can see up there. And we also support health equity through this partnership with the different local health departments, since there was like lack of access to testing more in the Central Valley and to vaccines, lower vaccination rates and Hispanic individuals were more prone to COVID-19 and had higher cases of hospitalizations and death. And so, we also develop new ways, so the part of this panel is predictive technology. So there's ways to use this data to predict hospitalizations or test positivity rate and infections.
And this is kind of a streamlined snapshot of the process, so I think I was talking to Mira at lunch and we were saying like soft serve ice cream almost. You open the tap and we take primarily like solids if you have a primary clarifier for the wastewater treatment plant and these solids concentrate the virus more, but you can also take it from the liquids and you pasteurize the sample, especially at UC Davis, I think we had pasteurization to keep everything, everyone healthy in the lab and you extract and concentrate and you do similar quantification that you do with clinical tests or your PCR test or if you remember the nasal swabs and you can get information within one day of a community. So Merced's about 91,000 people served by the sewer shed, so one wastewater sample can give us information about most of that community and not relying on test seeking behavior. And since this was more about like sensors and technology, you can also do this at the building level. So our faculty lead is Dr. Heather Bischel at UC Davis, she's on sabbatical in France, so enjoying her time there.
But early in the pandemic we were able to monitor at the different dorms and nodes. So you can see that you can actually put auto sampler down a maintenance hole and sample every like 15 seconds to get a representative sample of the day. And you can see that we could categorize the different dorms if they were higher levels or lower levels and you can alert the dorms for testing or isolation. At University of Arizona, they actually quantified the signal and were able to estimate that there were two students infected in the dorm.
They tested everyone, they found the two students, isolated them before they could spread it to the rest of the community or the rest of the university. And so, this is our data from the eight different cities over time. So you can see the waves of the pandemic and we're luckily in a lower tier right now from national level and this has been a culmination of a lot of work in the background to show the different rising and falling of the waves to alert the community.
And then we were able to match that, this wastewater data does align well with health metrics. So we lagged the data, we brought back, usually it is anticipating the hospitalizations by 14 days cases, by nine days in Merced and then, even with the decoupling, you can see in the second curve that the wastewater data is higher than the case data because more people are doing at home tests and like less PCR tests that are not reported to the health authorities. So then this has shown that even with like lack of clinical testing and these data in the central valley and like more data scarce areas, that we were able to use this tool. And now, so Dr. Miriam Nuno,
she's a UC Davis associate professor in biostatistics and her team, this is Dr. Maria Daza-Torres, has been developing different models with machine learning to relate wastewater data, are effective or also hospitalizations test positivity rate. And so we're able to kind of estimate and so our effective kind of shows for every infected person, how many people they may also infect. So as we combined it with California Department of Public Health different models and for Yellow County we're at one, so we're at a stable level, which is good news.
And so, we're able to kind of even project this data out into the future with the different curves of where the pandemic or infections might be going. And just bringing this all together, it's been a great collaboration with the local public health departments. We used to meet weekly, now we meet monthly, but we send weekly reports to them and we have interviewed our different health officers and they would tell us like how useful the data has been in their community.
And in Stanislaus County we're able to inform the hospitals, especially with shifting variants, that their monoclonal antibody treatments might not have been as effective. So it's, as an engineer, good to see that like some of this data from wastewater can be really useful to inform public health. So just, yeah, thank you, this is just a case study of how different environmental measurements can provide public health data between engineers, public health officials and like statisticians. But it's a huge collaborative effort with a lot of different universities and organizations, thank you.
(audience applauding) - So I was was part of this award previously and it was during the pandemic, so it was online. So I'm very happy to be in person with this community, I'm very happy about this. I would talk a little bit about my research, also probably following up this way to sense, but now probably a little more closer to the body and re-imagining how we all have different kind of sensors and reuse our two meter squared of skin as a platform for sense in the body. And I lead this lab that is Interactive Organisms Lab in the department of design and I want you to start thinking on like daily products and how we could reimagine them with the use of technology. One of them is makeup and if you think about makeup over that time it didn't change that much.
I have a realistic, my mom had a realistic, my grandma had a realistic, if we think about how we could envision these materials and in this case I will show you some examples of how we can use electronics as a way to have our skin as an interactive platform. I think I might have some issues with the videos, but I'll tell you a little bit more about that. In this way we can think about makeup and in this video I was showing how we could have different muscle interactions. So for example, you blink, control the lights or I have other projects I will show you, like with your fingernail you could pay the metro or how by just touching your hair you could send a message to the police asking for help. - [Participant 1] Well we have little text.
- [Participant 2] Access to your video. - I can see that. (all laughing) Well let me move on a little bit more on what I want to bring actually for today. That is how we could actually start thinking about the skin as a way to read. And this is a book that we brought that is called, "Beauty Technology." That is how we use these electronics on the skin, on the body.
And for me what's interesting to start thinking about that, for thinking about these objects that we could put around the body to have that access. But then I start right now in my lab thinking more about the body as an organism, as a way that we could have all these body fluids as similar as we were talking before, with Professor Collin that we could actually have access to that information that we usually don't have access to, unless you go to the doctor and have, for example, a blood test, so you could have access to your fluid data basically. And when I talk about this in terms of medical and wellbeing, in a way to monitor your body, you could think about different kind of analytes that we have already in our fluids. Our fluids like sweat, like saliva, like tears and in this way to transport you to rethink on the skin and to think about its possibility, the skin could be used as a display. And when you think about our cells, they could become somehow pixels.
And this project is called "The Dermal Abyss." It was a collaboration between MIT media lab and Harvard Medical School and we use biomarkers, like these biosensors that change color and instead of using traditional inks, we were using these other kind of inks. Then when you tattoo, then you could be in contact with the interstitial fluid and have access to that kind of data. So for example, we could think about glucose levels that change and we could make it it visible, like something like oil like over here.
So we think about these biosensor, we see how they could change colors or we could also have other ones that change fluorescence and we could include that in between the skin and to have that information with a computer vision device, for example. And that way this is another project we were working on and each of these tattoos could have different kinds of information. So for example, albumin, glucose or pH and depending of different kind of illness could be all of them connected or not.
Of course I cannot make a tattoo to you today, there is kind of like a lot of development to be done for having these biosensors inks. However we put it that as a proof of concept that how we could envision new ways that we have biosensors closer to the body and having access to data that you usually don't have access. In the same way, probably a little less invasive than a tattoo, we could think about saliva. Saliva is a way to transport also information that is transparent, it regenerates all the time, but also it has this direct contact with your metabolism and this other project called braceIO, we develop these dental ligatures, like for braces, that each of them already have different kind of colors, but what if they will be reacting depending of different saliva levels? Like for example, we were working with different kind of biosensors and you can imagine interactions with pH, sodium, nitric oxide for example. We develop different fabrication process as well and we could use also computer vision to have access to that information. We could also talk about, if we go deeper into fluids, we can also go into sweat and sweat again as a way to sample from the skin, a different kind of information.
In this way, we work with designers, generally designers, that instead of having traditional ways that we see these kind of biosensors and you could see it here on the right, how they traditionally look like, how we could change the shape and it could be generally, so in that way, instead of having all these ways that we throw away whenever we use these biosensors and if we imagine for example people with diabetes or other kind of diseases, they use it very often. What if we could repurpose those biosensors and convert them in jewelry pieces, like earrings or some pins. This is a project BioSparks and we were kind of like also making this connection with crafters and these techniques that already exist from designers and again, my work is a lot of how to bring these technology that are usually exclusively from science or engineering for being closer to users. And I'm interested and this is also part of the NSF career work, the grant that I'm working right now, in how to think about body modification technologies as a way to embed, or as a way to have it as a substrate of these biosensors, electrochemical, colorimetric or fluorescent. And that is why we could think about the skin as an interactive display that give us access to information we use.
We embed technologies in a way that is seamless to the body, like as much as possible, as closer as the way we wear technology and how we are and also how instead of, if we think about different devices, wearable technologies, that is my main field, we have to press a button, you have to to touch it or there is kinda like some particular interaction. I'm interested in thinking about that interaction is your own metabolism itself, that is interacting with the body. Just to thank my lab members and their amazing work, I'm also part of the graduate group of computer science. So a lot of my students goes from design, computer science, chemical engineer that come together to rethink this ideal with human device and biosis, thank you.
(audience applauding) (panelists laughing) - [Dr. Garcia] Our next presenter is Dr Julie Fruetel. - Thank you, it's my pleasure today to be here. Can you all hear me? - I think so - [Crew Member] You should be good. - I'm good? Okay. So I don't have any slides today, so hopefully you can still stay awake.
(all laughing) It's hard to follow Katia, I understand. (all laughing) Wonderful picture. - Thank you. Let's see, so I work at Sandia National Laboratory. So I'm gonna give a little bit of a different perspective for AI monitoring. So Sandia is a national security laboratory.
It has a very broad national security mission. So mostly known for nuclear deterrence but we also have programs, significant work in things like energy security and climate security and biodefense. And so when I came to Sandia and it's been a while now, in the late nineties I worked on a project to leverage Sandia's micro-fabrication capabilities, so they have significant capabilities that derive from developing radiation hardened microelectronics that need to fit into hypersonic systems. But we have people that came in and said what can we do with this for bio-detection. Could we use this kind of technology to think about aerosol and air monitoring? And so, we had an internally funded program, it was a grand challenge, where we invested capability to develop the ability develop microfluidic chips and hired people like me, biologists, to develop assays for proteins, bacteria, viruses and toxins.
And together we worked on putting these assays on chip, building the device to have secure and leak proof connections to the chip and putting electronics on the chip so we could put high voltages on there to run electrophoresis. And so, it was in the early days of microfluidics and it was a really exciting time. We had a diverse team that came together, cross-disciplinary, it was kind of one of those experiences where I hadn't really experienced something like that before.
I was used to being an individual researcher in a laboratory, so it was a very powerful experience. So Sandia has other technologies in bio-detection that they've done, other things, microfluidics, looking at not chips with channels for electrophoresis but round devices that can be sent on a centrifuge and spun with radial channels on them. Introduce your sample in the middle, it migrates to the end and reacts with reagents and the detection points on the outside, they call it spin DX. That's found some applications in medical diagnostics. Other technologies are something we call Bad X.
So this was a case where they developed a chip size microfluidic device for low resource environments. So it was part of a mission space at the lab where staff would go out and work with other countries to secure biological pathogens and ensure biosafety and biosecurity in public health labs in the world. And they saw that in low resource environments, typically assays were still back to culture, they might not be using PCR for example or molecular diagnostic assays.
And so they came up with an idea and developed it and ultimately developed IP and transferred the technology to a company where you can introduce the sample onto a chip and run a culture on the chip for bacillus anthracis, so the causative agent of anthrax. You could let that run overnight, open a valve, it would run a lateral flow assay, give you your result and then there was a button on the chip that you could push that would release bleach and it would completely decontaminate the chip. So you weren't generating more pathogen by assaying it and that was really kind of the goal, was to come up with those ideas and then at the end they could just throw it away because it had been decontaminated. So these are some of the kinds of microfluidic devices we were interested in. The microfluidic field now is like an over $20 billion industry, it's really taken off in a lot of applications.
But we were kind of working on that early edge where we were thinking about this, that it could have really good applications for national security events, excuse me, and leverages some of Sandia's capabilities. So kind of in the lab, in a lab environment, our motto was sort of, "Smaller, faster, cheaper." We were really excited about taking the analytical laboratory and bringing it to the chip and so as small as we can make it was was very exciting, handheld was the goal. Faster, we wanted results in seconds to minutes and not hours. So you could literally be out in the field, potentially, and measure something. And then cheaper, ultimately the goal was to think about materials like plastics that could be easily manufactured, could be disposable, like the Bad X cartridge type of thing.
So I wanna contrast that experience with where I went to next in the lab. This was all before we got to where Lorena was talking about my pedigree and I was really back in the lab in the beginning, but I went from there to a systems analysis group where we did a lot of modeling and analysis and the space that we worked in was sort of counter-terrorism. So we worried about aerosol releases of pathogens and the harm they could cause to a population. And so we had looked at what did these scenarios look like and what are possible ways to prevent them or at least detect them and then by detection ultimately mitigate them and try to preserve as many lives as possible. And so, we had found that if you could detect a release of a pathogen, say, and this is true for certain pathogens, but if you could give people treatment, before they started to show symptoms, so in that incubation period between exposure and symptoms, then you could potentially save a number of lives or you could at least lower the infectious dose, they wouldn't get as sick. So the modeling results suggested that if you could detect something early before they got sick, then you could potentially save lives.
And so that was a motivation for a program called BioWatch where that system is a series of deployed aerosol collectors with laboratory analysis of the collector filters and they would look for various pathogens. And so, you looked at that system, it wasn't smaller or faster and cheaper, it wasn't small, the aerosol collectors are large, 'cause they draw large volumes of air. It wasn't fast, particularly, it was typically hours, you had to pull filters and analyze them, so we were talking hours not minutes and cheap, there were a lot of people involved. So there was a lot of costs associated with labor in this system. So here we thought, okay, surely smaller, faster, cheaper, that's the solution and if you ask the end users, they're like, yeah, that's not the issue. So what was the issue? And the issue for the end users was to try to understand how to use the information to make public health decisions.
So the end users ultimately were public health. Typically public health officials are used to seeing disease in a community, by case reports. So you get sick, you go to the doctor, you get a diagnosis, particularly for certain pathogens and diseases, there's a case report that gets logged, we can see those, public health official can see those cases and then they can use epidemiology techniques to try to understand potentially the source, depends on what it is or maybe understand if it's contagious, if it's spreading in the community, they could use those kinds of techniques to understand are our protective measures effective, for example. So here we came up with a system and said, "This is great, look it, you get information before that, you can have this huge benefit."
And they said, but they didn't have their gold standard method. They didn't have their usual way of understanding disease in a community, where we're taking them outside their comfort zone. And so what we ended up doing was spending a lot of time working with the end user, developing concepts of operation or conops, what are you gonna do if you get this information? What additional information would you like to have? When could you get it? How would it fit into the picture of what you would decide to do? Could you do some kind of low consequence things, like pre-stage antibiotics or vaccines or whatever your treatment was gonna be? And so at least then you take in some steps to speed up the potential response. So I just make this point like, we were in the lab so excited about technologies and then when we worked with the end user, we found out like, they had a whole different set of requirements for thinking about the impact of the technology. So I put that out there, to think about developing these technologies and who is your end user? What's the use case? How is it gonna be used? So those parameters you're thinking are very important.
Maybe cheap is really in the wash compared to, I need high accuracy, high precision, I need the assays validated, I might need FDA approval, to be able to act on those kinds of results. And so just one last reflection on this. So I mentioned a national security laboratory where we're interested in trying to anticipate threats to national security or that might have national security impacts. And so often we're working in a paradigm where before there's any kind of market.
ChemLab was an example, the microfluidic device project where we were sort of working before there was a market, people were thinking about it, although that came along too, but we were kind of working at the beginning of the field and so often you're kind of high risk events, but they're high consequence, but maybe low likelihood, but they could still be overall high risk if the likelihood was enough or it was credible. But if you had asked me back when we started working in this area, the value of indoor aerosol detection of a SARS coronavirus, I would've called that a low likelihood high consequence event and now as you know, the world has changed. Now we've had the pandemic, I saw in, "Nature," there's a nice proof of concept article about an aerosol collector to an antibody based assay for SARS-CoV-2. So you could find out in minutes if a room had any SARS in the air. So now that was a proof of concept, it's still the market, but now it seems it's kind of a different paradigm where now it's kind of clear that that kind of technology would be interesting.
So I just kind of to conclude my thoughts on developing these, one, understand the use case and the end user and also think about what are those competing technologies and gold standards that you either need to improve upon or address a gap that they're not addressing, that may be useful. And then finally that it may be that your new technology is gonna require a paradigm shift, it's a different kind of thinking about it, in order to have the impact and the adoption that you envision it to be. So anyway, those are some thoughts for today. - Thank you. - Thank you. (audience applauding) - Well I certainly have quite a few questions.
I hope you do too. We're now moving on to our question and answer period. I have several questions, but I will begin with the first one so that we can start off and thinking about all of the work and the research that you're doing, how can advanced science techniques, particularly health technology contribute to more effective environmental and public health strategies? And you're all welcome to answer, who would like to start? - Yeah, I guess we can go in order, still.
(Colleen laughing) Well I showed like the case for wastewater, like it has been more effective that we can test like a wider pool, like even Los Angeles' wastewater treatment plant serves millions of people so we can get information about a large population and then kind of put out messaging a little quicker and more tailored to increases or decreases in different pathogens than we would from traditional public health metrics, but usually we work in tandem together. So it's really helped us like kind of serve a wider population with this tool. - Yeah, a lot.
Yeah, I definitely don't work with a large population but I think that a lot of the projects we develop that are kinda like proof of concept of future or different devices that coexist, we always kind of ask different kind of questions, even during the review process of the projects of what these projects could mean for the population and just to bring kind of like one idea, if we think about the two projects, that it came out like very kind of publicized probably, we have like hundreds of emails of people, mainly with diabetes that would like to use this kind of tattoos, even like when it not in that traditional population of the tattoo, were like users, like for example someone that were more elderly or like two years old kids, just to have access to that information. But it also comes with a question of what it means to reveal that personal information from your body, like you are kinda like showing these tattoos that are changing colors, what that means that your personal and private data in some ways is kind of like visible for other people and we came out with some kind of idea, so how to contrast probably, like not following very much kind of like what is in the lab settings right now, what's possible. But it's kind of like something that I wanna bring into discussion because I feel that when we develop these kind of projects, privacy, it comes a lot into that conversation of what it means to have this available, that is part of you and sometimes you need to do that, you want to do that, but it's also health information that I think it's interesting to discuss.
- I think some of my thoughts on that, I covered a little bit of my talk of thinking about these technologies and engaging the end user up front. So, for wearable technologies it might be, would you be willing to share that information? Should it be something under your sleeve or something only you can see. Is that information and that data useful to your doctor? Is it something your doctor could monitor? Would they want that sort of information? So I think about in my line of work, we often walk through scenarios. If you think about a technology or something that you're trying to detect, how would it play out? What would it look like? What are the considerations? And then, who are the various stakeholders? And then talking with those stakeholders to understand what their needs are, what their willingness is, and then kind of cycle that back, to see if that would change the way you designed it or the implications that it would have for that. So I think those are key. Privacy is obviously huge, trying to translate from individual health to population health is a huge question and I think that's one of your follow up questions, Lorena too, so...
(Julie laughing) So what are some of the issues and ethical issues involved? - Absolutely, but I also wanted to make sure that we opened our questions to the audience. Do any of our audience members have a question or questions? We have a mic, yes. And yes, Dr. Otten, I do have a question on ethical considerations. (all laughing) - [Participant 3] Hey, can you hear me up there? Is that fine? - Yes. - Uh, huh.
- [Participant 3] Yeah, okay, cool. So I would just wanted to ask like a lateral question of you all and maybe this is a little hypothetical, but last year EPA put out an FOA to ask about bio-technologies and biomaterials and other types of nano-materials for doing PFAS remediation detection and like degradation. But in terms of talking about biological sensing, if EPA were to put out like another FOA to ask how can we do something very interdisciplinary and very cross-cutting in order to look at very large sensing networks for looking for a spillover event. We're in the middle of one, so more to come.
Do you have like considerations about what the types of technologies should be in order to facilitate something like that? Like would it be quantum sensing? Would it be something that had other types of like performance criteria that you felt? I mean as you just mentioned, Julie, that it was the users who needed to make some type of informed decision on these and that it's great if something's really advanced, but if it doesn't help the users make really useful informed decisions, that's a major bottleneck. - Okay, I think the trend, if I understand the question, to think about, if EPA put out a call for broad area monitoring for a release, is that right? What would that look like or what would need to happen in order to really understand how the technologies, what technologies might play into it. Is that generally your question? - Well, yeah, I mean actually, if new technologies are coming out, new data materials and other types of sensing technologies are still coming out, do you think that there should be some type of prerogative or there could be some type of lateral prerogative on looking at large testing networks, such as wastewater testing integrated with like other things? Would there be like a uniform set of criteria that could be put out to say this is in fact what is really useful for putting out a network like this? - I see, I see. I think, yeah, I mean if the problem space is studied well, to understand what are possible technology use cases and it's understood sort of what would be the requirements from all the various stakeholders, that would be playing into the scenario, it could be, I mean it would depend a lot on the specifics, but it could be, I mean that would be the process I would think that they would need to go through.
So if, yeah, you have to define the problem to understand what technologies could play or what the requirements for technologies could play and then given those requirements, we really need information quickly in this case, we need information that looks something like this, then you can think about what technologies might be able to play into that. But it's important to understand all the various stakeholders and then things like the data and who owns the data, what decisions are gonna be made and actions are gonna be made, based on the data, will they have access to that, the context for the decision making. So it's a complicated thing. This type of problem is something I worked on in my systems analysis career, but I also would comment that if you came up with technologies that are appropriate for one use, or even requirements for technologies in one use, would they be applied to other uses? Maybe or maybe not, right? It might be something completely different. It might be that technology's developed over in one space, people use them. One time we had a panel with our handheld devices and we met with a group of firefighters and hazmat to understand what would it look like? We showed them our technologies, they talked about their day in the life of what it would like look like and this was post 9-11, post the anthrax letters, so white powder spills were very common, or not common, but they happened, and fire departments, first responders were often the first people on the scene and they had to understand what to do about that.
And so they commented like, we figured it out, when we go to a fire, we kinda have a playbook in our head. What's going on? Where's people? Where's exits? Where's smoke? Where's what? We put together a plan of action. So he says, we'll do it with what we've got, but if we have something, technology that's well designed for what we've got, that'd be great. But sometimes we have technology that we use for certain things and other pieces for other things. So those kind of concepts of operation are pretty interesting. So it's really trying to understand how people use the information from the technologies to figure out what really is gonna help.
It's a fascinating, fascinating problem, so. - Thank you, Dr. Naughton, Dr. Vega, did you have any? - Yeah, I wouldn't like to talk really, because I'm not totally sure if I understand correctly the question, but you were also talking about whether like multidisciplinary teams or what could be kind of like a strategy that could exist to think about how these implications coexist for the technologies that are happening. And I think that that's something that when we are developing these devices is that worries us a lot. And right now we are kind of like developing some workshops that, as Professor Julie was mentioning, like how to make these stakeholders come together.
So for the co-design workshops that involve bio technologies and users or bio technologies and designers, so in that way we could rethink about these different wearable factors, implications that could exist in these technologies and also different possibilities, also to fabricate the same kind of devices, that I think that that's, at least for me is something very interesting, because we are at least in that area I'm researching right now, it feels that there is not a consideration very much on the user and to rethink about like how that person will be using this device, like for example, a user of these diabetic badges, that they are kind like on the air interstitial fluids all the time or they have to pinch their self many, many times. Is there a way that we call still support the users with other form factors or physical aspect of this? And as I'm interested more in the physical aspect of the device, it's one aspect to to see, but as you also were mentioning of what it means that information that could be actually connected to different kind of population, I think that it could maybe be expanded, by bringing the real users to the fabrication of those devices. And I think that that's kind of like something that we are lacking now in science and engineering labs to have that closer view of the human aspect.
Yeah, thanks for bringing that question. - I think just adding, since you specifically asked about PFAS, I mean it was landmark legislation that we're finally getting to those family of chemicals that are causing a lot of really bad health effects for the population and we have the technology, to detect it is challenging, since there's so many different chemicals and then when you ban one, then the company makes a slightly different one that's sometimes worse. So we have those monitoring networks, we've found it everywhere, it is, yeah, in the wastewater everywhere, it's in remote places, it's in Alaska, so it's more, I think, yeah it'd be good to help with more detection but also just as an environmental engineer we look at like pollution prevention and kind of going towards the company so we don't generate the waste in the first place. So a lot of times the wastewater treatment plants then get stuck with this like regulation to treat it and the treatment technologies are really energy intensive and difficult. So it's a very complex problem and you're right that the technology needs to be useful to the end user and a lot of times we have the technology to treat it or manage it, but it's like kind of cost and economics and society.
- Thank you, do we have other questions from our audience? Right here at the front. - [Participant 4] I think this may be more for Dr. Vega. The biosensors area looks really promising, but I also have seen so many like false positives in this area, so I'm just curious, what are a few areas that look promising to you based on what's not already there in the market today? - Yeah, 100% agree with that and there is a lot of research coming out and now that I'm trying to understand more because I mentioned probably before, my background's in computer science and now I'm in department of environmental design, I kind of learned about biosensors and with this goal of bringing these biosensors closer to humans, we come out with this also question that we have to deal right now in my lab is that why we're creating these, for example, specific sweat biosensors targeting a specific anolyte that the range of that anolyte is totally off for the biosensor range that we could have in sweat for example. So that's something that I'm still kind like working on thinking about like what could be like a specific field. Right now we are working in a new project that will be following up what could be these wearable factors that we need to consider, taking into consideration those anolytes, because for example, and there are many of these temporary tattoos for sweat that are in diabetic, for diabetics for example. However the glucose levels in sweat is very, very small.
I think it also needs to be probably more evolution in the biotech labs to have that probably sense level how can like achieve for the those kind of wearable devices for example. And in terms, just following up your question, on what could be a good area- - [Participant 4] What are some of the one or two good use cases that you think have really good promises? - Yes, so following up, like what could be kind of like some directions that at least in my case we're trying to follow, we are right now investigating a specific anolyte probably closer to cortisol for example, to kind of like have a relationship also with different kind of mental levels on women particularly. And also thinking about, (Dr. Vega talking in a foreign language)
and it's kinda like a large project for us, but it's something that I think. that it could be kinda like an interesting path that is not very well explored. There is a lot of projects that are mainly about pH and we also follow that and I think that, at least for me more interesting, is to bring together different biosensors and not specialize in just one, because most illness needs to follow up kind of like a group of biosensors. So I think like the multiplexing aspect, I think it's also kind of like another interesting path to follow. I don't know if that helps. - Dr. Fruetel, did you have?
- Nothing. - No? - Nope. - [Participant 4] I'm also interested to know what water filtration system should I be using at home, (all laughing) to make sure other things can be detected. And even lead, actually the lead levels I've heard are much higher than what our kids really need to be exposed to.
So I'm just curious if you have any thoughts on that? - Yeah, well at least for SARS-CoV-2, it's like viral fragments and like largely inactivated virus, so don't worry about it like going into your water, it's largely respiratory aerosols that it's spread. But yeah, other contaminants, lead especially is really challenging. But like the EPA and like all of your water utilities, if you are on like a municipal water system have like standards and you can look up your consumer confidence reports and they're supposed to report about lead and nitrates and so trust your tap water, to an extent if you have a domestic well or like others like get your water tested.
But I often still advocate that tap water is still very good with these standards that we have and people turn to bottled water, which there was just a landmark study like quantifying all the nanoplastics and everything that you get in your water. But yeah, I know that with the recent legislation that they're trying to replace a lot of the lead service lines in your house. So something to like just watch out for, depending on the age of your home, if you have an older home, it'd be good to get it tested, since that contamination often happens after the water comes from the treatment plant.
But luckily in California it's not as huge of an issue as like the east coast, but it's still definitely in like LA and San Francisco area. - And I did want to segue into that one question that I did have in regards to ethical concerns. So what considerations, now that we're addressing that issue, the collection, the use and the sharing of health data should we consider? - I can also start. I think that in wearable devices, it comes out a lot of the way we collect the data and there are many parts that could go through the answer to that question. That one could be like the materials we use, like a lot of them are chemical materials, so how we encapsulate them or how to make them safe for a user, that comes also like we can like put that in a larger population, if we think about the skin, we could also have like different kind of allergies, different kind of reactions that could happen.
So the use of biochemical biomarkers, needs to have a lot of different implications for the population. Other thing that I already mentioned a little bit before is about what it means to reveal the data. You have a tattoo that is showing your data or a lipstick that's changing color and revealing that information.
We thinking about, as also Professor Julie was mentioning, of how that data could be actually safe. There are some strategies that we were discussing of how we could use a cream with some of these calibration solutions that change its color so you actually activate it whenever you need it. So it's kind of like hiding in some way or you are the one that kind of like just know what are the different colors, that comes probably with some evolution of biosensors and other aspect of that already exists, because we already have all these wearables, is how the data that you are having is protected. So now I'm saying like, oh I could collect all this data from your body, how the data will be actually saved, that for example, like there are very famous cases that start to be kind of like forbidden. Like what it means that your data is not revealed for a insurance company for example.
So they cannot kind of like target yourself in a specific way. So I think that these kind of like three main things are some of the factors that we need to consider if we are moving wearables even closer and more intimate to the body. - So Dr. Fruetel, I know we had discussed it, you have.
- Yeah, I was just reflecting on air monitoring and some of the considerations we've had in our studies too, thinking about sensitive populations. So I said we do modeling and often, you just take an average or an ensemble, but there are special factors of thinking about different populations will have different impacts by air monitoring pollutants or contaminants. And so how do you share that information? What are the implications of that information? Privacy issues, but also should you share it or should you not? If children are more susceptible, a lot of public health communications, in terms of we've done, it's a huge piece and I feel like the pandemic has kind of really illustrated clearly things that we sort of, not took for granted, but it seemed straightforward before it came real, was how do you communicate with the public about things and now with disinformation or misinformation or just so many channels of giving the information, having the one trusted source of information that that's the only one is no longer true. So how do you handle those kinds of implications for people? Are some of the things that I think, yeah, come up in my mind, is some of the conundrums for this, at least from my perspective. - Dr Naughton? - Yeah, no, like it's similar with air monitoring, with wastewater data and just your end use of the data and are you going to be helping the population or are you using it that it? What are kind of the implications that could happen? Especially as we move into like monitoring illicit drugs and opioids more, like, it can be very useful to provide resources to communities, but not more like incarceration or policing. So we just have to be careful and mindful as we set up these programs, how the data's gonna be used, how it's going to be stored.
I think CDC, with their national wastewater surveillance system, like say that they own all the samples, because they said as the technology improves, we have all these stored wastewater samples. So maybe the technology will improve later where we can identify more information. But usually it's a large pooled sample so we're not identifying individuals but we keep it at like sewer sheds or like populations of 3,000 or more.
So getting to the dorm testing or building level testing, you have to be very sensitive. I know they've talked about uses at like Google headquarters or others of like you can detect whether people are depressed or like stress levels or things like that and that gets a little challenging when it's like not for infectious disease or other applications, but usually like the ethics I also focus on is just equity, like we didn't have this monitoring in the Central Valley, it was mostly like urban and coastal California. So it took a lot of like CITRIS pride, some like seed grant money and then like we were able to get some philanthropic funds, but still only 6% of low income countries have wastewater monitoring.
So it's also the ethics of who has access and like is it just still just serving the higher income like populations? And then I was just trying to make the data open so people can use it. So for infectious disease and pandemics, it should be like open data, but as we get to like other targets, then we have to have that discussion of if it's open, how it's shared. Like in Texas, I think they're monitoring schools for fentanyl now, so that data might be a little sensitive. - Well thank you and I did want to just wrap up quickly and given your diverse backgrounds in regard to health technology, from engineering, computer science, pharmacology, toxicology. So what do you believe are competencies that are essential for the next generation of data scientists? (all laughing) The next two, three minutes? - Just just to think about, like, I will say that in terms of wearables in general, technology, there are already these multidisciplinary things that exist, that brain scientists engineered together. And one thing that, at least for me what's more interesting is to bring together all the stakeholders that were not in discussion before.
For example, the projects I was doing previously with cosmetics and electronics, I was always working with nail salons or makeup artists to try to understand how they actually use their products. Even like the tattoo projects, we had different performances to understand how these inks that are different to their traditional inks, they could see them, they could have sense of the skin or with their tattoo guns and what it means for them to have this different material for making those kind of art and technology. So I think that kind of like bringing together the other stakeholders that are closer to humans and to these kind of ways to interact with the body or with different devices, even thinking about landscapers or architects, why they have to be out of the equation. - Things that come to mind, you'll hear my bias, you've already heard, sort of systems thinking.
So this idea of your data is not, think of it not just in isolation, but think of it in the context of what it's trying to inform, what systems it might be used for. Thinking about the uncertainties in your data, if there's an algorithm you're running and analyzing it for, how confident are you in that? And I think this has come up in the other discussions about AI and all this sort of black box, if you will, sort of like data analysis. What are the implications of that? How transparent does it need to be? How verified and validated does it need to be? Understanding the uncertainties in the data and how that plays through, are there alternate interpretations of your data? So that plays into the decision making from these technologies as well. So those are a couple of thoughts there. - Yeah, I think my two things for this question were like lifelong learning and listening. Just as engineering, we always like advocate for that, but when working with the biostats team, they're like using wastewater data, like we had to like explain to them like a wastewater treatment plant, like where the data came from.
So I think in earlier panels, like all the students, data science students, like just wanna get the data and use it, but like actually understanding where it comes, where its limitations are, and then your point about the end user with listening, it's just like you can create all these fancy models and everything, we've changed our code so many times for the public health departments of like levels and like percent change, like just listening to like what's actually gonna be useful to them is really important. - Well thank you. I'd like to thank each of our panelists for such an insightful discussion. I'd also like to thank our audience and it was certainly an honor to serve as moderator for this panel. Thank you so much. (audience applauding)
2024-07-02 15:34
