 Hello, and welcome to my talk, You Are What You Measure, Digital Biomarkers for Insights in Personalized Health. My name is Irina von der Brug, and I'm excited to share with you my journey that started two years ago when I decided to quit my career in molecular biomarker research and dive into the world of data science and digital health. Since then, I started my Quantified Self project and the freelance business named You Are What You Measure. Through this project, I try to track as many facets of my life as possible on the one hand to see if and how I can optimize my own health using this health data and on the other hand to use all the data to improve and practice my data science and data visualization skills. So to do so, I'm using a wide variety of wearables, apps and sensors like this ring, for example, that measures my activity, sleep and readiness. My watch is tracking my steps and my heart rate. When I'm sleeping, I'm wearing a headband that is measuring my brain activity. I have this sensor in my upper arm that measures my glucose, and everywhere I go, I take my air quality sensor or my light intensity sensor with me just in a few things. I also use different type of apps like the Strava app to record my workouts, the rescue time app that monitors the time that I'm using my mobile device or my laptop and the chronometer app to record a food diary. So in this talk today, I'm really excited to share with you two things and I hope to inspire you in at least one of two ways. First goal is to demonstrate the potential of digital biomarkers to its funds, personalized health, whether for personal use or for implementation in your clinical trial or clinical practice. The other goal is to share with you my enthusiasm about R to collect, analyze and in particular visualize digital health data. With digital biomarkers, I refer to objective and quantifiable data collected by digital devices such as apps, sensors and wearables that provide information about a physiology, behavior or environment. Typically, such digital devices allow continuous, if not at least more frequent, collection of data and thereby instead of comparing a single time point to a population mean and a more individualized feedback, even in real time. A perfect example of molecular biomarkers gone digital are the reasons that founders in blood glucose monitor. Various sensor-based devices now exist that allow continuous or flash measurement of blood sugar levels. This helps patients in managing their diabetes but can also provide insights in an individual's response to different food types. Another digital biomarker that deserves specific mention is heart rate variability. Heart rate variability or HRV is the variation in the time between consecutive heartbeats. This variation in heart rhythm is a good thing. It keeps the heart flexible so it can easily adapt to certain changes or stresses. When we are under stress, the variability in our heart typically lowers. HRV is relatively easy to measure with commercially available heart rate monitors. I use HRV myself as an aid to better listen to my body by taking a three minute short-term HRV measurement every morning and then awakened using a heart rate monitor on my finger in combination with the HRV Portraying app. In this chart, you can see my HRV fluctuating over time, indicating a period of stress about here in the middle of May, as well as a strong response of my body to my second COVID-19 vaccination, indicated by a strong drop in HRV. In a similar way as HRV, I track my long-term trends for various auto parameters. For most of these, I don't really care about the numbers at the moment. I just want to keep track and build my personal data bits. But that's not always easy. At the moment, I first analyzed my step count. I tracked with the Apple Watch, which was about here after a year of wearing the Apple Watch in October last year. I was shocked by my sedentary lifestyle. As you see, I hardly reached the 5,000 steps a day. From that moment on, I started to almost obsessively track my step counts to make sure I hit at least 10,000 steps a day, as you can see later on. In the same way, actually, in the name of science, I measure my body weight every single day using the cardio smart scale. Although I try hard to not look at the scale every day, I can't help to do so. I then feel a little bit frustrated now and then. But I really like to keep on connecting this data and just see how my weight evolves over time. Probably the easiest thing to track is your sleep. You don't have to do anything for it. Just sleep, and your data is ready when you wake up. Therefore, I'm using the dream hatband that measures brain activity. And besides tracking the amount of sleep, motion and heart rate, the product algorithm can make a pretty accurate prediction of my sleep phases. So I use the sleep phases to gain more insights in my sleep. I, for example, plotted the frequency of my sleep phases along the night. And here you can see that sleeping in is not likely to get more deep sleep, which is needed for recovery. As deep sleep is this purple band, it occurs mostly in the beginning of the night. So for increasing my amount of deep sleep, I better make sure to go to bed in time. In addition, I keep track of my nightly patterns in HRP, which should show an increase during the night as a sign of recovery, as well as my heart rate. I also measured with the oro ring, which should show an early decrease as a sign of recovery during the deep sleep. Another interesting thing about sleep, or lifestyle in general, is granularity. Going to bed and waking up at about the same time each night is important to keep our bodies by a logical clock functioning correctly and not to suffer from a so-called social jet lag. As you can see, during the one and a half year of the pandemic, my sleep pattern has been nearly excellent. No late night parties in the weekend, no needs to sleep in, and no early morning flights to Keshe. My eating patterns on the other hand is a different story. Working from home has especially messed up my lunchtime. On some days, I'd have my lunch already at 11 a.m., whereas on other days, I completely forget about lunch until late in the afternoon. Talking about working from home and for myself, I got some other interesting insights using the rescue app for my laptop and mobile devices. As you see, I start to use the mobile phone almost first thing in the morning and, yes, I also seem to go to bed with it. You can also see that I had a hard time to close my laptop in the evening or put aside my work in the weekends. So it's clear that I got hooked on data science and data visualization. It's actually pretty addictive. And with so much data that I'm collecting, there's always something more to visualize, to analyze, but not all our blog posts to read or, I must admit, another coding issue to be solved. As another data visualization example, I'd like to end my talk with this data visualization of patterns around my morning period. In this chart, where each ring represents another cycle, you see how the oer ring picks up the rising body temperature after ovulation. Although not very surprising, I was more surprised to see a similar pattern for my breathing rate during sleep. Apparently, some hormonal influences cause my breathing rate to be higher at the end than the beginning of each new cycle. And the same hormonal influences would be seen for the base that I noted a headache. Although I still don't know what is triggering the headaches, it is fairly obvious from that plot that my headaches that occur only during the first half of my menstrual cycle. And with that observation, I'd like to end my talk. And I hope you agree with me that you are what you measure. And if I am what I measure, then, well, I sleep, I move, I work, and I use R. A lot for about two hours, almost two hours a day, every day, including the weekends and holidays over the past year. But I'm excited to keep on working. I look forward to update and update my skills further to automate and share the data upload analysis and visualization tools that I have worked on so far. I hope to create an interactive shiny dashboard. But most of all, I really look forward to working together on some awesome digital health and data-fist projects. So please get in touch. If you have any ideas on those projects, I would be happy to hear from you. For now, I'd like to thank you for listening. And again, please get in touch. I'd love to hear from you. Bye.