Topic Archives: Videos
Bryan Ausinheiler was experiencing gastrointestinal issues for years and decided it was time to figure out what was causing it. By precisely controlling his diet – eating exactly the same quantities at exactly the same time – for a month and then measuring the quality of his stool in a self-designed spreadsheet he was able to create a baseline dataset to better understand his issues. Bryan then developed an experimental protocol that included “elimination and diet variations to figure out the cause of my frequent (3-5x/day) loose stools.” It turns out that “eating too many sunflower seeds was the main culprit.” Watch Bryan’s fascinating talk, presented at the Bay Area QS meetup group, to learn more about his process, and how he tackled self-experimentation and data collection.
Cathal Gurrin is a researcher at Dublin City University and the University of Tsukuba. He’s also an expert in the field of visual and data-driven lifelogging. Since 2006 he’s collected over 14 million passively collected images from different wearable cameras. Add his other sensors and he’s nearing over 1TB per year of self-tracking data. In this talk, presented at our 2014 Quantified Self Europe Conference, Cathal describes what he’s learned over the last eight years and what he’s working on in his research group including search engines for lifelogging as well as privacy and storage issues.
“How can I define what makes me happy and what makes me sad, what is good for me?”
In 2012 Benjamin Bolland was finishing up his undergraduate degree and working on a new start-up. He found that his moods were constantly changing and wondered if there was something he could do to make sense of them as they moved “up and down.” He began tracking his mood with an simple self-designed Google Form. Each day at 11AM and 6PM he reported his mood on a 1-10 scale and wrote a quick descriptive note. After 1.5 years of doing this nearly every day he realized that he didn’t really know what was making him happy or sad so he decided to update his Google Form to include a variety of different categories that he thought might affect his mood including physical health, sport participation, and many others. In this talk, presented at the Berlin QS meetup group, Benjamin describes his process and how he’s used this mood tracking process to be more reflective and mindful during his daily life.
“There was nothing in my life pushing me to to have these more intimate relationships, the few people I actually care about.”
When Akshay Patil was putting together the guest list for his wedding he realized that it had been a long time since he’d spoken with some of the the people he was inviting. Even with his good friends, he surprised by his lack of communication, his inability to stay connected. As anyone faced with this realization he decided to try and change, but the realities of life quickly crept back and as they say, old habits die hard. When he left his last job and began looking for projects to work on, this troubling area of his life crept back to the fore. Maybe there was something he could do better track and change his communication and relationships. Using his development skills, and the ability to gather data from his Android phone, he decided to build a system that helped him stay in touch with the people that mattered most to him. In this talk, presented at the New York QS meetup group, Akshay talks about what’s he’s learned from using this app, including when it fails.
Siva Raj was interested in lowering his blood pressure. With a family history of cardiovascular disease and heart attacks he was worried about slightly elevated blood pressure (pre-hypertension). As someone engaged with understanding and building fitness applications he thought he would be able to lower his blood pressure by staying on track with a regular exercise program that focused on cycling. Interestingly his blood pressure measurement didn’t respond to his constant exercise or weight loss. After reading more research literature about the link between fitness and cardiovascular health Siva decided to change his training to improve his fitness. He decided to incorporate a increased intensity into his routine. After a short period of time he had increases in this fitness and was able to observe the reduction in blood pressure he was looking for. In the video below, filmed at the Boston QS meetup group, Siva explains his methods and talks about how he was able to track his body’s response to different fitness routines.
Jamie Williams found himself with almost two years of self-tracking data including physical activity, blood pressure, and weight. Because of his interest in data visualization and coding he decided to learn how to access it the data and work on visualizing and understanding some of the trends and patterns. In this talk, presented at the QS St. Louis meetup group, he takes a deep dive into his activity and step data as well as his blood pressure data to learn about himself and what affects his behavior and associated data.
What Did Jamie Do?
Out of pure interest in seeing what the data would reveal, Jamie utilized a combination of devices to track his physical activity, blood pressure, heart rate, weight, numbers of drinks, and automobile travel. He then went on to explore ways in which he could pull down, integrate, visualize, and ultimately make sense of what he collected.
How Did He Do It?
In order to obtain his data on a minute-level resolution, Jamie had to email FitBit for a specialized use of their API. He then employed Mathematica to develop a number of (beautiful) visualizations of his activity – along with other key moments in his life (moving to St. Louis, changing job location, preparing for a Half Marathon, etc.). Jamie was able to compare his data not only to his peers through FitBit, but also to others of his demographic in the U.S. using the publicily available NHANES data set.
What Did He Learn?
Through Jamie’s Quantified Self collection and analysis efforts, he learned a lot not only about the patterns and changes in his activity, but why they were the case. He also presented great feedback about one’s mindset when comparing to peers vs. the general population.
Withing Blood Pressure Cuff
Thank you to QS St. Louis organizer, William Dahl, and Jamie for the original posting of this talk!
Sue Lueder had a mystery stomach ailment that started after a vacation to Spain in 2011. When she returned from her trip she was beset by consistent and frequent burping attacks. After visiting her physician and receiving a diagnosis for heart burn, which she didn’t trust. she began to track her attacks and her diet. In this talk, presented at our 2013 Global Conference, Sue how she tracked he symptoms and used the data to make sense of this mystery food allergy.
What Did She Do?
Sue tracked her diet and the frequency and severity of her attacks.
How Did She Do It?
Sue was able to explore the data she was entering in to her self-designed spreadsheet tracking system. She used a few of the analytical tools and visualizations built into Excel to explore her data.
What Did She Learn?
Her analysis was able to pinpoint that dairy was probably the main culprit responsible for her attacks. Sue found out that she was able to improve her “good” days from 32% to 51% of the days she was tracking when she reduce dairy in her diet. When she experimented with adding dairy her findings were confirmed.
In 2009 Tim Ngwena switched on Last.fm and he’s been running in across all his devices ever since. Earlier this year he decided to take a deep dive into his listening data to see what he could learn.
I realized that I was listening to the same old thing and I began to think about changing what I was listening to. But how can I change? Where can I start? I also wanted to learn something about my music, what I was listening to and who was behind the sounds. I decided to focus on music because it was doable.
In this talk, presented at the London QS meetup group, Tim explains how he was able to make sense of almost five years of data and learn more about himself and his listening habits.
What Did Tim Do?
Tim explored his music data along side additional information such as location data from Moves to learn about his musical tastes, listening habits, and explore new visualization and data analysis techniques.
How Did He Do It?
Tim exported his data, used the Last.fm API and some data cleaning and organizational tools to create a simplified and extensive database of his music listening history and associated data. He then visualized that data using Tableau.
What Did He Learn?
Tim learned a lot about himself and what the music he listens to says about him. He describes a few of the most interesting below,
Basically 80% of my listening comes form 10% of the artists that I have in my library.
I’ve listened to Erykah Badu for over a week (7.2 days). It led me to ask what is she saying to me?
Monday is my jam time. I’m listening from the morning into the evening.
I listen to music mostly when I’m walking.
Tim also learned a lot through the process of designing and creating his data visualization. The visualization, which you can explore here, made him think about being able to see the big picture when he has so much linked data.
I think context is important and you need to see all that information in one place and the tools I’m using allows me to do this.
Shawn Dimantha is always looking for easier ways to track his health. He uses a variety of self-tracking tools, but a few months ago he became interested in exploring what he could do given his engineering and health IT background. He was inspired by immersion, an MIT-developed email analysis tool, which helped him understand who he was communicating with, and by Wolfram Alpha’s Facebook analysis tool. Focusing on Facebook and the wealth of image-based data in his profile he asked himself if images could be a window into his health. After reading a research paper on the use of images to predict body mass index he decided to see what he could learn my implementing a similar procedure on his own images.
What Did Shawn Do?
I used photos from my Facebook account to track my health, the reason I did this because I wanted to see how a simple heuristic I used for tracking my health daily could be implemented in the online world given the huge amount of photos that are and have been shared on a daily basis. I notice when I gain or lose weight, am stressed or relaxed from my seeing my face in my mirror. I was partly inspired by the self-photo collages presented on YouTube.
How Did He Do it?
I selected photos of my face from my Facebook account, cropped out my face and used some software and manual tagging to measure the ratio of different fiducial points on my face (eye-eye length, and cheek to cheek length) over time to help serve as a proxy for my health.
What Did He Learn
Facial image data needs to be cleaned and carefully selected. Face shapes are unique and need to be treated as such. Data that is not present is often more telling than what is present. Life events effect my weight and should be put into context; however causation is harder to determine than correlation. By being more conscious of my score and I can change my behavior before things get off track.
Right now I’m turning this into a product at Enfluence.io where I’m focused on using it to help with preventive health.
Facebook (my own images)
Python / OpenCV
Slides from Shawn’s talk are available here.
One interesting aspect of personal data is how it can reveal what is unique about you. Nowhere is this more true than with genetic information coming from DNA testing kits. However, people are still at an early stage on how they apply that information to their lives. Ralph Pethica, who has a PhD in genetics, was interested in what his DNA could tell him about how to train more effectively. His findings were presented as an ignite talk at the 2014 QS Europe Conference.
What did Ralph do?
Ralph loves to surf. When it is the off-season, he trains so that his body will be in good condition for when the warm weather rolls back around. He used genetic research to inform how he designed his training plans.
How did Ralph do it?
Ralph used a 23andMe kit to find out his genetic profile. He researched those genes that have been found to have an impact on fitness to see his body should respond to exercise. For example, did he possess genes that gave him an advantage in building muscle with resistance training? He then modified his training routines to take advantage of this information and monitored his results (using the Polar watch and a Withings scale) to see whether his assumptions held up.
What did Ralph learn?
Ralph found out that he has genetic disadvantages when it came to strength training. This told him that progress in this area depended more on his lifestyle. In particular, he found that eating immediately after working out was important.
When it came to cardio exercise, he had a number of genetic advantages. The unexpected downside to this is that his body adapts quickly to any training regimen, resulting in a plateau. To get around this, he varied his training plan and monitored his results. On one day, he would cycle at a steady rate, while the next, he would use high-intensity intervals. His body seemed to respond to the varied training plan and he hit fewer plateaus. Without knowing which genes he possessed, and reading current research on those genes, it is unlikely that he would have discovered these effective customizations to his training plan.
Ralph has taken what he’s learned and built a tool called Genetrainer to help people use their genetic information to inform their fitness plains. You can check it out here.
Tools: Genetrainer, 23andMe, Polar RCX5, Withings Smart Body Analyzer