Technical Excellence Award
Proctor & Gamble
In September I received the 2017 P&G 1st Place Technical Excellence Award and scholarship, which goes to one senior in the College of Engineering.
As finalists for the award, we were given 5 days to create a 20 minute presentation on a technical achievement to give to the judges. More than anything, I was just so excited to get to tell them about what I've done- a project that I had ownership over, spent months working on, and am so proud to have created. (Check out the Cowscope Page to find out more about the project I presented!)
This experience solidifies three things that have become important to me as I finish my incredible four years at Cornell. The first two are pieces of advice that my mentor at the wearable tech startup I worked at over the summer gave to me. And I hope to keep these in mind in all aspects of my life and career:
The first is to be excellent.
Whatever you do, even if it’s not an excellent job or excellent circumstances, be excellent at it.
The second is that success depends not only on what you create, but how you share it. I feel unbelievably lucky that I am so proud and excited to share the work that I’ve done. I truly believe in the creativity of my solutions and the potential they have to change the world. And when I can walk out of the presentation room feeling on top of world, I know I’m on the right path.
Successful people have good ideas and are good at sharing them.
And lastly, over the summer when the CEO of my company complimented me on my confidence, I was completely blown away. But since then I’ve decided that if that’s how someone so amazing and inspiring to me can see me, then that’s who I’ll become. So today I’m owning my confidence- in the work I produce, in the journey to get here, and in everything yet to come.
Presentation Slides
Transcript of my presentation
[COW] My name is Sachi and today I’m going to tell you about how I created a better detection system for a costly dairy cow disease.
[iGEM] For the last two and a half years, I have been lucky enough to be part of Cornell’s International Genetically Engineered Machines Project team. This is a team entirely run and organized by 25 undergraduates like myself, and we are among over 300 teams from all over the world. Each competition season we pick a local and/or global problem that we think we can solve in a novel way using synthetic biology - it’s completely open ended. We spend the spring semester researching and proposing dozens of problems, ideas, and solutions, taking into account relevance, impact, and feasibility. And it can be anything - health, environmental, manufacturing, even food products.
But it’s not just a synthetic biology solution. There are all sorts of other components to the project that we add to create as holistic and applicable a solution as possible, including hardware, devices, apps, outreach, business plan and project website. We spend the entire summer then formulating, creating, and testing our solution to bring to competition in the fall. In the two seasons I was on the team, we won a silver medal, a gold medal, Best Environmental Project, Best Supporting Entrepreneurship, Best Applied Design, and were nominated for Best Integrated Human Practices.
Last year, I pitched the original idea that became our 2016 iGEM project.
[Legendairy] Bovine mastitis is an infection of cows udders caused by several types of common bacteria. An infection can be chronic or short term. It causes irreversible udder damage to the cows. [COW]And it causes milk contamination where somatic cells are shed from the cows’ udders and wind up in the milk. All milk containing high concentrations of these cells must be thrown away. Current treatments rely on long-term antibiotics that takes 6 weeks if it’s successful. Not only does this treatment have the drawback of antibiotic resistance, but all the antibiotics end up in the milk as well. That milk is also thrown away.
Okay, so why is this such a big problem?
[30%]Mastitis happens in every dairy farm- every single one. And at a rate of 30% a year. 30 out of every 100 cows. And here’s the kicker - one case of mastitis costs the farm 400 Dollars - that’s the cost of six weeks of treatment, the cost of needing to continue to milk the cow during that time, and the lost money from throwing away six weeks worth of milk. That’s a 2 billion dollar net loss in the US dairy industry alone.
This is a problem we decided we could help eliminate.
[SO WHO AM I?] My name is Sachiye Katrina Koide, and in 2016 I was the Computer Science and Electrical Engineering Product Development Lead. I am neither a CS or ECE major; I am a biological engineer minoring in business- but boy can I learn and create, and boy do I love solving problems. I handpicked my subteam of 10 extremely talented engineers, and for 5 months, we took it upon ourselves to redesign the cornell iGEM engineering website, create our entire project wiki, but most importantly, create something that would revolutionize bovine mastitis care - whether that meant prevention, detection, or treatment. [Prevention, Detection, Treatment]
As a full team we actually accomplished all three, but I specialized in detection. [Detection].
[Point A] The first step to getting from point A to point B is finding point A in the first place. This was the first step in the story of our process.
[Design Thinking Workshop] My goals were always very user based. I wanted to create the best possible solution for our specific market. And I wanted to do this right.
So I started our research process with 3 steps. The first was a design thinking workshop.
I taught myself stanford design thinking method and then ran a workshop for my team in which we went through every step of the UX process in an hour- starting from interviewing each other to identify a problem, getting to the root of the thoughts and feelings of the customer, brainstorming as many ideas as we possibly could, conducting more interviews, and creating prototypes out of post it notes and popsicle sticks. This mini lesson set up the structure of the entire summer of continual research and reiteration.
The second research process was web based. We read any information we could find, from published scientific papers to Dairy Farmer magazines, to informational pamphlets about mastitis care. And then we conducted a competitive analysis by looking at all the products already out there to treat and detect the disease and identified what they did well and could improve on.
[Web Research and Interviews] And finally, we reached out to dozens of farms around Ithaca and interviewed as many dairy farmers as we could- big farms, small farms, family run, Cornell owned, as well as researches at the vet school.
[Farms we Interviewed] And after all this, what did we learn? We spent an afternoon compiling all the information we had gathered about the disease and organizing it into a problem statement and brainstorming all the specs that would be required of a useful and effective product
We learned 1) That yes, this a huge problem that does not yet have an effective solution. Because the milk doesn’t get tested for high somatic cell counts until after all of it has been collected into the 3000 gallon milk tank, ne undetected infection could mean the entire tank gets poured out onto the grass. I saw one farmer cry when that happened.
[Need, Say, Feel, Think, Do] We stuck to this problem statement: Farmers need to detect disease sooner because of money and time. It needs to be easy to use, affordable, easy to manufacture, durable, and mobil,e to name just a few of the requirements we came up with.
[Specs] From our specs we created quantifiable measures of success vs. failure. If it’s easy to use, it needs to be between 3 and 5 steps. If it allows for rapid detection, it needs to take fewer than 12 hours.
[Creative Brainstorming = Ideas] And then the real fun began, in which we came up with as many ideas as we could to meet the needs of dairy farmers across the country. Then with 45 or so ideas to work with, we grouped them together, combined them, and voted on our favorites.
And this was a great starting point, but our ideas were still half formed, and when we brought some of them to farmers and researchers, they weren’t too enthused.
It wasn’t until I talked to Dr. Daryl Nydam, Director of Quality Milk Production Services at the Cornell University College of Veterinary Medicine, that I came up with an idea that I knew had huge potential.
[More Interviews = Key Information] According to the farmers and researchers we met with, the only way for farmers to get a quantitative SCC for each of their cows is to send the milk to facilities where lab technicians can analyze the samples. This can take between 18 hours and several days
In New York State, where there is a huge emphasis on agricultural research, many farms are fortunate enough to only be an hour away from the nearest facility. However, farms in more rural areas as well as in the west and mid-west often live much further away. Because it can take very long to package and transport samples and wait for the results, farms can only check the SCC of their cows once a month. Since it’s inefficient to check the SCC for a few individual cows, this is often neglected. Many farms may instead opt to have a technician come to their farm once a month to check the SCC of each cow in their herd. Overall, the process is time consuming, infrequent, and expensive [2].
[A New Idea] One of the ways lab technicians get the SCC, I learned is through direct microscopy using a dark blue stain on the nuclear mass and a hemocytometer [3]. This is a paperI found detailing the process. They recognize the nuclear masses in the hemocytometer grid and manually count them to get the average [3]. It’s that simple!
When I learned this I immediately thought
[How Do I Bring This Process Cowside?] How do I make an inexpensive customized cow-side microscope that lets the farmers take a somatic cell count themselves?
[Prototypes] Plywood and a smartphone camera! To start. And an app that could use photo recognition to count the cells on the hemocytometer the same way they do it in the lab. This was the very first prototype we built. (not a novel idea, we found this online)
[Cowscope] So at the next meeting, I proposed this idea to my team, and later we did a preliminary design review.
[The Build]
[The Final Iteration] And it’s so interesting to me that after all that information collecting and ideating on post-it notes, the final design came together so quickly.
We created an inexpensive, 3D printable, customized cow-side microscope that utilizes a smartphone camera to take the somatic cell count within our app. We went through the same process for our smartphone application as we did for the hardware in terms of ideating, interviewing, and refining. I was in charge of the final design of our features, but did not code the app itself. But I created the Cowscope hardware.
[Scope] The microscope uses a simple light, 30 mm lens, and smartphone camera to take pictures within the app of samples placed in a hemocytometer. It has adjustable plexiglass surface to put the sample into focus, a platform to place your phone securely, and storage area for the haemocytometers and dye. This allows a 375x magnification of a sample, which is enough to visualize on a cellular level. This model I created in Autodesk fusion, and it can be 3D printed or machined for under $15. It’s designed to be durable, simple to use, easily cleanable, and sturdy. All things farmers need cowside.
[The App] In the somatic cell counter section of the Cowscope app, farmers can image their samples and save their microscope photos in one place for fast, efficient somatic cell counting. The idea was to have a photo recognition feature which uses the same parameters the lab technicians use to recognize and count the stained somatic cell nuclear masses within the top and left borders of the hemocytometer grid. [3]
The Somatic Cell Counter not only allows farmers to take and save their cell images, but photo-recognition software of the stained nuclei allows the app to take the cell count for the farmer, eliminating the need for the time-consuming individual counting. It then calculates the SCC concentration, and adds the information to the Cow Data section that keeps track of individual cow health and history.
[DATA] The Cow Data section shows farmers trends that may be overlooked in the day-to-day care of farms. It shows graphs such as infection by cow and milk production so that farmers can identify clusters of cows that have recurring infections, and see how productive each cow is. We intend for the Cow Data section to highlight trends such as budding mastitis outbreaks, so that farmers can take preventive measures as early as possible. This section was created at the enthusiasm of farmers we talked to about the idea, and we will continue expanding the scope of the Cow Data application in order to create as holistic a picture of the disease on the farm as possible. Soon we may be able to compare other local farm data to identify regions where outbreaks are more common and may begin affecting other farms. Our goal is to create a local, regional, and even national network to identify trends in disease and create the foundation for more regionally targeted and efficient cures.
[COST CALCULATOR] The Cost Calculator aims to give farmers some quick metrics to weigh the potential financial costs of an infected cow. The Keep page allows farmers to calculate the cost of keeping the cow. As a preliminary simplification we have included factors such as treatment time, milk production, and milk cost, but we acknowledge that there are more factors that determine the cost of taking care of an infected cow. The Discard page acts like the Keep page, except it takes in factors that determine the cost of “letting go of” or culling the cow. Finally, after inputting the data, the app will highlight the less costly option in green, which farmers can use in their decision for treating an infected cow.
[Cowscope] Overall, these three features of our detection solution aim to help farmers stop mastitis in its tracks before costing them valuable money and time.
[Thank you]Thank you for listening, and feel free to ask me about the the treatment and prevention system we designed as well, and check out the wiki as well! Because I was in charge of that, and it’s a really cool project!
