Last May, Ultimaker supported the Weill Cornell Medicine Clinical and Translational Science Center (CTSC) 2nd Annual Health Innovation Hackathon, equipped with three Ultimaker 3 printers, spools of PLA, TPU, PVA, and Breakaway support, and buckets of 3D printing knowledge.
A Hackathon is an event in which teams made up of individuals with different expertise concentrate on a specific problem to create something usable within a specific time frame. The CTSC Health Innovation Hackathon was the brainchild of medical student Birra Taha, founder and student leader of Tech-in-Medicine at Weill Cornell Medical College, who approached My Linh H. Nguyen-Novotny, the Assistant Director of CTSC two years ago to propose such an event. The CTSC 2nd Annual Hackathon included students, staff, and faculty across different disciplines from Hunter College, Cornell Tech, Cornell Ithaca, Weill Cornell, Memorial Sloan Kettering, Hospital for Special Surgery, Rockefeller, and The Cooper Union.
The purpose of the event was to stimulate disruptive thinking, emphasize innovation, and leverage new and emerging technologies to solve medical research questions that might ultimately benefit human health. Dr. Julianne Imperato-McGinley pointed out in her talk at the kickoff event that the Hackathon had the potential to increase the “free flow of information, and to break down the silos that exist between disciplines, between institutions, between companies.” Dr. Imperato-McGinley reminded participants before launching into problem-solving to think about what Albert Einstein said, “We can not solve the problems with the same thinking that created them in the first place.”
While the actual working session of the hackathon occurred over a weekend, participants met a week before the event to participate in a SCRUM exercise, hear about the challenges, receive team assignments, and find out which challenge they would be asked to meet.
Alexandra Kramer, Cornell Alumna and Manager of Digital Experience Design at Johnson & Johnson, led the SCRUM workshop. As a way to introduce the participants to this framework, she challenged the entire group to pass as many balls as they could around the room, with the caveat that each ball had to touch everyone in the room, could not be passed to one’s neighbor, and had to make its way back to the starting point. She gave the participants a time limit of two minutes to pass as many around as possible. As a group, the participants had to estimate how many balls would successfully be passed through the entire circuit. After each round, participants could discuss strategy, and had to predict the outcome of the lates one. With each iteration, the system functioned better and better and the group’s ability to predict the actual outcome improved.
This team event was a great way to quickly demonstrate the following concepts:
- Self-organizing teams
- Flow of work from Product Owner -> Scrum Master-> Team
- Estimation accuracy improves over time
- Velocity becomes more consistent over time
- Continuous improvement
According to Kramer, SCRUM provides a framework “within which people can address complex adaptive problems, while productively and creatively delivering products of the highest possible value.” This was a perfect exercise and lesson for those about to jump into a hackathon.
After the SCRUM activity, the facilitators presented the problems for the hackathon. Questions were held to the end. After one of the organizers, Milestone-Lab’s Oded Shorer, provided a brief introduction to Design Thinking and a breakdown of how the Hackathon would be organized. Each team had to have a point of contact responsible for reporting and communicating the team’s needs to the organizers, a treasurer responsible for the team’s wishlists and receipts, and a person to document the team’s progress. The teams were assigned and the challenges were defined.
After a week of research and communicating using the UNI Communication Platform, an online application, two of the original ten challenges were taken off the table and some participants shifted to different teams.
By the following Friday, these were the working challenges:
How might we develop a device that can help someone living alone who is experiencing choking?
The Sensible Hand
How might we create a low-cost functional prototype for a myoelectric hand prosthesis to maximize independence in activities of daily living (ADL) for individuals with upper extremity amputations. (As well as minimizing the potential for prosthesis rejection.)
How might we objectively, reliably, and immediately determine the mental state, between sleep and wake, of an individual in real time?
How might we develop a cost-effective endoscopic tool to safely remove tissue debris resulting from pancreatic necrosis, gastroparesis, foreign objects, and clots in the stomach to improve health outcomes for patients with digestive diseases?
How might we apply subvocal speech recognition for silent communication to and via cell phones?
How might we design an insertion device with a wide-angle lens that will safely and securely fit a CMOS digital camera (that is approximately 5 mm or less in diameter and approximately 20-25 mm in length) for the purpose of producing high-resolution images from the ear, nose, and mouth regions with a field of vision (FOV) of at least 90 degrees?
How might we encourage breastfeeding in refugee camps?
How might we measure or find obstructions in Ventriculo-Peritoneal shunts?
After the kickoff, participants broke out into their teams and sent representatives to the documentation and treasurer workshops.
In the Documentation workshop, Oded asked those in the room if they had previously participated in hackathons and asked them what happened to their projects. A common thread was that there was a prototype but that it did not go beyond the event. As Oded said, “It’s a great experience, people come up with amazing ideas, and then it’s over. It’s over and nobody has time to follow up on this. There is no ‘the day after’ plan.” He lamented that “people working so hard, investing their time to come up with these amazing ideas, and then that’s it, it fades. And it’s a shame. You should not let that go.” Oded wanted participants at this hackathon to follow a more sustainable structure.
The setup was as follows: every team selects one person to be responsible for documenting the project. All evidence of progress (sketches, code, CAD files, photos, videos, and notes) would be kept in one place, and accessible to all team members. Each team has a folder in Google Drive, and Oded provided each team with a Documentation template to make keeping track of the project as simple and painless as possible. The idea was that these folders should serve the teams, but they also provided a starting point for others who would like to continue to develop these ideas after the hackathon. Because, as Oded explained, if you document properly, other people can pick up the project.
In the treasurer meeting, the team treasurers were instructed how to handle purchasing supplies and submitting for reimbursement. Each team was allowed up to $100 to purchase necessary supplies.
The last organized activity on Friday afternoon was the first progress, needs, and goals (PiNG) session of the weekend. During this activity, the point of contact (POC) from each team reported to the organizers and mentors their team’s progress, needs, and goals. Feedback from the mentors and organizers was provided to each team on Friday afternoon and throughout the weekend.
Everyone broke for the evening at 7pm, and returned the next morning at 8:30am. Throughout the day there were PiNG sessions, material runs, and concurrent workshops covering scanning and 3D printing.
During the 3D printing session that I led, I talked about different fabrication materials and how to prepare models for 3D printing with Ultimaker Cura. I talked about wall thickness, and how a higher value for wall thickness might create a sturdier model (and decrease the chance of leaks), but how a lower value might significantly decrease the print time and filament costs. I then introduced the participants to the alternate extra Walls setting (found under shell). This setting lets one add strength to the outer walls without putting too much burden on print speed. When this setting is enabled, Ultimaker Cura will add one extra interior shell to every other layer.
I covered a few other settings with the event in mind. To keep the prototypes progress on track, I spoke about the horizontal expansion setting. A positive horizontal expansion value will add to the dimensions of a model. Participants could use a positive value when their printed models were smaller than they expected, something that might occur because of their material selection. I also spoke about different infill options, as well as ironing. At the end of the sessions I handed out two versions of the Everyday 3D Printing checklist (http://ultim.kr/ChecklistAdvanced, http://ultim.kr/ChecklistSimple) to everyone new to 3D printing.
Since time is so limited during hackathons, I also talked about strategies for printing faster, like using larger nozzle diameters and performing feature tests, rather than printing their entire part over and over.
To help with the documentation process, I encouraged participants to save each Ultimaker Cura file as a project. This step would allow each team to keep track of settings, materials, and part orientations with the least amount of effort on their part. This also helped with troubleshooting when participants encountered unexpected results.
Sunday continued the activities of Saturday, as the projects continued to advance: PiNG sessions, material runs, and concurrent workshops. The 3D printers were moved to the CTSC 3D Printing Lab between 3pm and 5pm so that prints could be completed after the hackathon ended at 5pm. Participants continued to have access to these printers until the Final showcase on June 1st.
On Friday June 1st, the teams, organizers, mentors, and judges reconvened for the final showcase. Teams delivered their final presentations, judges deliberated, and the awards were to be announced.
In the time between the Sunday of the hackathon, and the final showcase, teams solidified their ideas, selected final project names, continued to take advantage of the 3D printers, and prepared their presentations.
Pancreosound became the more descriptive Endoscoop, and their project became a revolutionary device used to remove dead tissue found inside the pancreas. eNT Camera became Digital Doctor’s Bag. What started out as a single tool became several, collected together to provide cost effective and user-friendly aids for telemedicine.
Anthony M. C. Brown, PhD, Director of Medical Student Research, and Assoc. Professor of Cell & Developmental Biology at Weill Cornell Medicine, announced the winners. The judges decided to split the grand prize of $10K into two, with a 6 to 4 ratio.
Third place went to Endoscoop. Second went to Zleep. First went to Milky Way. Sensible Hand received the first grand prize of $4,000. And the grand grand prize winner with $6,000 was MyophonX.
The Sensible Hand project, the winner of the smaller grand prize, revolved around a modified version of an open source 3D-printed hand prosthesis called the K1, originally created by Evan Kuester and made available via the e-Nable community. The modifications that the Sensible Hand team made to the device included adding an infrared proximity sensor embedded in the middle finger with an LED indicator to cue the user about how close they are to a desired object. The team added haptic technology to the prosthesis to offer the user sensory feedback regarding the strength of their grip. Also added to the hand were boreholes and a spring loaded turret universal tool adapter to accommodate any position necessary for the attachment of tools. This modification was included to provide the user with greater independence in daily activities.
The MyophonX team create a project that enabled those who cannot communicate with sound to have a voice executed via muscle signals. They chose to work with a laryngectomy patient at Weill Cornell who provided input towards the direction the team pursued. The laryngectomy patient suggested that he found communicating through mouthing out the words more natural. They tailored their device to this patient’s specific needs. They scanned his head and created a 3D printed mask that covered one side of his jaw and was equipped with eight electrodes/sensors. They measured eight muscle signals in parallel and also established wireless communication from the device to a computer. The goal was to use some AI to differentiate the muscle signals and to differentiate between those muscles used to produce the names Ed and Ted, the latter being the name of the patient. Their ultimate goal was to have the ability to customize their device for individual patients.
The event was a success. Ideas were created and relationships were formed. We are sure to see a 3rd Annual CTSC Health Innovation Hackathon. And the days spent collaborating inspired the organizers, mentors, and participants to think about how they might find other opportunities to work together — and how Design Thinking and the diverse skills that were brought to the table from the various institutions could be combined in deeper and longer engagements.