We are three seniors at Washington University in St. Louis majoring in Biomedical Engineering. This project will serve as our capstone design project for BME401.
Researchers working to minimize developmental issues in premature babies caused by stress, resulting from an isolated, noisy environment, are in need of a small, lightweight device to be used in a study that measures limb movements indicative of stress to determine if listening to recordings of a mother’s voice reduces stress and improves the medical outcomes of the child.
Our client requested a device that will measure the movements of incubated premature infants. This device will be used in a study that attempts to reduce stress levels by playing recordings of the infant’s mother’s voice. When the infants feel stress, they move more and burn calories, which is dangerous for premature infants of that size. The physiological impact of stress can also lead to developmental issues as the baby grows. Existing solutions are too large to be worn by a premature infant or not accurate enough for a scientific study. The device must be safe for a premature baby without hindering its movements, while being sensitive enough to accurately detect those movements. In April, we will be delivering a device prototype with accompanying software, circuit diagrams, and mechanical drawings to Dr. Bozzette.
Weight | Any component physically attached to the baby must weigh less than 1oz (28 g) |
Compact | The shape and volume of the device does not impede the movements of the baby |
Sanitizable | Can be sprayed with disinfectant (70% ethanol) or washed or washed in a washing machine if fabric-based |
Reusable and Consistant | Will work the same way every time for a range of movements and infant sizes over multiple trials |
Durable | Can be used for 30-40 trials |
Safe | Exterior of the device is smooth, durable, and non-toxic |
Stress | Does not add to the patient's stress |
Interference | Does not interfere with hardware/software of the other medical devices in the room |
Safe | Exterior of the device is smooth, durable, and non-toxic |
Easily Operable | Can operate independently once turned on; fast and easy for a nurse to set up |
Fluid Resistant | Device is resistant to dripping fluids; classified as IPX2 (Ingress Protection Code) |
Distance | Records each time the infant’s limb changes direction and moves more than 2 cm |
Duration | Records movements that last longer than 500ms |
Acceleration | Records accelerations between 2.5 cm/s^2 and 400 cm/s^2 |
Sampling Frequency | Minimum 50 Hz (20 ms/sample) |
Precision | The device should count movements with at least 95% accuracy when compared to a manual observer. |
Continuously Operable | 3 hours of continuous use |
Accessibility of Data | Data is downloadable |
Range | Website can collect data from 5 meters away and with a plastic barrier between the device and the computer |
Budget | Maximum materials cost of $250 |
Week | Report |
---|---|
3 (9/13 - 9/17) |
We have agreed to work with Dr. Bozzette on a device for measuring limb movements in premature infants. We completed our project scope, including a need statement, scope, and design specifications. Finally, we created our web page. |
4 (9/20 - 9/24) |
We met with Dr. Bozzette to discuss design specifications. She sent us some information to influence our design, including photos of premature babies and a growth chart. We are re-working the project scope and beginning to work on the preliminary report. |
5 (9/27 - 10/1) |
This week, we received and implemented feedback on our project scope, including updating the need statement to clarify the goal of the project, as well as a complete overhaul of the design specifications to be less solution-oriented. We used our usual meeting time to outline responsibilities for completing the preliminary report. We also researched background information and existing solutions to complete the report. Lastly, we created a tentative timeline for project development. |
6 (10/4 - 10/8) |
On Monday 10/4, Gillian presented the preliminary report to the class. We met the following day to go over the project timeline in preparation for the progress report, and we made sure LabArchive was up to date and organized in preparation for the upcoming notebook check. |
7 (10/11 - 10/15) |
We did not meet this week due to fall break. In the meantime, Gillian completed the website, and we started individually brainstorming potential solutions in preparation for a group brainstorming session. |
8 (10/18 - 10/22) |
At our group meeting, we reviewed feedback from the preliminary report. We continued to brainstorm potential solutions and set a timeline to ensure we are working towards completing the progress report on time. We agreed to finish brainstorming potential solutions by next week. |
9 (10/25 - 10/29) |
This week, we discussed potential solutions. We further discussed our design specifications and considered how the potential solutions aligned with them in preparation for making our Pugh chart. |
10 (11/1 - 11/5) |
This week, we had two team meetings. The first was dedicated to going over feedback of our LabArchive notebook and subsequently adding to and organizing the LabArchive notebook. In the second meeting, we wrote down a list of potential ideas and reviewed the schedule for the rest of the semester to ensure we are on track. |
11 (11/8 - 11/12) |
This week, we first met to further brainstorm potential solutions. We discussed the pros and cons of each, and how well they fit the design specs. We met another time to determine the weight criteria for each design spec to start making the Pugh Chart. |
12 (11/15 - 11/19) |
This week, we met with Professor Klaesner to discuss how to rate potential solutions regarding each design spec and how to weight each design spec. Following this meeting, we continued rating potential solutions to build our Pugh chart. We also discussed our schedule for the next week to make sure enough work is completed over Thanksgiving, and we started planning our Progress Report. Lastly, we emailed with our client, Dr. Bozette, to better understand the environment the premature infants are in so that we can determine which potential solutions will be best. |
13 (11/22 - 11/26) |
We met as a group to finalize our Pugh chart and emailed Professor Klaesner to ask for feedback. We each individually worked on the Progress Report and delegated sections for each member to work on. We plan to meet tomorrow to check-in on report progress. |
14 (11/29 - 12/3) |
This week, we met several times to work on the Progress Report. We finalized the Pugh Chart and list of potential solutions, and did market research to fully explore and understand each design. While finalizing the Pugh Chart, we continued to discuss, clarify, and update our design specifications. In addition, we researched and discussed our chosen solution. We planned the budget and developed a high-level schematic to start the design process. The completed report was submitted to Canvas on time. |
16 (1/15 - 1/21) |
This week, we had our first meeting of the semester via Zoom to touch base after break. We decided on a preliminary meeting schedule for the semester. We also ordered parts and submitted for reimbursement via Workday. Finally, we read through and discussed feedback from our progress report and design notebook. |
17 (1/22 - 1/28) |
This week, we met on Zoom once again to do the "Yellow Stickies" activity. This helped us plan our design process as we determined each step that had to be done and the order they should be completed. We set a loose timeline for the rest of the semester, with the goal of finishing the prototype in March with V&V testing to follow. Lastly, we ended our meeting by assigning tasks to each member of the group; while we plan to work together, whoever is assigned to a task is in charge of the planning needed to get it done. |
18 (1/29 - 2/4) |
This week, we met in Professor Widder's lab to unpack our materials and start to learn how they work. We did this by watching online tutorials and experimenting with the equipment hands-on. We also weighed them to ensure that the total weight would not be over 1 oz/28 grams (from our design specifications). In the lab, we practiced soldering so that we could solder our equipment without making mistakes. We planned to meet again to start building, but unfortunately we were unable to meet due to the weather. |
19 (2/5 - 2/11) |
This week, we made our first prototype. We were able to connect the accelerometer to an Arduino, and a laptop read the output. While it is still too large to fit on a bracelet for a premature infant, this first prototype allowed us to work on the software component of the product. In addition, we improved our understanding of how the accelerometer works, so we can continue to refine the hardware design. This week, our software work mainly focused on counting and defining movements instead of the user interface. We also determined that a calibration period will be necessary to obtain a baseline value of the accelerometer at rest, which will then be subtracted from the output when the accelerometer is in motion, so we implemented this code as well. |
20 (2/12 - 2/18) |
TThis week, we worked on making our prototype smaller so that it could fit on a bracelet for a premature infant. We did this by connecting our accelerometer to the microcontroller instead of the breadboard, but found that the microcontroller did not give an output in real-time. We then tried using a mini Arduino, which seemed to work better, as we got a live read of the accelerometer. Further testing is necessary to make sure the output is accurate. |
21 (2/20 - 2/25) |
This week, we met on Sunday to try to work the power source and battery into the circuit, replacing the need for the laptop to be connected. While we were able to get the power source/battery to work, we were not yet able to integrate it with the circuit, as we still need the laptop to run code to the Arduino. We then met on Wednesday to discuss how we plan to go about completing the V&V report. Carley began doing FDA research, Julia worked on CAD and researched ways to make the circuit wireless, while Gillian worked on building the user interface. We also started working with conductive thread, fabric, and foam to try building the physical bracelet. Moving forward, we're planning to use next week to finalize our pre-V&V prototype and complete the V&V report. The plan for this prototype is to have just the accelerometer on the baby's wrist, and connect with with very long wires to the rest of the circuit which will be on a breadboard that sits beside the baby. |
22 (2/26 - 3/4) |
This week, we made progress on our prototype. We lengthened the wires connecting the accelerometer to the rest of the hardware and made a wristband for the accelerometer so that the amount of circuitry connected to the baby is minimized, thus reducing the weight. We also 3D printed a case for the accelerometer to protect the circuitry from fluids. In addition to the progress on the hardware end, we further developed the software needed to calibrate the device and count movements. Lastly, we worked on our Verification and Validation plan and report. |
23 (3/5 - 3/11) |
This week, we focused on planning ahead for the rest of the semester. Our biggest priority is developing the software needed for the device to function, so Gillian will take the prototype with her over spring break to work on it. We also brainstormed ways to improve our device; we decided to reprint the case housing the accelerometer to make it smaller, more lightweight, and circular. We will also attempt to make the device wireless using the transmitter. Lastly, Julia presented our V&V report to the class this week as well. |
25 (3/5 - 3/11) |
This week, we continued to refine our prototype. We designed and 3D printed a new circular case that was smaller and had rounded edges, and we determined that we needed to make the diameter 1 mm bigger and increase the height of the case to make the prototype fit. These updates have been made and the updated plans have been sent for printing. We also worked on the software, and tried to figure out the best way to stream the data to make the important information most accessible to the nurses. This is something we will continue to work on next week, but we discussed different programming languages and importing the raw data into a csv file. Lastly, we discussed some design competitions to enter our prototype in, like the DEBUT competition. |
26 (3/5 - 3/11) |
While we had to cancel one of our meetings due to multiple team members being sick, we were still able to meet later on in the week. On Monday, a new print of the case was picked up. It's height and diameter seemed to fit the accelerometer, however, the hole for the wires was placed incorrectly and the bottom was broken/contained gaps (possibly due to trying to make the edge more rounded). We made improvements to our design and sent the next (and hopefully final) version of our case to be printed. We also worked on the software component of the device. Our current approach is to use the Pythagorean theorem to get a singular, timestamped value that represents the acceleration measured in each of the three directions. We will then use that to count movements. We also discussed other approaches, such as calculating the velocity based on time and acceleration values, or creating a "score" based off of the average value of acceleration recorded that can be used to compare between the control and experimental group. |
27 (3/5 - 3/11) |
This week, we met several times to continue working on the prototype. We upgraded the wristband design by using two strips of velcro stuck back-to-back, which made the connection to the case more secure and the size more adjustable. We also finalized our case for the accelerometer to make it as small as possible while still fitting the hardware. Lastly, we continued developing the software. We are able to get raw data and are trying to determine the best way to count movements. We tried using the acceleration values to calculate velocity in addition to trying to determine movements from the acceleration values. To interface with the nurses best, we are looking into an interactive website that would be able to capture and analyze data while giving instructions to the nurses. |
28 (3/5 - 3/11) |
This week, we are finishing the final touches on our prototype and we are continuing to do our verification. We met with one of the web development professors on Thursday to see whether we could create a live website for our device. Additionally, we bought a replacement microcontroller because we have been borrowing one for all of our research up until now. It has a bluetooth component, which we will try to integrate into our design if time permits, but otherwise the prototype is complete. We will be meeting this weekend to finish up final testing and start writing the final report. |