Week 1: First Steps in Developing a New Prototype

Geoju A -

Hello, everyone! My name is Geoju, and welcome to my Senior Project Blog! I am a high school senior at BASIS San Antonio Shavano Campus, conducting my senior project research under Dr. Dang and Priyabrata Dash at the LUCENT Lab at UTSA. Over the next 11 weeks, I will be working in their lab to complete my research project. Specifically, I plan to develop a prototype ammonia breath sensor that detects fluctuations in ammonia concentration in a patient’s breath, helping individuals monitor their health and potentially prevent serious complications.

This past week has been an eye-opening experience, giving me insight into the complexities of undergraduate and graduate-level research. I had the opportunity to attend a lab-wide meeting, where I saw firsthand the diverse projects being developed. One group worked on a Raspberry Pi cluster to function as a server for running larger AI models, while another focused on optimizing photonics calculations. Despite working on seemingly separate projects, each team contributed to the lab’s overall mission of advancing AI and photonics research. The LUCENT Lab primarily explores AI systems and photonics, emphasizing applying these technologies to smaller-scale devices—where my project fits in. My goal is to create a compact healthcare device that integrates efficiency programs developed by the lab, enabling AI models to run on smaller devices without requiring significant power.

This week, I focused on familiarizing myself with the lab environment and researching previous studies related to my project. A significant challenge has been selecting the right sensor for my prototype. There are multiple ways to measure ammonia concentration, including spectroscopy, chemo-resistive materials, and biological sensors, each with distinct advantages. Beyond the sensor, I also need to consider the other components that will make the device functional and practical. For example, the battery must be small enough to avoid overheating issues while providing sufficient power. I am also considering using a Raspberry Pi as the device’s processing unit, as it balances compact size and performance. Additionally, I need to design the screen and frame to ensure ease of use and durability. Over the next few weeks, I will finalize my materials and begin testing and calibrating the sensor to optimize its accuracy and efficiency.

Thank you for reading, and I look forward to sharing my progress next week!

More Posts

Comments:

All viewpoints are welcome but profane, threatening, disrespectful, or harassing comments will not be tolerated and are subject to moderation up to, and including, full deletion.

    james_l
    Hey Geoju! Your project seems to be really innovative and impactful! I was wondering if you have been thinking about potential future applications for some of the technology you are developing. I know that there are specific biomarkers in the body that can tell if someone is about to develop something like cancer. Do you think that the technology you are building could be implemented to detecting cancer biomarkers or biomarkers for other diseases and would be able to become kind of like a portable breathalyzer that can tell if someone might get a certain disease? Also with multiple sensor options available for ammonia detection, what specific factors are guiding your decision in selecting the most suitable one? Are you prioritizing sensitivity, cost, or another factor? Thank you and I look forward to reading your next blog!
    March 19, 2025 at 9:08 pm - Reply
    mariska_k
    Hey Geoju! This is a super interesting project! You frequently mention the importance of the device's small size; do you have an estimate of its final dimensions? Also, regarding the different ways of measuring ammonia concentration, do these sensors work similarly to alcohol breathalyzers? If so, can external factors like prescription medications, food, etc. create artificial errors? Thanks!
    March 20, 2025 at 11:07 am - Reply
    geoju_a
    Hey James, thank you for your comments! I believe the technology I'm working with could be used to detect cancer and other disease biomarkers. The key requirement for detection is that the sensor must have a catalyst capable of accurately reacting to the biomarkers at low concentrations while distinguishing potential noise from other compounds in the air. That said, I wouldn't necessarily use breath sensing for more complex illnesses like cancer because false results could have serious consequences. For such cases, I'd assume that more invasive testing methods—such as blood samples, urine samples, or even biopsies—would be necessary to ensure accuracy. I chose breath sensing for the ammonia level in the body because it's much less invasive and significantly faster than many other diagnostic techniques while still producing reliable results. As for the factors guiding my decision to select the most suitable sensor, I'll likely prioritize cost during this project. Since the goal is to develop a prototype for commercial use, the sensor, and other components must be affordable and reliable without compromising the model's accuracy. I hope that answers your questions. Thanks for reading!
    March 20, 2025 at 12:42 pm - Reply
    mariana_g_p
    Hi Geoju. Your project sounds like it has the potential to be extremely beneficial to a lot of people. I was hoping you could expand on what health conditions this could help determine. Also, could you expand on the role of a Raspberry Pi in your overall product? I'm not super knowledgeable about what a processing unit does. Looking forward to reading about your progress!
    March 20, 2025 at 6:30 pm - Reply
    geoju_a
    Hey Mariska, thanks for your questions! For the final dimensions, I aim to keep the device compact—ideally around the size of a standard alcohol breathalyzer. These typically resemble a small box with a nozzle for the user to blow into, and I plan to follow a similar design. However, my device may be slightly more significant because it includes a Raspberry Pi and additional sensors. With further prototyping and testing, I hope to refine and shrink the design to match the portability of a breathalyzer. There are several potential sources of error regarding external factors affecting ammonia concentration measurements. I discuss this in my Week 2 post, but since the sensor collects data from the mouth, even factors like dental hygiene (specifically bacterial presence) can influence breath ammonia levels. One of the biggest challenges is ensuring accurate readings despite external noise from diet, medications, and other variables. With further testing and possibly integrating an AI model, the device could improve accuracy by learning to filter out inconsistencies and standardize measurements across different users. I hope that answers your questions!
    March 25, 2025 at 11:24 am - Reply
    geoju_a
    Hey Maria, thank you for your questions! This device has the potential to aid in diagnosing and monitoring various health conditions. Ammonia breath detection has been used to identify Helicobacter pylori (H. pylori), a bacterium that can cause peptic ulcers and other illnesses. Elevated ammonia levels in the body can be harmful to patients, as ammonia is a toxic byproduct that, if not properly regulated, can lead to severe health complications. In particular, high ammonia levels in the breath can indicate impaired kidney function since the kidneys filter ammonia from the bloodstream. By monitoring ammonia concentration, this device could help track a patient's condition and potentially prevent further complications. As for the Raspberry Pi, it serves as the device's processing unit. Essentially, it takes the raw data from the ammonia sensor and processes it before displaying the results meaningfully. Instead of directly outputting the sensor's raw readings, I plan to write code that interprets the data and categorizes ammonia levels as healthy, borderline, or potentially dangerous. This would make it easier for users to understand their results. Additionally, the Raspberry Pi allows for future expansion—I want to integrate other input parameters, such as a patient's personal and medical history, to refine the accuracy of the readings. I hope that answers your questions!
    March 25, 2025 at 11:52 am - Reply

Leave a Reply

Your email address will not be published. Required fields are marked *