Medicine and Health Projects at the 8th Symposium of Rising Scholars

Student researchers at Polygence conduct original research under the guidance of a mentor, and each project culminates in some type of research showcase. Over 160 students showcased their work at Polygence's 8th Symposium for Rising Scholars this past March. In this post, I will discuss projects related to medicine and health from the symposium. There were a wide range of projects in this field, from music therapy to artificial intelligence to bioengineering. Read on to learn more about this incredible work!

Projects about Brain Diseases and Artificial Intelligence 

Melinda Yao discussed how and why music can help people with Alzheimer's. Alzheimer’s is a neurogenetic disease that causes problems with learning, memory, and thinking and is eventually fatal. Since pharmacological drugs only reduce symptoms and come with side effects, researchers are studying using nonpharmacological treatments alongside pharmacological drugs. Music therapy is one such type of nonpharmacological treatment. Musical memory is the last type of memory to be degraded by Alzheimer’s, and it is partially separate from other memory systems. Researchers have found that music therapy improves patients’ verbal fluency, memory, and psychiatric symptoms.

Melinda conducted a literature review on the neurological mechanisms underlying music’s effect on the brain. She found that music increases levels of a certain type of protein in the brain. This increase in turn impacts neuroplasticity, structural changes in the brain that strengthen neural connections, and neurogenesis, the generation of new neurons. In the future, Melinda hopes that researchers will explore how specific elements of music, such as genre, tempo, pitch, and rhythm, affect the brain. This understanding will lead to more targeted treatments for Alzheimer’s.

Sucheer Maddury investigated how to use Artificial Intelligence (AI) to detect Huntington's Disease, which is another neurogenetic disease. Huntington's Disease (HD) is genetic, which means that people have it from birth. However, symptoms only appear in middle age. These symptoms are characterized by motor, cognitive, and/or behavioral disturbances. 

Currently the only early detection method is genetic testing, which is very expensive. Thus, people are usually diagnosed in the early symptomatic stage. Early intervention can improve outcomes, so there need to be reliable and less expensive detection methods. 

Since the genetic abnormality is present before symptoms manifest, Sucheer wanted to investigate a way to detect HD early through biomedical signals. Biomedical signals are a way of tracking something in the body over some period of time. He used data collected at the Neurological Clinic in Slovenia which had biomedical signals from HD patients and control patients. Sucheer evaluated the accuracy of various machine learning methods in detecting patients with HD, and he found that Extra Trees and Random Forest were the most successful methods. This work is the first piece of research that computationally analyzes biomedical signals in HD patients, and it shows that biomedical signals are a useful biomarker for HD. You can read the full paper here.

Similarly, in her presentation called “Artificial Intelligence: Competitor or best friend of breast cancer radiologists?”, Yutonia Tang  investigated how AI can be used in detecting breast cancer. She was inspired to conduct her research by her mom, who is an oncologist, and her grandmother, who was recently diagnosed with breast cancer. Radiologists use mammogram screenings to detect breast cancer early. However, the radiologists must subjectively interpret the screenings. This subjectivity leads to women being diagnosed with breast cancer when they do not actually have it. Thus, researchers are investigating how to use AI to aid in this detection process.

Yutonia collected data from seven previous studies that compared how AI by itself, radiologists by themselves, and AI-informed radiologists performed in detecting breast cancer. She did a meta-analysis on these studies, a type of primary research where a researcher statistically analyzes the results from many different studies and sees what the overall results are. Overall, AI performed better than radiologists by themselves, and the combination of AI and radiologists was better than AI by itself. While these results were not statistically significant, they show that the collaboration between radiologists and AI is promising. Yutonia acknowledged the pitfalls of AI, such as the fact that it is based on data from specific subpopulations that may not be representative of the larger population, and called for researchers to conduct more inclusive studies. 

Projects that Involve Medical Hardware and Products  

Some Polygence scholars focused on physical products that could be used to address health issues. Rishab Bhattacharya shared his ongoing work in his talk, “Band to prevent compulsive motion.” Obsessive-compulsive motions are a class of disorder that results in a compulsive tendency to do motions, such as skin picking and hair pulling, which can be psychologically and physically harmful. Rishab is currently developing an easily usable device that detects and stops compulsive movements by using vibrations or (non-harmful) electric shocks. Thus far, he has created a proof of concept from both the hardware and software side. On the hardware side, he used an arduino, a gyroscope, and an accelerometer to detect when there are certain motions. He has also written code that records data from the accelerometer and gyroscope and compares it to previous values in order to detect when someone is engaging in compulsive behavior. 

Currently, Rishab is working on expanding on the hardware and software aspects of the device. He will make the hardware smaller by putting it into a 3D-printed container along with a commercial band strap. To prevent the software from detecting that someone is doing the behavior when they actually are not, he will incorporate a more advanced machine learning algorithm. He also wants to give users the option to have a vibration rather than an electric shock. Rishab hopes to one day make this product commercial and accessible!

Similarly, Alisha Gupta proposed a new and more accessible treatment for Osteoarthritis (OA) in her presentation. OA is characterized by the wearing of cartilage in joints and most frequently affects hands, hips, and knees. This pain is a significant cause of mobility impairment in older adults. Treatment of OA often involves full replacement of the joint, i.e., hip or knee replacement surgery. Alish aimed to create a treatment that is minimally invasive and more accessible for people with end-stage OA. 

Alisha’s proposed treatment focuses on bioengineering. After consulting with a PhD student at Georgia Tech and a professor at Duke, she decided that it would be best to use an injectable component made of scaffold hydrogel. The scaffold is 3D-printed and developed for each specific patient. It gives the mechanical structure to the component and keeps the joint open so that the hydrogel can do its job. The hydrogel makes up 75% of the component and regenerates the cartilage tissue. Previous research has found that a similar combination of components as Alisha wants to use in the hydrogel were successful in growing cells. She hopes that the results of her injectable treatment would be similar. This research would provide an accessible and lower-cost treatment for OA, which could have significant implications in the treatment of other diseases. 

Conclusion

There are dozens more projects that relate to the industry of biomedicine and health sciences - and it’s simply wonderfully impressive what our students have been able to achieve in the span of their projects. If you’re reading about these passion projects and feel an itch to start you brown, check out our article on the many different ways that you can creatively explore a topic you’re interested in. Apply today to do research with the guidance of a Polygence mentor. Conducting research really gives you an edge on your college admissions.