AI methods for Epilepsy Detection/Application of Sleep structures
Project by Polygence alum Briana
Project's result
I completed a research paper investigating whether machine learning can predict individual responses to temazepam using baseline polysomnographic sleep characteristics. I also created a presentation summarising the methods, findings, and limitations, which I presented at the Symposium of Rising Scholars. The paper was also accepted by the Research Archive of Rising Scholars (RARS) as a preprint and submitted to The Curieux Review for peer review, with a decision pending.
They started it from zero. Are you ready to level up with us?
Summary
Sleep disorders affect millions of people worldwide, who then commonly get prescribed benzodiazepines such as temazepam, helping improve sleep quality. However, individual responses to these medications can vary substantially. This makes it difficult to identify which patients are most likely to benefit from treatment. Thus, this study aims to investigate the extent to which machine learning can be used to predict individual responses to temazepam using baseline polysomnographic (PSG) sleep characteristics.
Through the use of paired placebo and temazepam overnight sleep recordings from 22 participants, multiple sleep metrics were extracted. These sleep metrics were then compared between treatment conditions. Exploratory analyses evaluated changes in overall sleep architecture, variability between individuals, and demographic subgroups. AI model development followed an 80/10/10 train-validation-test split to maximise predictive performance while minimising overfitting. Ridge Regression and Random Forest Regression were tested using leave-one-out cross-validation (LOOCV) within the development cohort, with change in sleep efficiency defined as the primary treatment-response outcome.
Temazepam was associated with a significant reduction in N1 sleep (p = 0.024) and increase in N2 sleep (p = 0.027). Conversely, changes in overall sleep efficiency, total sleep time, and Wake After Sleep Onset (WASO) were not statistically significant. Ridge Regression achieved an LOOCV Mean Absolute Error (MAE) of 3.63 percentage points, compared with 3.79 for Random Forest Regression. However, Random Forest produced lower prediction errors in the validation set and was therefore chosen for independent testing, where it produced an MAE of 3.03 percentage points.
These findings suggest that baseline PSG characteristics do contain information relevant to individual temazepam response. Predictive performance, however, was relatively limited due to a smaller sample size and varied across evaluation stages. Thus, the findings should be considered preliminary and support the feasibility of individualised treatment-response modelling rather than be seen as a clinically reliable prediction.
Rahul
Polygence mentor
PhD Doctor of Philosophy
Subjects
Psychology, Neuroscience, Biology, Computer Science
Expertise
Neuroscience, Biology, Engineering, Genetics, Data Science
Check out their profile
Briana
Student
Graduation Year
2027
Project review
“The project exceeded my expectations in terms of how much independence I had in developing my project while still receiving consistent guidance from my mentor. I especially valued learning new methods in data analysis and machine learning, and developing the project into a complete research paper. The mentorship and feedback throughout the process were extremely helpful and allowed me to take the project further than I initially expected.”
About my mentor
“I had an excellent experience working with Rahul. He was very knowledgeable, supportive, and easy to communicate with throughout the entire project. I learned a huge amount from working with him. I would highly recommend him to any student interested in conducting a research project in this area.”
Check out their profile