Improving AI methods for analyzing clinical language data
DeconDTN: Deconfounding Deep Transformer Networks for Clinical NLP
This study is working on improving how computers understand doctors' notes and speech about changes in thinking, so that they can help make better decisions in healthcare, especially by reducing any unfair influences that might confuse the results.
Quick facts
| Grant type | R01 grant |
|---|---|
| Study type | NIH-funded research |
| Funding institution | University of Washington NIH-funded |
| Lab location | 1 site (Seattle, United States) |
| Project ID | NIH-11044988 on NIH RePORTER |
What this research studies
This research focuses on enhancing Natural Language Processing (NLP) techniques used in clinical settings, particularly through the use of advanced deep learning models known as deep transformer networks. These models aim to accurately interpret clinical notes and speech samples that reflect cognitive changes. However, the research addresses the challenge of confounding biases that can affect the accuracy of these models, especially when trained on diverse medical data. By developing methods to mitigate these biases, the project seeks to improve the reliability of AI applications in clinical decision-making.
Who could benefit from this research
Good fit: Ideal candidates for this research are individuals undergoing cognitive assessments or those with conditions that affect cognitive function.
Not a fit: Patients with stable cognitive function who are not undergoing any assessments may not benefit from this research.
Why it matters
Potential benefit: If successful, this research could lead to more accurate AI tools that assist healthcare providers in diagnosing and monitoring cognitive conditions.
How similar studies have performed: Previous research has shown promise in using AI for clinical NLP, but addressing confounding biases in deep learning models is a relatively novel approach.
Where this research is happening
Seattle, United States
- University of Washington — Seattle, United States (Active)
Researchers
- Principal investigator: Cohen, Trevor — University of Washington
- Study coordinator: Cohen, Trevor
About this research
- This is an active NIH-funded research project — typically early-stage science, not a clinical trial accepting patient enrollment.
- Some NIH-funded labs run parallel clinical studies or seek volunteers for related work. To check, contact the principal investigator or institution listed above.
- For full project details, budget, and progress reports, visit the official NIH RePORTER page below.