Personalized treatment predictions from scans and medical records
Deep Learning for Individualized Treatment Effect Inference with Multimodal Depiction
This project builds AI that uses medical images and clinical data to predict how different treatments might change outcomes for individual patients with conditions like glioblastoma.
Quick facts
| Grant type | R21 grant |
|---|---|
| Study type | NIH-funded research |
| Funding institution | Yale University NIH-funded |
| Lab location | 1 site (New Haven, United States) |
| Project ID | NIH-11307164 on NIH RePORTER |
What this research studies
Researchers will combine pre-treatment brain scans and clinical records from many patients to train deep learning models that estimate individual treatment effects. The models are designed to compare multiple treatment options and predict both actual and hypothetical (counterfactual) outcomes, including time-to-event and ordered outcomes. The team will add measures of reliability and explanations so clinicians can understand and trust the predictions. The work focuses on multimodal data (imaging plus clinical information) to make the predictions more complete and clinically useful.
Who could benefit from this research
Good fit: Ideal candidates are people with brain tumors such as glioblastoma who have pre-treatment imaging and clinical records and are considering different treatment options.
Not a fit: Patients without available digital scans or clinical records, those with very rare conditions not represented in the data, or people whose care does not involve the treatments modeled may not see direct benefit.
Why it matters
Potential benefit: If successful, this could help doctors and patients choose treatments more likely to extend survival or improve outcomes for each person.
How similar studies have performed: Some prior machine learning methods have shown promise on retrospective patient datasets, but this multimodal, multi-treatment, uncertainty-aware approach is relatively new and not yet proven in routine clinical care.
Where this research is happening
New Haven, United States
- Yale University — New Haven, United States (Active)
Researchers
- Principal investigator: Liu, Xiaofeng — Yale University
- Study coordinator: Liu, Xiaofeng
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.