Explainable AI to predict non-cancer causes of death for people with breast, colorectal, prostate, or lung cancer
SCH: Screening and confirmatory machine learning for explainable modeling of non-cancer deaths in cancer patients
This project uses a two-step, explainable AI approach on genomic and medical-record data to better predict non-cancer causes of death for people living with breast, colorectal, prostate, or lung cancer.
Quick facts
| Grant type | R37 grant |
|---|---|
| Study type | NIH-funded research |
| Funding institution | Rutgers Biomedical and Health Sciences NIH-funded |
| Lab location | 1 site (Newark, UNITED STATES) |
| Project ID | NIH-11180985 on NIH RePORTER |
What this research studies
Researchers will build a two-step system where a very sensitive screening AI first flags patients at risk for non-cancer death and a specific confirmatory AI then reduces false alarms, with both models designed to produce clear explanations clinicians can understand. The team will train and test these tools using genomic ('omic) data and electronic health records from people with breast, colorectal, prostate, and lung cancers. The focus is on achieving both very high sensitivity and very high specificity so clinicians receive fewer missed risks and fewer false positives. The project will also produce workflows and methods intended to make these tools usable with clinical data.
Who could benefit from this research
Good fit: Ideal participants are adults with breast, colorectal, prostate, or lung cancer who have electronic health records and available genomic ('omic) data or who can provide samples and medical-record access.
Not a fit: People without linked medical records or genomic data, children, or patients with cancer types outside breast, colorectal, prostate, or lung are unlikely to benefit directly from this project.
Why it matters
Potential benefit: If successful, this work could help clinicians spot and prevent non-cancer health risks in cancer survivors, potentially reducing avoidable deaths.
How similar studies have performed: Some machine-learning work has improved clinical predictions, but separating screening and confirmatory explainable tools for non-cancer death in cancer patients is relatively novel and not yet proven.
Where this research is happening
Newark, UNITED STATES
- Rutgers Biomedical and Health Sciences — Newark, United States (Active)
Researchers
- Principal investigator: Zhang, Lanjing — Rutgers Biomedical and Health Sciences
- Study coordinator: Zhang, Lanjing
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.