# Screening and confirmatory machine learning for explainable modeling of non-cancer deaths in cancer patients

> **NIH NIH R37** · RUTGERS BIOMEDICAL AND HEALTH SCIENCES · 2024 · $105,528

## Abstract

Abstract
This supplement proposal is closely related the parent grant for two major reasons: 1). Breast
cancer patients with disabilities fared worse than those without disabilities. The same occurred in
non-small cell cancer or cervical cancer patients with disabilities. However, it is unclear how to
best use ML tools to address this health disparity issue. As a direct extension of the parent grant’s
overarching goal that focuses on all cancer patients, the proposed work will exclusively focus on
the cancer patients with disabilities and plan to improve ML performance and survival rates in
these patients. 2). Based on reported incident cancer cases in patients with disabilities and all
incident cancer cases about 27-31% of the colorectal, lung or prostate cancer patients in the
SEER-Medicare dataset had disabilities at the time of cancer diagnosis. Therefore, the proposed
research will shed light on how modelling a minority population differs from modelling the whole
population in a real-world dataset using our novel ML algorithms. Thus, it is a direct extension of
parent grant’s Aim 1.2 that focuses on impact of imbalanced label-distribution.
 There are four major barriers that we feel significantly limit the research and career
potential of the candidate, who is an Early Stage Investigator (ESI) and has a documented
disability. Without this supplement focusing on these barriers, she will unlikely successfully
compete for a R01 grant or become an independent investigator. Upon completion of this
supplement, we expect to overcome the four major barriers, build a solid research foundation and
generate ample preliminary data to successfully compete for R01 grants.

## Key facts

- **NIH application ID:** 11061641
- **Project number:** 3R37CA277812-03S1
- **Recipient organization:** RUTGERS BIOMEDICAL AND HEALTH SCIENCES
- **Principal Investigator:** Lanjing Zhang
- **Activity code:** R37 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2024
- **Award amount:** $105,528
- **Award type:** 3
- **Project period:** 2022-09-01 → 2026-08-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/11061641

## Citation

> US National Institutes of Health, RePORTER application 11061641, Screening and confirmatory machine learning for explainable modeling of non-cancer deaths in cancer patients (3R37CA277812-03S1). Retrieved via AI Analytics 2026-05-23 from https://api.ai-analytics.org/grant/nih/11061641. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*