Research Areas

Pathology Image Analysis for Precision Oncology

A major focus of our lab is developing new machine learning methods to analyze and extract information from histopathology images including standard H&E-stained slides and multiplexed immunofluorescence imaging. We are investigating a variety of approaches to obtain comprehensive spatial charazterizaion of the tumor microenvironment through single-cell analysis (Sali et al. JNCI 2024). We are working with a team of oncologists and pathologists toward the goal of predicting immunotherapy response and patient outcomes for personalized cancer therapy.

Radiology Image Analysis for Precision Oncology

Another focus area has been extracting information from radiology images for clinical outcome prediction. Our approach incorporates deep learning as well as knowledge-based radiomics profiling of the intratumoral subregions/habitats and surrounding parenchyma. We have conducted imaging biomarker studies on brain, breast, lung, head and neck, gastric, and rectal cancers, using several modalities including PET, CT, and MRI (Cui et al. Radiology, 2016; Wu et al. Radiology, 2016, 2018; Jiang et al. Annals Surgery, 2020, Jin et al. Nature Communications, 2021; Jiang et al. Lancet Digital Health 2022). In a Nature Machine Intelligence study, we developed a unifying radiological tumor classification system that demonstrates prognostic relevance and therapeutic implications across multiple cancer types.

Integrative Imaging and Immunogenomics

We are also exploring the biological basis of clinically relevant radiologic phenotypes by correlating imaging with molecular data (Wu et al. Radiology, 2017; Wu et al. Clin Cancer Res, 2017). In more recent studies, we developed a noninvasive radiographic approach to evaluate tumor immune and stromal microenvironment, and showed these imaging signatures could predict immunotherapy response and survival outcomes (Jiang et al. Annals Oncology, 2020; Lancet Digital Health, 2021; Nat Comms 2023; Cell Reports Medicine 2023).

Molecular Analysis

In addition to imaging, we are also developing molecular biomarkers to predict therapy response and prognosis in cancer. We are particularly interested in how various components of the tumor microenvironment (including immune and stroma) interact with each other and contribute to cancer progression or control. Previously we developed an individualized immune gene expression signature that predicts survival of patients with non-small cell lung cancer (Li et al. JAMA Oncology, 2017).  In another work, we integrated radiation sensitivity and immune signatures to predict which patients are most likely to derive survival benefit from adjuvant radiotherapy in breast cancer (Cui et al. Clin Cancer Res, 2018). In a study of pan-squamous cell carcinomas (including esophagus, head and neck, lung, and cervix), we identified 6 immune subtypes associated with distinct molecular characteristics and clinical outcomes (Li et al. Clin Cancer Res, 2019).