Our ultimate goal is to improve the lives of cancer patients by integrating artificial intelligence (AI), machine learning, computational methods, and experimental approaches. Our interdisciplinary research spans spatial biology, single-cell analysis, digital pathology, biomarker discovery, and therapeutic development, including immunotherapy and cellular therapy strategies.
We actively develop AI-driven solutions for 3D tumor atlas models, which enable us to explore the complex interplay between cellular and molecular components. This approach guides the identification of novel, clinically actionable biomarkers, therapeutics, and personalized treatment strategies. By synergistically combining cutting-edge technologies and methodologies, the Hwang Lab strives to advance personalized medicine and therapeutic development, focusing on the tumor immune microenvironment, ultimately benefiting cancer patients.
These are a few examples for our current projects:
AI and Machine Learning-driven Investigation of Treatment Resistance in Cancer Using 3D Tumor Atlas Models
The goal of this project is to understand treatment resistance mechanisms in solid tumors by integrating subcellular-resolution 3D tumor atlas models with advanced imaging and spatial transcriptomics/proteomics data. We employ machine learning and AI algorithms to uncover hidden patterns and interactions in the tumor immune microenvironment (TIME) that contribute to treatment resistance. This research aims to identify novel biomarkers, therapeutic targets, and strategies to enhance the efficacy of emerging therapeutics, ultimately improving patient outcomes.Enhancing Cellular Therapy Efficacy with AI and Machine Learning-driven Single Cell Analysis
The goal of this project is to uncover the mechanisms behind suboptimal cellular therapy outcomes, such as those seen with CD19 CAR-T treatments. We develop and apply cutting-edge machine learning and deep learning algorithms to analyze single cell RNA and protein expression data from CD19 CAR-T clinical trials. Our research enables the discovery of novel combinatorial therapeutic strategies and the identification of ideal CAR-T cell subsets to improve patient outcomes. A prime example of our work is our recent publication in Cancer Discovery (2022), where we successfully identified TIGIT as a biomarker and developed and validated a complementary therapeutic approach. This project is a collaborative effort with David Wald's lab at Case Western Reserve University, and together, we continue to investigate additional biomarkers and therapeutic targets to enhance the effectiveness of cellular therapies.Developing Multimodal AI and Machine Learning Algorithms for Personalized Cancer Treatment
The goal of this project is to create innovative AI and machine learning algorithms that integrate multimodal data, including imaging, spatial, single cell genomic data, and electronic health records (EHRs), to identify predictive biomarkers and guide treatment decisions for advanced or refractory cancer patients. By focusing on the development of cutting-edge algorithms and tools, we aim to better stratify patients and optimize treatment selection across various modalities such as immunotherapy, targeted therapy, and chemotherapy. Our research harnesses the power of AI and machine learning to drive advancements in personalized medicine, improving patient outcomes and transforming cancer care.