Dr. Chang Su, assistant professor of population health sciences and Walsh McDermott Scholar in Public Health, and director of the Weill Cornell Medicine MS in Health Informatics and Artificial Intelligence program, has received an R01 to develop an artificial intelligence (AI)-enhanced integrated pipeline aimed at improving early detection of Parkinson’s disease (PD) and accelerating disease-modifying therapeutic treatment development.
PD is the second most prevalent neurodegenerative disorder worldwide, affecting an estimated 1.1 million people in the US. PD, caused by the death of dopamine-producing neurons, is associated with movement, speech, and swallowing dysfunctions, as well as diverse non-motor symptoms such as depression, anxiety, sleep disorders, and cognitive problems, substantially impacting daily activities and quality of life. Although treatments are available to manage symptoms, no disease-modifying therapy currently exists to slow, halt, or reverse the course of PD.
Parkinson’s is typically diagnosed only after movement symptoms appear, but the disease pathology may begin decades earlier. By then, nearly half of the brain’s dopamine-producing neurons have already been lost, making it much harder to develop treatments that can slow or stop disease progression.
“The prodromal phase of PD represents a critical and promising therapeutic window, during which the underlying PD pathologies may still be reversible or more amenable to intervention,” said Dr. Su. This phase indicates the onset of disease and is characterized by early, nonspecific signs and symptoms. “However, a major barrier is the lack of robust, scalable tools to accurately identify individuals in the prodromal phase of PD.”
To address this challenge, Dr. Su’s team will develop an AI framework to harmonize and integrate multimodal patient-level data, including clinical records, neuroimaging, and multi-omics data such as genomics, transcriptomics, and proteomics. This framework will support more accurate identification of individuals in the prodromal phase of PD, laying the foundation for earlier intervention strategies and improved design of disease-modifying therapy trials.
“Today’s vast multimodal health data provide an unprecedented opportunity for Parkinson’s disease research,” said Dr. Su. “AI enables us to integrate these complex data in new ways, overcome challenges such as data heterogeneity and missing modality, and identify populations at risk before irreversible neurodegeneration has occurred.”
The project will also use real-world electronic health records to emulate disease-modifying trials among the prodromal PD population and identify candidate drugs for repurposing. Promising candidates will then be further screened and validated using human midbrain organoids and mouse models. Human midbrain organoids, derived from human midbrain cells, can model key aspects of brain tissue and PD biology, providing a biologically relevant platform for evaluating therapeutic potential.
“It’s important to integrate advanced AI and data science techniques with wet lab experiments to accelerate treatment development,” explained Dr. Su. “Using mouse models and human organoids, we can mimic PD pathology and evaluate whether certain drugs stop cell death or neurodegenerative processes.”
Through this work, Dr. Su and his team aim to create a scalable AI-driven pipeline that bridges early disease detection, real-world drug repurposing, and experimental validation, with the long-term goal of advancing precision prevention and treatment strategies for Parkinson’s disease.
This project brings together expertise in health AI, health informatics, biostatistics, neuroscience, and neurology. Dr. Su will work with two other principal investigators: Dr. Fei Wang, associate dean of data science and artificial intelligence, and Dr. Roberta Marongiu, assistant professor of genetics and neuroscience in neurological surgery. His co-investigators include Dr. Shuibing Chen, Kilts Family Professor of Surgery and professor of chemical biology in biochemistry and biophysics; Dr. Harini Sarva, associate professor of clinical neurology; Dr. Yifan Peng, associate professor of population health sciences and radiology; Dr. Wodan Ling, assistant professor of population health sciences, and Dr. He Sarina Yang, associate professor of clinical pathology and laboratory medicine. Dr. Su extends gratitude to members of the Department of Population Health Sciences, the Division of Health Informatics and Artificial Intelligence, the Institute of Artificial Intelligence for Digital Health, and the grants and finance team for their continued support.
“Early identification of brain and neurodegenerative disorders like Parkinson’s and Alzheimer’s is critical but very difficult,” said Dr. Su. “Our pipeline is designed to be flexible and encourages collaboration across disciplines. We intend for it to be adapted to address other chronic, progressive diseases that lack effective treatments.”
- Highlights