Syed Ahmad Chan Bukhari
St. John's University-New York, NY·Computer Science
Medicine produces huge amounts of data, from drug and gene databases to electronic health records, but turning that data into useful decisions is hard. These projects build artificial intelligence (AI) systems that link facts into knowledge graphs, networks that connect drugs, diseases, genes and medical concepts, and combine them with machine learning.
- Fields
Computer Science
Medicine & Clinical Research
Pharmacology
- Topics
artificial intelligence
neuro-symbolic AI
knowledge graphs
- Grants
- 2 known grants, through Jun 2029
What they work on
- Build RESPOND, a hybrid neuro-symbolic AI that combines knowledge graphs with machine learning, to rank FDA-approved drugs likely to act on disease pathways in pulmonary arterial hypertension and idiopathic pulmonary fibrosis.
- Test top drug candidates from RESPOND in laboratory (in vitro) experiments and animal (in vivo) studies to see whether they slow progression of PAH and IPF.
- Integrate varied electronic medical record data into a single knowledge graph and develop algorithms that build patient context from these records for medical code assignment.
- Develop an AI coding framework in which a retriever and a generator interact over repeated rounds to refine billing code recommendations, guided by feedback from healthcare professionals.
More about the research
One project applies this to finding new uses for existing drugs in lung disease. The other applies it to the paperwork side of healthcare: assigning the billing codes that describe each patient visit.
Why it matters
Patients with pulmonary arterial hypertension and idiopathic pulmonary fibrosis have limited treatment options; repurposing already-approved drugs could speed new therapies to them and lower costs. Separately, more accurate automated medical coding could reduce billing errors, ensure correct reimbursement claims, and free doctors and coders to spend more time on patient care.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Meng Jiang
University of Notre Dame, IN·Computer Science·Early-career
Develop and compare natural language generation models, one data-driven and one based on theories of abduction, for comparative summarization and argument generation in scientific domains.
Yong Huang
University of Virginia, VA·Medicine & Clinical Research
Test whether reducing lung stretch by paralyzing one side of the diaphragm slows fibrosis progression in the outer and lower lung of mice with bleomycin-induced lung injury.
Rishikesan Kamaleswaran
Duke University, NC·Biomedical Engineering
Discover 'physiomarkers' in continuous ICU monitor data, including high-fidelity waveforms, that predict the onset of sepsis in patients who develop secondary infections after ICU admission.
The school
St. John's University-New York: 2 labs on PI Pages, $5M a year on research, 140 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.