PRoBE Laboratory

Pattern Recognition from Biomedical Evidence Laboratory is designed:

• • To harness prior knowledge for effective knowledge discovery from biomedical data.
  • To design and develop novel machine learning algorithms using symbolic, probabilistic and hybrid approaches to solve bioinformatics problems of clinical importance such as biomarker discovery and disease classification.
  • To develop complex pattern recognition tools that can be plugged into computer-aided diagnostic systems to facilitate evidence combination from heterogeneous sources such as data from imaging, de-identified clinical information and biochemical profiling.

Research Mission:

AI approaches to solving real-world problems:

• • • Include human in the loop as needed (machine + human combinations)
    • Human interpretable predictive modeling (explainable AI)
    • Integrative model development from diverse data types and knowledge sources (multimodal collective intelligence)

Diseases Studied (Biomarker Discovery / Early Detection):

–Amyotrophic Lateral Sclerosis

–Alzheimer’s

–Lung Cancer

–Breast Cancer

–Esophageal Cancer

–Inflammatory Bowel Diseases Ulcerative Colitis, Crohn’s

–Coronary Artery Disease

–Pediatric Cardiomyopathy

–Helminth’s infections

–Beryllium Exposure

–Type 2 Diabetes

–Cardiovascular Disease

–Intervertebral Aging

Grants as a Principal Investigator:

• NIH/ NIGMS K25 Mentored Quantitative Research Development Award: Intelligent Aids for Proteomic Data Mining (5 years)
• NIH/NLM R01 Bayesian Rule Learning Methods for Disease Prediction and Biomarker Discovery (3 years)
• NIH/NIGMS R01 Transfer Rule Learning for Knowledge Based Biomarker Discovery and Predictive Biomedicine (3 years)
• NIH/NIGMS R01 (Renewal) Transfer Rule Learning with Functional Mapping for Integrative Modeling of Panomics Data (3 years)
• PHDA Grant: CADidME: Coronary Artery Disease intelligent detection via Metabolomic Expression (1 year)

Other Notable Research Activities:

• Investigator: NSF CMU Language Technologies Institute – Biological Language Modeling” (4-5 years)
• Investigator: TATRC/DoD – Bioinformatics core (3 years)
• Co-director of Bioinformatics and Biostatistics Core: NCI Hillman Cancer Center – Lung SPORE (Specialized Programs of Research Excellence) (9 years)
• Inventor: US Patent App. 15/107,444, 2016 – Methods for the detection of esophageal adenocarcinoma

Current Collaborators:

Edith Luhanga
Assistant Research Professor
CMU-Africa

Current Graduate Students:

William Reynolds
(Rafael Ceschin: Primary Advisor)

Past Collaborators:

Jaime Carbonell
Professor

Nam Vo
Professor

Blair Jobe
Surgeon
AHN

Featured Post Doctoral Advisees:

Mari Mori
Clinical Geneticist

Doctoral Student Graduates from PRoBE Lab (dissertations)