Projects

This project focuses on building a reproducible computational foundation for biologically informed risk stratification in Burkitt lymphoma. The work integrates genomic, transcriptomic, and clinical data across multiple Burkitt lymphoma cohorts in order to support pathway inference, subgroup assignment, and development of the proposed M2-BLAIPI risk framework. The broader aim is to move beyond purely clinical prognostic tools by incorporating biologically meaningful features that better capture disease heteroge

Closely related to the broader Burkitt lymphoma program, this project examines how inferred metabolic pathway activity and literature-derived biological features can contribute to clinically useful predictive modeling. The work uses transcriptomic pathway analysis, feature engineering, and multivariate modeling to identify pathway signatures associated with Burkitt lymphoma genetic subgroups and to evaluate their contribution to expanded risk prediction. This project reflects Dr. Jarso’s interest in using computational biology and quantitative feature selection to improve clinical decision support.

This area of work focuses on the emerging use of artificial intelligence and whole-slide imaging in cervical cancer screening. The project area is motivated by the promise of AI-assisted cytology to strengthen screening workflows, expand access to expert review, and support lower-cost diagnostic systems in both remote and resource-constrained settings. Dr. Jarso’s interest in this space aligns with a translational agenda that combines imaging, low-cost hardware development, and clinically deployable decision support for cancer screening.

This project area addresses the challenge of late diagnosis and limited access to prostate cancer care in Uganda. The work is framed around understanding how patients encounter barriers across the care pathway, including screening, referral, treatment access, and follow-up. Using a structured access-to-care framework, this project seeks to identify measurable dimensions of healthcare access and to support more effective assessment of prostate cancer care delivery in low-resource systems. 

This implementation-focused project explores how circulating tumor cell measurement and liquid biopsy technologies might be incorporated into systemic treatment planning and monitoring for late-stage breast cancer patients in Uganda. The work is centered on the clinical decision problem of how to use new biomarker data to stratify patients into more meaningful risk groups and improve treatment monitoring.

Across these projects, Dr. Jarso’s work is shaped by a consistent methodological approach: identifying clinically meaningful problems, developing quantitative or engineering-based solutions, and working toward tools that can be translated into real-world practice. His projects often combine feature selection, imaging or molecular data interpretation, risk modeling, and context-aware design for use in low-resource environments. They also frequently involve collaboration across disciplines, including engineering, oncology, pathology, computational biology, and global health. 

Many of Dr. Jarso’s projects are connected to long-term work in Uganda and sub-Saharan Africa, where cancer care challenges are often intensified by constrained infrastructure, late presentation, limited screening access, and reduced availability of advanced diagnostic systems. His project portfolio reflects a sustained commitment to building tools, frameworks, and collaborations that are technically rigorous while remaining responsive to local clinical realities.