Explainable AI Models to Predict Metabolic Cancer Biomarkers

Cancer’s persistent prevalence calls for innovative steps in early detection and therapeutic interventions. Metabolomics offers a comprehensive view of the metabolites that measure the function of cellular processes in this context. As explainable AI (XAI) emerges, it sheds light on these intricate systems’ decision-making processes and outcomes, making them accessible to human understanding.
The project aims to address the critical need for early cancer detection and treatment by advancing the field of metabolomics. Metabolic data will be analyzed to identify possible cancer diagnosis and prognosis biomarkers. The project will employ XAI techniques to enhance the interpretability of the prediction models used in medical diagnostics. This will ensure that healthcare professionals can understand and trust these complex systems’ decision-making processes. The interns will apply XAI models to metabolite features, validate their significance through pathway analysis, and strive to improve the transparency and applicability of these models in clinical settings. The ultimate goal is to contribute to personalized medicine by bridging the gap between data analytics and clinical practice.

Faculty Supervisor:

Abedalrhman Alkhateeb

Student:

Partner:

Inönü University

Discipline:

Computer science

Sector:

Education

University:

Lakehead University

Program: