Quantitative histopathologic methods for assessing novel imaging biomarkers of cancer invasion: application to brain tumours
Supervisors:
PhD Project Summary
A major factor for treatment failure in high-grade brain tumours is the ability of tumour cells to infiltrate normal brain regions, compromising complete surgical resection and contributing to high recurrence rates. Novel imaging biomarkers of marginal tumour invasion are urgently needed as conventional imaging fails to probe marginal infiltration, reducing the accuracy of radiotherapy target delineation and surgery planning. One of the major challenges for the deployment of these biomarkers to clinics is their validation.
Our groups have introduced and assessed imaging biomarkers of tumour invasion, combining novel MRI acquisition methods with new histological processing and image analysis pipelines. To account for imaging slice orientation and thickness, multiple histology sections are cut in the MRI imaging plane, and averaged to produce stacked in-plane histopathology maps acting as ground-truth for imaging biomarker assessment.
Building on this strong interdisciplinary experience, the proposed PhD project will aim at producing new robust methods for quantitative histopathologic assessment of novel tumour infiltration biomarkers based on a combination of adapted tissue handling methods with purpose-coded image analysis techniques. The resulting gold-standard tumour cell density maps will be used to assess clinically-relevant MRI-based biomarkers of tumour infiltration and AI machine learning models predicting spatial tumour progression.