Diagnosis and prognosis of urothelial cancer are performed via assessment of specific tissue and cellular features by pathologists. Importantly, intra and inter variability of pathologists can affect diagnosis. There is a need to develop methods for reducing the inconsistency of pathologic assessment.
Computer-aided Diagnosis (CAD) is a promising tool for automation of the diagnostic processes and reducing intra/inter variability in urothelial cancer diagnosis.
CAD is being utilized in image analysis too. However, there is no CAD currently available that employs pathological features in HR-NMIBC patients for diagnosis and prognosis. Our goal is to improve HR-NMIBC patient’s diagnosis by developing deep learning algorithms and using artificial intelligence to aid in the diagnosis of HR-NMIBC patients via digital pathology. We aim to provide pathologists and clinicians with an accurate, affordable, and reproducible diagnostic and prognostic tool. Our research project CLARIFY (Cloud Artificial Intelligence For pathologY), is funded by the European Commission.