DiCoL – Digital Colposcopy
Cervical cancer remains a major global health challenge, particularly in low to middle income countries (LMIC) where access to advanced diagnostic tools is limited. The DiCol project aims to integrate artificial intelligence (AI) into colposcopy to enhance the detection of dysplastic and malignant cervical lesions. The DiCol project, led by ATB at Heidelberg University Hospital in collaboration with PAICON GmbH, has established a robust data pipeline designed to support the development of an AI-assisted colposcopy system. This pipeline integrates global data collection capabilities, enabling the aggregation of diverse colposcopic image datasets from multiple international clinical partners. The integrated annotation platform ensures standardized labeling, with a rigorous expert review system involving medical experts to guarantee high-quality ground truth data. The annotated and reviewed datasets are seamlessly integrated into the AI algorithm, allowing the enhancement of the algorithm’s diagnostic accuracy and generalizability to diverse population of LMICs, ensuring deployment in digital colposcopes as well as smartphones for clinical settings worldwide.
Our partners
