Validation of the Practicability of Logical Assessment Formula for Evaluations with Inaccurate Ground-Truth Labels: An Application Study on Tumour Segmentation for Breast Cancer
This provides a practical validation for AI evaluations in medical imaging where ground-truth labels are often inaccurate, though it is incremental as it tests an existing theory in a new domain.
The study validated the logical assessment formula (LAF) for evaluating AI models with inaccurate ground-truth labels by applying it to tumour segmentation tasks in breast cancer histopathology, finding it performed reasonably on a difficult task but not confidently on an easier one.
The logical assessment formula (LAF) is a new theory proposed for evaluations with inaccurate ground-truth labels (IAGTLs) to assess the predictive models for artificial intelligence applications. However, the practicability of LAF for evaluations with IAGTLs has not yet been validated in real-world practice. In this paper, we applied LAF to two tasks of tumour segmentation for breast cancer (TSfBC) in medical histopathology whole slide image analysis (MHWSIA) for evaluations with IAGTLs. Experimental results and analysis show that the LAF-based evaluations with IAGTLs were unable to confidently act like usual evaluations with accurate ground-truth labels on the one easier task of TSfBC while being able to reasonably act like usual evaluations with AGTLs on the other more difficult task of TSfBC. These results and analysis reflect the potential of LAF applied to MHWSIA for evaluations with IAGTLs. This paper presents the first practical validation of LAF for evaluations with IAGTLs in a real-world application.