Validation of the Practicability of Logical Assessment Formula for Evaluations with Inaccurate Ground-Truth Labels: An Application Study on Tumour Segmentation for Breast Cancer

TL;DR

This study validates the logical assessment formula (LAF) for evaluating AI tumor segmentation models in breast cancer, demonstrating its effectiveness with inaccurate ground-truth labels in real-world histopathology image analysis.

Contribution

First practical validation of LAF in real-world medical image analysis with inaccurate ground-truth labels, specifically for breast cancer tumor segmentation.

Findings

LAF performs well on difficult tasks with inaccurate labels.

LAF is less confident on easier tasks with inaccurate labels.

This work confirms LAF's potential in real-world medical evaluations.

Abstract

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.