Tumor Cellularity Assessment of Breast Histopathological Slides via Instance Segmentation and Pathomic Features Explainability

Abstract

The segmentation and classification of cell nuclei are pivotal steps in the pipelines for the analysis of bioimages. Deep learning (DL) approaches are leading the digital pathology field in the context of nuclei detection and classification. Nevertheless, the features that are exploited by DL models to make their predictions are difficult to interpret, hindering the deployment of such methods in clinical practice. On the other hand, pathomic features can be linked to an easier description of the characteristics exploited by the classifiers for making the final predictions. Thus, in this work, we developed an explainable computer-aided diagnosis (CAD) system that can be used to support pathologists in the evaluation of tumor cellularity in breast histopathological slides. In particular, we compared an end-to-end DL approach that exploits the Mask R-CNN instance segmentation architecture with a two steps pipeline, where the features are extracted while considering the morphological and textural characteristics of the cell nuclei. Classifiers that are based on support vector machines and artificial neural networks are trained on top of these features in order to discriminate between tumor and non-tumor nuclei. Afterwards, the SHAP (Shapley additive explanations) explainable artificial intelligence technique was employed to perform a feature importance analysis, which led to an understanding of the features processed by the machine learning models for making their decisions. An expert pathologist validated the employed feature set, corroborating the clinical usability of the model. Even though the models resulting from the two-stage pipeline are slightly less accurate than those of the end-to-end approach, the interpretability of their features is clearer and may help build trust for pathologists to adopt artificial intelligence-based CAD systems in their clinical workflow. To further show the validity of the proposed approach, it has been tested on an external validation dataset, which was collected from IRCCS Istituto Tumori “Giovanni Paolo II” and made publicly available to ease research concerning the quantification of tumor cellularity.

Radiometric Infrared Thermography of Solar Photovoltaic Systems: An Explainable Predictive Maintenance Approach for Remote Aerial Diagnostic Monitoring

“…The pursuit of interpretability and explainability, via the adoption of XAI techniques [65], particularly in DL models, is crucial for understanding how these models arrive at their predictions [66][67][68][69][70][71]. Two broad categories of methods are commonly employed to achieve explainability and interpretability [66,67]: perceptive explainability and mathematical interpretability.…”

Machine learning in onco-pharmacogenomics: a path to precision medicine with many challenges

“…This tool reduces the time required to assess lymph nodes, which in turn improves the diagnostic process and treatment decisions ( Kindler et al, 2023 ). Another DL tool has been developed to assist clinicians in digital pathology by assessing the tumor cellularity of histopathologic hematoxylin and eosin sections ( Altini et al, 2023 ). Apart from the importance that this algorithm may have in the clinical setting, it is important to point out that its use may also be useful in research, as it allows pathologists to share valuable information with researchers in an automated manner.…”

An Explainable Radiogenomic Framework to Predict Mutational Status of KRAS and EGFR in Lung Adenocarcinoma Patients

“…Medical imaging techniques can offer a non-invasive way to visualize phenotypic differences between different kinds of tumors and lesions. In this context, radiomics is a rapidly expanding approach that involves extracting a large number of quantitative features from images to determine the phenotype of regions of interest, possibly through the use of intelligent automatic systems [10][11][12][13][14]. These features encompass a range of properties, including shape characteristics, textural features, and pixel intensities and can provide information about areas affected by a disease, also allowing the identification and quantitative description of tumor patterns and characteristics [15][16][17][18].…”