FOXP3-stained image analysis for follicular lymphoma: Optimal adaptive thresholding with maximal nucleus coverage

Area of Science:

• Immunology
• Computational Pathology
• Oncology

Background:

• Regulatory T cells (TR), identified by FOXP3, are vital for immune homeostasis but can be co-opted by tumors to suppress anti-tumor immunity.
• In follicular lymphoma (FL), the prognostic significance of tumor-infiltrating TR remains debated.
• Accurate enumeration of FOXP3+ TR in FL biopsies is essential for resolving this controversy.

Purpose of the Study:

• To develop and validate a novel computational method for detecting and quantifying FOXP3+ TR nuclei in FL tissue.
• To address the limitations of existing methods in accurately segmenting and counting nuclei in complex tissue images.

Main Methods:

• A novel Optimal Adaptive Thresholding (OAT) method was developed for initial nuclei segmentation.
• An Elliptical Arc and Line Segment Detector (ELSD) was integrated to refine segmentation and split merged nuclei.
• Simple Linear Iterative Clustering (SLIC), a super-pixel method, was used to further resolve clustered nuclei.

Main Results:

• The method demonstrated high accuracy in detecting both negative and positive FOXP3+ nuclei.
• Sensitivity ranged from 83-100% for negative nuclei and 90-95% for positive nuclei.
• Precision values were exceptionally high, ranging from 98-100% for negative and 99-100% for positive nuclei.

Conclusions:

• The developed computational method provides a robust and accurate approach for enumerating FOXP3+ TR in FL biopsies.
• This tool can facilitate further research into the prognostic impact of TR in follicular lymphoma.
• Accurate quantification of FOXP3+ cells is key to understanding their role in FL pathogenesis and patient outcomes.