Conformal Prediction for Uncertainty Quantification and Reliable HER2 Status Classification in Breast Cancer IHC Images

Accurate assessment of Human Epidermal growth factor receptor 2 (HER2) status in breast cancer is crucial for determining treatment eligibility for targeted therapies. Although machine learning approaches have shown promise in automating HER2 status classification from immunohistochemistry (IHC) ima...

Full description

Saved in:
Bibliographic Details
Main Authors: Surayuth Pintawong, Shanop Shuangshoti, Tikamporn Jitpasutham, Somruethai Shuangshoti, Kulachet Wiwatwarayos, Thananop Kobchaisawat, Thanarat H. Chalidabhongse
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10933960/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850207324821520384
author Surayuth Pintawong
Shanop Shuangshoti
Tikamporn Jitpasutham
Somruethai Shuangshoti
Kulachet Wiwatwarayos
Thananop Kobchaisawat
Thanarat H. Chalidabhongse
author_facet Surayuth Pintawong
Shanop Shuangshoti
Tikamporn Jitpasutham
Somruethai Shuangshoti
Kulachet Wiwatwarayos
Thananop Kobchaisawat
Thanarat H. Chalidabhongse
author_sort Surayuth Pintawong
collection DOAJ
description Accurate assessment of Human Epidermal growth factor receptor 2 (HER2) status in breast cancer is crucial for determining treatment eligibility for targeted therapies. Although machine learning approaches have shown promise in automating HER2 status classification from immunohistochemistry (IHC) images, they struggle to maintain reliable accuracy in equivocal cases. This study introduces a novel application of Conformal Prediction (CP) framework that, instead of making potentially incorrect predictions for uncertain cases, explicitly identifies them for further examination. The CP framework achieves this by generating prediction sets whose size reflects uncertainty, inherently aligning with the clinical practice of identifying cases that require confirmatory testing. Using dual in situ hybridization (DISH) results as the gold standard for HER2 status, we developed our system using handcrafted features&#x2014;including color intensity, local binary patterns, and Haralick features&#x2014;combined with tree-based classifiers. In experiments on 2,570 IHC images (94 cases) from the Institute of Pathology, Bangkok, Thailand, our framework demonstrated the ability to make highly accurate predictions on unambiguous cases while automatically filtering out uncertain ones. At a significance level of 0.05, the XGBoost-based conformal predictor achieved 90.1% (95% CI: 87.4-92.2%) accuracy on the equivocal cases, while identifying 69.8% (95% CI: 67.7-71.9%) of cases as requiring additional testing. This selective prediction approach represents a practical step toward reliable automated HER2 assessment by focusing on high-confidence predictions. The source code of our implementation is available at: <uri>https://github.com/Surayuth/her2cp.git</uri>.
format Article
id doaj-art-77b5e87b570d42569a3ce1bc0c4b8521
institution OA Journals
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-77b5e87b570d42569a3ce1bc0c4b85212025-08-20T02:10:34ZengIEEEIEEE Access2169-35362025-01-0113516455165910.1109/ACCESS.2025.355293410933960Conformal Prediction for Uncertainty Quantification and Reliable HER2 Status Classification in Breast Cancer IHC ImagesSurayuth Pintawong0https://orcid.org/0009-0006-7507-9827Shanop Shuangshoti1Tikamporn Jitpasutham2Somruethai Shuangshoti3Kulachet Wiwatwarayos4Thananop Kobchaisawat5Thanarat H. Chalidabhongse6https://orcid.org/0000-0002-7249-6135Department of Computer Engineering, Faculty of Engineering, Chulalongkorn University, Bangkok, ThailandDepartment of Pathology, Faculty of Medicine, Chulalongkorn University, Bangkok, ThailandDepartment of Pathology, Faculty of Medicine, Chulalongkorn University, Bangkok, ThailandDepartment of Medical Services, Ministry of Public Health, Institute of Pathology, Bangkok, ThailandDepartment of Medical Services, Ministry of Public Health, Institute of Pathology, Bangkok, ThailandEikonnex AI Company Ltd., Bangkok, ThailandDepartment of Computer Engineering, Faculty of Engineering, Chulalongkorn University, Bangkok, ThailandAccurate assessment of Human Epidermal growth factor receptor 2 (HER2) status in breast cancer is crucial for determining treatment eligibility for targeted therapies. Although machine learning approaches have shown promise in automating HER2 status classification from immunohistochemistry (IHC) images, they struggle to maintain reliable accuracy in equivocal cases. This study introduces a novel application of Conformal Prediction (CP) framework that, instead of making potentially incorrect predictions for uncertain cases, explicitly identifies them for further examination. The CP framework achieves this by generating prediction sets whose size reflects uncertainty, inherently aligning with the clinical practice of identifying cases that require confirmatory testing. Using dual in situ hybridization (DISH) results as the gold standard for HER2 status, we developed our system using handcrafted features&#x2014;including color intensity, local binary patterns, and Haralick features&#x2014;combined with tree-based classifiers. In experiments on 2,570 IHC images (94 cases) from the Institute of Pathology, Bangkok, Thailand, our framework demonstrated the ability to make highly accurate predictions on unambiguous cases while automatically filtering out uncertain ones. At a significance level of 0.05, the XGBoost-based conformal predictor achieved 90.1% (95% CI: 87.4-92.2%) accuracy on the equivocal cases, while identifying 69.8% (95% CI: 67.7-71.9%) of cases as requiring additional testing. This selective prediction approach represents a practical step toward reliable automated HER2 assessment by focusing on high-confidence predictions. The source code of our implementation is available at: <uri>https://github.com/Surayuth/her2cp.git</uri>.https://ieeexplore.ieee.org/document/10933960/Breast cancerconformal predictiondigital pathologyhandcrafted featuresHER2 statusimmunohistochemistry
spellingShingle Surayuth Pintawong
Shanop Shuangshoti
Tikamporn Jitpasutham
Somruethai Shuangshoti
Kulachet Wiwatwarayos
Thananop Kobchaisawat
Thanarat H. Chalidabhongse
Conformal Prediction for Uncertainty Quantification and Reliable HER2 Status Classification in Breast Cancer IHC Images
IEEE Access
Breast cancer
conformal prediction
digital pathology
handcrafted features
HER2 status
immunohistochemistry
title Conformal Prediction for Uncertainty Quantification and Reliable HER2 Status Classification in Breast Cancer IHC Images
title_full Conformal Prediction for Uncertainty Quantification and Reliable HER2 Status Classification in Breast Cancer IHC Images
title_fullStr Conformal Prediction for Uncertainty Quantification and Reliable HER2 Status Classification in Breast Cancer IHC Images
title_full_unstemmed Conformal Prediction for Uncertainty Quantification and Reliable HER2 Status Classification in Breast Cancer IHC Images
title_short Conformal Prediction for Uncertainty Quantification and Reliable HER2 Status Classification in Breast Cancer IHC Images
title_sort conformal prediction for uncertainty quantification and reliable her2 status classification in breast cancer ihc images
topic Breast cancer
conformal prediction
digital pathology
handcrafted features
HER2 status
immunohistochemistry
url https://ieeexplore.ieee.org/document/10933960/
work_keys_str_mv AT surayuthpintawong conformalpredictionforuncertaintyquantificationandreliableher2statusclassificationinbreastcancerihcimages
AT shanopshuangshoti conformalpredictionforuncertaintyquantificationandreliableher2statusclassificationinbreastcancerihcimages
AT tikampornjitpasutham conformalpredictionforuncertaintyquantificationandreliableher2statusclassificationinbreastcancerihcimages
AT somruethaishuangshoti conformalpredictionforuncertaintyquantificationandreliableher2statusclassificationinbreastcancerihcimages
AT kulachetwiwatwarayos conformalpredictionforuncertaintyquantificationandreliableher2statusclassificationinbreastcancerihcimages
AT thananopkobchaisawat conformalpredictionforuncertaintyquantificationandreliableher2statusclassificationinbreastcancerihcimages
AT thanarathchalidabhongse conformalpredictionforuncertaintyquantificationandreliableher2statusclassificationinbreastcancerihcimages