Attention-guided classification and tumor localization of combined hepatocellular-cholangiocarcinoma

Xiaowen Liang; Hélène Cazier; Aurélie Beaufrère; Miguel Albuquerque; Valérie Paradis; Cédric Wemmert

cs.CVOct 22, 2026

Attention-guided classification and tumor localization of combined hepatocellular-cholangiocarcinoma

Authors: Xiaowen Liang, Hélène Cazier, Aurélie Beaufrère, Miguel Albuquerque, Valérie Paradis, Cédric Wemmert

Organizations: Université de Strasbourg, CNRS, ICube UMR 7357, Illkirch, France · Université Paris Cité, INSERM, UMR-S1149, Centre de Recherche sur l’Inflammation (CRI), Laboratoire d’Excellence Inflamex, F-75018 Paris, France · Pathology Department, Hôpital Beaujon, AP-HP, Clichy, France

Abstract

Combined hepatocellular-cholangiocarcinomas (cHCC-CCA) are a type of cancer that share components of two other primary liver cancers: hepatocellular carcinoma (HCC) and cholangiocarcinoma (CCA). Accurate diagnosis of cHCC-CCA, of which the behavior may resemble that of either HCC or CCA, remains a challenge due to strong intratumoral heterogeneity. Since the prognoses differ between liver cancer types, it is necessary to improve their morphological characterization and recognition in order to accurately identify cHCC-CCA. In this work, we use whole slide images (WSIs) of the two main types of liver cancer (HCC and CCA) and cHCC-CCA, stained with Hematoxylin-Eosin-Safran (HES), to train a classification model that can distinguish between these subtypes and provides a potential solution for localizing different types of tumors in the combined liver cancer images.

Figures & tables

Figure 1: Attention-guided three-class MIL training.

Figure 2: Joint learning of the three-class model and the dual model. The 3-class MIL block is the same as described in Figure 1 . Red dotted lines indicate the generation of cHCC-CCA attention heatmaps used to compute the auxiliary loss $\mathcal{L}_{\mathrm{aux}}$ , comprising union alignment loss $\mathcal{L}_{\mathrm{Union}}$ , background suppression loss $\mathcal{L}_{\mathrm{BG}}$ , and foreground-guided complementary loss $\mathcal{L}_{\mathrm{comp}}$ . The three-class model and both branches of the dual model are trained with classification and attention supervision as described previously.

Model	CCA	HCC	TM	Macro F1
	Recall	Recall	Recall
w/o attn. sup.	0.915	0.868	0.867	0.880
w/ attn. sup.	0.884	0.871	0.871	0.872
Joint	0.903	0.863	0.871	0.876
Dual	0.918	0.768	0.816	0.834

Table 1: Classification performance of the three-class models (rows 1–3) and the dual-branch model (row 4).

Class	TAC		Slides with $\overline{\mathrm{TAC}}<1$
	w/o AS	w/ AS	w/o AS	w/ AS
CCA	$3.04\pm 1.63$	$5.66\pm 1.12$	$0$	$0$
HCC	$1.75\pm 0.51$	$6.50\pm 1.72$	$6$	$0$

Table 2: Attention plausibility on 30 slides per class, without (w/o AS) and with attention supervision (w/ AS).

\overline{\mathrm{TAC}}

denotes per-slide TAC averaged across the ten fold models.

Panel of Figure 3 — Figure 3: HCC slide with annotated tumor regions and attention heatmaps without and with attention supervision.

Model	CCA/HCC attention ratio
Three-class model	$0.677\pm 0.090$
CCA-like branch	$2.483\pm 0.873$
HCC-like branch	$0.480\pm 0.102$

Table 3: Component selectivity on 15 annotated cHCC-CCA slides, measured by the CCA/HCC attention ratio and averaged across ten fold models.

Panel of Figure 4 — Figure 4: cHCC-CCA slide with (a) annotated tumor regions (CCA in green, HCC in red) and attention heatmaps of (b) 3-class model, (c) branch 1 and (d) branch 2.

Cancer Subtype Classification

See all 15 papers

A Digital Pathology Resource for Liver Cancer Quantification with Datasets, Benchmarks, and Tools
Apr 22, 2026Ying Xiao, Shimiao Tang, Xitong Ling +11Computational Pathology Image Segmentation
Department of Pathology, Beijing Tsinghua Changgung Hospital, School of Clinical Medicine, Tsinghua Medicine, Tsinghua University, Beijing, China · Shenzhen International Graduate School, Tsinghua University, Shenzhen, China
Semantic Segmentation-Driven Image-Level Diagnosis of Liver Cancers in Hematoxylin and Eosin Histopathology Images
Jul 3, 2026Ivica Kopriva, Dario Sitnik, Arijana Pacic +3Computational Pathology Histopathology Image Segmentation
Division of Computing and Data Science, Ruđer Bošković Institute, Bijenička cesta 54, Zagreb Croatia · Sitnik Ventures UG, Lautensackstr. 25, 80687 Munich, Germany · Department of Pathology and Cytology, Dubrava University Hospital, Avenija Gojka Šuška 6, 10000 Zagreb, Croatia +2
An interpretable deep learning method for early hepatocellular carcinoma diagnosis on multiphase CT
Oct 21, 2026Emma Sarfati, Alexandre Bône, Pietro Gori +8Medical Diagnosis Medical Image Classification
Guerbet Research, Villepinte, France · LTCI, Télécom Paris, Institut Polytechnique de Paris, Saclay, France · MAP5 (UMR 8145), Université Paris Cité, 45 rue des Saints-Pères, Paris, France +7

Apr 22, 2026J/K move · Enter open · S save

A Digital Pathology Resource for Liver Cancer Quantification with Datasets, Benchmarks, and Tools

Ying Xiao, Shimiao Tang, Xitong Ling, Weiming Chen+10

Department of Pathology, Beijing Tsinghua Changgung Hospital, School of Clinical Medicine, Tsinghua Medicine, Tsinghua University, Beijing, China · Shenzhen International Graduate School, Tsinghua University, Shenzhen, China

Computational Pathology Image Segmentation Histopathology Image Segmentation Oncology Histopathology Image Classification Medical Image Benchmarks+1

Open paper AskPDF arXiv

Liver cancer, especially hepatocellular carcinoma (HCC), imposes a substantial global disease burden. Accurate diagnosis and prognostic assessment directly influence treatment selection and patient survival, and pathological examination remains the gold standard for liver cancer diagnosis. Identifying diverse tissue components and pathological subtypes on histopathology slides is crucial for estimating postoperative recurrence risk and overall prognosis. However, most publicly available resources are still provided at the whole-slide image (WSI) level, and well-annotated datasets for fine-grained tissue component identification in liver cancer are scarce, which hinders reproducible model development and the deployment of quantitative analysis tools. To address this gap, we release HepatoBench, a patch-level image database for liver cancer with annotations for seven key tissue categories. Based on HepatoBench, we train and open-source a deep learning classification model as a tissue recognition tool. Furthermore, we train a WSI-level tumor/non-tumor segmentation model to automatically localize lesion regions across entire slides. By integrating the patch-level tissue classifier with the WSI-level segmentation model, we build HepatoQuant, an end-to-end, disease-specific regional quantification tool for liver cancer, enabling a unified workflow from WSIs to tissue composition parsing and quantitative statistics. We also open-source HepatoBench, the benchmarking protocol, and supporting tools, providing a solid foundation for automated regional quantification and fair method comparison in liver cancer pathology.

Abstract

Cancer Subtype Classification

Explore similar work

A Digital Pathology Resource for Liver Cancer Quantification with Datasets, Benchmarks, and Tools

Semantic Segmentation-Driven Image-Level Diagnosis of Liver Cancers in Hematoxylin and Eosin Histopathology Images

An interpretable deep learning method for early hepatocellular carcinoma diagnosis on multiphase CT