A multimodal explainable ai framework to assist in the differential diagnosis of head and neck reactive follicular hyperplasia and follicular lymphoma: an international multicentre study

The differential diagnosis between reactive follicular hyperplasia (RFH) and follicular lymphoma (FL) in head and neck tissues represents a diagnostic challenge, particularly given the rarity of extranodal FL at these anatomical sites. We developed and evaluated an explainable multimodal artificial intelligence framework to assist in this differential diagnosis. This international multicentre study assembled 108 cases (54 RFH, 54 FL) from 10 centres across 4 continents, representing one of the largest collections of head and neck follicular lesions. We developed a multimodal framework integrating convolutional neural networks (CNNs: AlexNet, VGG16, ResNet18), vision transformers (CellViT + +), graph neural networks (Cell-GNN for spatial analysis), and gradient-boosted decision trees (XGBoost) with morphometric features. Explainability was achieved through Grad-CAM visualisation and SHAP analysis. The best-performing model (CellViT + +) achieved 95.7% accuracy on the internal test set. External validation demonstrated accuracy of 80.5% (ResNet18, Cohort 1) and 69.0% (VGG16-Seg, Cohort 2), with performance variation reflecting the challenge of generalisation across centres. Explainability analysis revealed that the multimodal framework integrated morphometric features (nuclear area, eccentricity) with epidemiological context (patient age, consistent with known FL demographics). Cell-GNN spatial analysis identified significant architectural reorganisation in FL, quantified by Hedges' g = -6.65, representing the loss of normal follicular polarity. This study demonstrates the feasibility of a multimodal explainable AI framework for assisting in the differential diagnosis of rare head and neck follicular lesions. The integration of morphological, spatial, and clinical features provides a foundation for future collaborative validation studies and potential diagnostic support in specialised pathology settings.

Authors

Institutions

Publication Details

Journal
Archiv für Pathologische Anatomie und Physiologie und für Klinische Medicin
Published
2026-05-02
DOI
https://doi.org/10.1007/s00428-026-04527-w
Primary Topic
Salivary Gland Tumors Diagnosis and Treatment
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

A multimodal explainable ai framework to assist in the differential diagnosis of head and neck reactive follicular hyperplasia and follicular lymphoma: an international multicentre study

Pedro Bandeira Aleixo, Paul Hankinson, M L Martins, Hanya Mahmood et al.
Archiv für Pathologische Anatomie und Physiologie und für Klinische Medicin
Salivary Gland Tumors Diagnosis and Treatment
article

A multimodal explainable ai framework to assist in the differential diagnosis of head and neck reactive follicular hyperplasia and follicular lymphoma: an international multicentre study

Pedro Bandeira Aleixo, Paul Hankinson, M L Martins, Hanya Mahmood, Talita de Carvalho Kimura, Leandro Luiz Lopes de Freitas, Matheus Cardoso Moraes, Fábio Luiz Coracin, Ahmed Hagag, Felipe Paiva Fonseca, João Felipe Leite Bonfitto, Ciro Dantas Soares, Sibele Nascimento de Aquino, Zhiyang Chen, Giovanna Calabrese dos Santos, Sheila Aparecida Coelho Siqueira, Fernanda Viviane Mariano, Alan Roger Santos-Silva, Thaís Cerqueira Reis Nakamura, Carolina Louzada Menna-Barreto, Itzia Araceli Torres Torres, Marcio Ajudarte Lopes, Cinthia Veronica Bardález López de Cáceres, Igor Kimura, Pablo Agustin Vargas, Nathalia Gomes Rodrigues, Lucas Lacerda de Souza, Syed Ali Khurram, Adalberto Mosqueda-Taylor, Hélder Antônio Rebelo Pontes, Hannah Walsh
article en

Abstract

The differential diagnosis between reactive follicular hyperplasia (RFH) and follicular lymphoma (FL) in head and neck tissues represents a diagnostic challenge, particularly given the rarity of extranodal FL at these anatomical sites. We developed and evaluated an explainable multimodal artificial intelligence framework to assist in this differential diagnosis. This international multicentre study assembled 108 cases (54 RFH, 54 FL) from 10 centres across 4 continents, representing one of the largest collections of head and neck follicular lesions. We developed a multimodal framework integrating convolutional neural networks (CNNs: AlexNet, VGG16, ResNet18), vision transformers (CellViT + +), graph neural networks (Cell-GNN for spatial analysis), and gradient-boosted decision trees (XGBoost) with morphometric features. Explainability was achieved through Grad-CAM visualisation and SHAP analysis. The best-performing model (CellViT + +) achieved 95.7% accuracy on the internal test set. External validation demonstrated accuracy of 80.5% (ResNet18, Cohort 1) and 69.0% (VGG16-Seg, Cohort 2), with performance variation reflecting the challenge of generalisation across centres. Explainability analysis revealed that the multimodal framework integrated morphometric features (nuclear area, eccentricity) with epidemiological context (patient age, consistent with known FL demographics). Cell-GNN spatial analysis identified significant architectural reorganisation in FL, quantified by Hedges' g = -6.65, representing the loss of normal follicular polarity. This study demonstrates the feasibility of a multimodal explainable AI framework for assisting in the differential diagnosis of rare head and neck follicular lesions. The integration of morphological, spatial, and clinical features provides a foundation for future collaborative validation studies and potential diagnostic support in specialised pathology settings.

Archiv für Pathologische Anatomie und Physiologie und für Klinische Medicin
Universidade Federal de Juiz de Fora (BR), Universidade Federal de Minas Gerais (BR), Arab Open University (OM), Universidade Federal do Rio Grande (BR), Universidade Federal do Rio Grande do Sul (BR), Universidade de São Paulo (BR), Universidade Estadual de Campinas (UNICAMP) (BR), Universidad Autónoma Metropolitana (MX), Santa Casa Hospital (BR), Hospital de Câncer de Barretos (BR), Arab Open University (KW), Arab Open University (EG), Universidade Federal do Pará (BR), Universidade Federal de São Paulo (BR), University of Sheffield (GB)
Fundação de Amparo à Pesquisa do Estado de São Paulo, Conselho Nacional de Desenvolvimento Científico e Tecnológico, Fundação de Amparo à Pesquisa do Estado de Minas Gerais, Ministério da Saúde, Centros de Pesquisa, Inovação e Difusão, Fundação Amazônia Paraense de Amparo à Pesquisa
Partnerships for the goals
Openalex Percentile: Top 17%
Salivary Gland Tumors Diagnosis and Treatment
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.