Haiyang Yu, Wang Yu, Yuan Enwu, Jun Ma, Xin Zhao, Linlin Zhang, Fang Yang
{"title":"利用hnsc分类器提高数字化全切片组织学头颈癌检测的准确性:一种深度学习方法。","authors":"Haiyang Yu, Wang Yu, Yuan Enwu, Jun Ma, Xin Zhao, Linlin Zhang, Fang Yang","doi":"10.3389/fmolb.2025.1652144","DOIUrl":null,"url":null,"abstract":"<p><p>Head and neck squamous cell carcinoma (HNSCC) represents the sixth most common cancer worldwide, with pathologists routinely analyzing histological slides to diagnose cancer by evaluating cellular heterogeneity, a process that remains time-consuming and labor-intensive. Although no previous studies have systematically applied deep learning techniques to automate HNSCC TNM staging and overall stage prediction from digital histopathology slides, we developed an inception-ResNet34 convolutional neural network model (HNSC-Classifier) trained on 791 whole slide images (WSIs) from 500 HNSCC patients sourced from The Cancer Genome Atlas (TCGA) Head and Neck Squamous Cell dataset. Our pipeline was designed to distinguish cancerous from normal tissue and to predict both tumor stage and TNM classification from histological images, with the dataset split at the patient level to ensure independence between training and testing sets and performance evaluated using comprehensive metrics including receiver operating characteristic (ROC) analysis, precision, recall, F1-score, and confusion matrices. The HNSC-Classifier demonstrated exceptional performance with areas under the ROC curves (AUCs) of 0.998 for both cancer/normal classification and TNM system stage prediction at the tile level, while cross-validation showed high precision, recall, and F1 scores (>0.99) across all classification tasks. Patient-level classification achieved AUCs of 0.998 for tumor/normal discrimination and 0.992 for stage prediction, significantly outperforming existing approaches for cancer stage detection. Our deep learning approach provides pathologists with a powerful computational tool that can enhance diagnostic efficiency and accuracy in HNSCC detection and staging, with the HNSC-Classifier having potential to improve clinical workflow and patient outcomes through more timely and precise diagnoses, serving as an automated decision support system for histopathological analysis of HNSCC.</p>","PeriodicalId":12465,"journal":{"name":"Frontiers in Molecular Biosciences","volume":"12 ","pages":"1652144"},"PeriodicalIF":3.9000,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12353728/pdf/","citationCount":"0","resultStr":"{\"title\":\"Enhancing head and neck cancer detection accuracy in digitized whole-slide histology with the HNSC-classifier: a deep learning approach.\",\"authors\":\"Haiyang Yu, Wang Yu, Yuan Enwu, Jun Ma, Xin Zhao, Linlin Zhang, Fang Yang\",\"doi\":\"10.3389/fmolb.2025.1652144\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>Head and neck squamous cell carcinoma (HNSCC) represents the sixth most common cancer worldwide, with pathologists routinely analyzing histological slides to diagnose cancer by evaluating cellular heterogeneity, a process that remains time-consuming and labor-intensive. Although no previous studies have systematically applied deep learning techniques to automate HNSCC TNM staging and overall stage prediction from digital histopathology slides, we developed an inception-ResNet34 convolutional neural network model (HNSC-Classifier) trained on 791 whole slide images (WSIs) from 500 HNSCC patients sourced from The Cancer Genome Atlas (TCGA) Head and Neck Squamous Cell dataset. Our pipeline was designed to distinguish cancerous from normal tissue and to predict both tumor stage and TNM classification from histological images, with the dataset split at the patient level to ensure independence between training and testing sets and performance evaluated using comprehensive metrics including receiver operating characteristic (ROC) analysis, precision, recall, F1-score, and confusion matrices. The HNSC-Classifier demonstrated exceptional performance with areas under the ROC curves (AUCs) of 0.998 for both cancer/normal classification and TNM system stage prediction at the tile level, while cross-validation showed high precision, recall, and F1 scores (>0.99) across all classification tasks. Patient-level classification achieved AUCs of 0.998 for tumor/normal discrimination and 0.992 for stage prediction, significantly outperforming existing approaches for cancer stage detection. Our deep learning approach provides pathologists with a powerful computational tool that can enhance diagnostic efficiency and accuracy in HNSCC detection and staging, with the HNSC-Classifier having potential to improve clinical workflow and patient outcomes through more timely and precise diagnoses, serving as an automated decision support system for histopathological analysis of HNSCC.</p>\",\"PeriodicalId\":12465,\"journal\":{\"name\":\"Frontiers in Molecular Biosciences\",\"volume\":\"12 \",\"pages\":\"1652144\"},\"PeriodicalIF\":3.9000,\"publicationDate\":\"2025-08-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12353728/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Frontiers in Molecular Biosciences\",\"FirstCategoryId\":\"99\",\"ListUrlMain\":\"https://doi.org/10.3389/fmolb.2025.1652144\",\"RegionNum\":3,\"RegionCategory\":\"生物学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"2025/1/1 0:00:00\",\"PubModel\":\"eCollection\",\"JCR\":\"Q2\",\"JCRName\":\"BIOCHEMISTRY & MOLECULAR BIOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Frontiers in Molecular Biosciences","FirstCategoryId":"99","ListUrlMain":"https://doi.org/10.3389/fmolb.2025.1652144","RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2025/1/1 0:00:00","PubModel":"eCollection","JCR":"Q2","JCRName":"BIOCHEMISTRY & MOLECULAR BIOLOGY","Score":null,"Total":0}
Enhancing head and neck cancer detection accuracy in digitized whole-slide histology with the HNSC-classifier: a deep learning approach.
Head and neck squamous cell carcinoma (HNSCC) represents the sixth most common cancer worldwide, with pathologists routinely analyzing histological slides to diagnose cancer by evaluating cellular heterogeneity, a process that remains time-consuming and labor-intensive. Although no previous studies have systematically applied deep learning techniques to automate HNSCC TNM staging and overall stage prediction from digital histopathology slides, we developed an inception-ResNet34 convolutional neural network model (HNSC-Classifier) trained on 791 whole slide images (WSIs) from 500 HNSCC patients sourced from The Cancer Genome Atlas (TCGA) Head and Neck Squamous Cell dataset. Our pipeline was designed to distinguish cancerous from normal tissue and to predict both tumor stage and TNM classification from histological images, with the dataset split at the patient level to ensure independence between training and testing sets and performance evaluated using comprehensive metrics including receiver operating characteristic (ROC) analysis, precision, recall, F1-score, and confusion matrices. The HNSC-Classifier demonstrated exceptional performance with areas under the ROC curves (AUCs) of 0.998 for both cancer/normal classification and TNM system stage prediction at the tile level, while cross-validation showed high precision, recall, and F1 scores (>0.99) across all classification tasks. Patient-level classification achieved AUCs of 0.998 for tumor/normal discrimination and 0.992 for stage prediction, significantly outperforming existing approaches for cancer stage detection. Our deep learning approach provides pathologists with a powerful computational tool that can enhance diagnostic efficiency and accuracy in HNSCC detection and staging, with the HNSC-Classifier having potential to improve clinical workflow and patient outcomes through more timely and precise diagnoses, serving as an automated decision support system for histopathological analysis of HNSCC.
期刊介绍:
Much of contemporary investigation in the life sciences is devoted to the molecular-scale understanding of the relationships between genes and the environment — in particular, dynamic alterations in the levels, modifications, and interactions of cellular effectors, including proteins. Frontiers in Molecular Biosciences offers an international publication platform for basic as well as applied research; we encourage contributions spanning both established and emerging areas of biology. To this end, the journal draws from empirical disciplines such as structural biology, enzymology, biochemistry, and biophysics, capitalizing as well on the technological advancements that have enabled metabolomics and proteomics measurements in massively parallel throughput, and the development of robust and innovative computational biology strategies. We also recognize influences from medicine and technology, welcoming studies in molecular genetics, molecular diagnostics and therapeutics, and nanotechnology.
Our ultimate objective is the comprehensive illustration of the molecular mechanisms regulating proteins, nucleic acids, carbohydrates, lipids, and small metabolites in organisms across all branches of life.
In addition to interesting new findings, techniques, and applications, Frontiers in Molecular Biosciences will consider new testable hypotheses to inspire different perspectives and stimulate scientific dialogue. The integration of in silico, in vitro, and in vivo approaches will benefit endeavors across all domains of the life sciences.