V P Gladis Pushparathi, S R Sylaja Vallee Narayan, R S Pratheeba, V Naveen
{"title":"组织病理学图像分析和提高口腔癌检测的诊断准确性。","authors":"V P Gladis Pushparathi, S R Sylaja Vallee Narayan, R S Pratheeba, V Naveen","doi":"10.1080/07357907.2025.2559103","DOIUrl":null,"url":null,"abstract":"<p><p>Deep learning (DL) has transformed medical imaging, particularly in the realm of Oral Cancer (OC) diagnosis using histopathological images. Timely detection of OC is essential for enhancing precision medicine and saving lives. However, incorrect diagnosis may impede effective treatment. In this study, we have proposed a DL model for OC classification, enhanced diagnosis decision-making, and interpretability. We achieve this by starting with color normalization of histopathology images using the Vahadane Three-Stain Parameter Normalization and watershed segmentation method, followed by tiling and augmentation. Key features are selected using the Weighted Fisher Score (WFS) to address class imbalance. The U-Net classifier has been improved by using feature-based inputs instead of full images, reducing computational complexity and training time. The integration of Vahadane normalization for consistent preprocessing across samples, WFS, and Explainable Artificial Intelligence (XAI) addresses critical challenges in histopathological image analysis. The proposed model surpasses existing approaches with a classification accuracy of 99.54% and outperforms DenseNet201 and VGG10 in precision and reliability. The efficiency in handling imbalanced datasets and explainability features make it suitable for early precise OC detection, which can reduce diagnostic errors and enhance treatment outcomes..</p>","PeriodicalId":9463,"journal":{"name":"Cancer Investigation","volume":" ","pages":"1-14"},"PeriodicalIF":1.9000,"publicationDate":"2025-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Histopathological Image Analysis and Enhanced Diagnostic Accuracy Explainability for Oral Cancer Detection.\",\"authors\":\"V P Gladis Pushparathi, S R Sylaja Vallee Narayan, R S Pratheeba, V Naveen\",\"doi\":\"10.1080/07357907.2025.2559103\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>Deep learning (DL) has transformed medical imaging, particularly in the realm of Oral Cancer (OC) diagnosis using histopathological images. Timely detection of OC is essential for enhancing precision medicine and saving lives. However, incorrect diagnosis may impede effective treatment. In this study, we have proposed a DL model for OC classification, enhanced diagnosis decision-making, and interpretability. We achieve this by starting with color normalization of histopathology images using the Vahadane Three-Stain Parameter Normalization and watershed segmentation method, followed by tiling and augmentation. Key features are selected using the Weighted Fisher Score (WFS) to address class imbalance. The U-Net classifier has been improved by using feature-based inputs instead of full images, reducing computational complexity and training time. The integration of Vahadane normalization for consistent preprocessing across samples, WFS, and Explainable Artificial Intelligence (XAI) addresses critical challenges in histopathological image analysis. The proposed model surpasses existing approaches with a classification accuracy of 99.54% and outperforms DenseNet201 and VGG10 in precision and reliability. The efficiency in handling imbalanced datasets and explainability features make it suitable for early precise OC detection, which can reduce diagnostic errors and enhance treatment outcomes..</p>\",\"PeriodicalId\":9463,\"journal\":{\"name\":\"Cancer Investigation\",\"volume\":\" \",\"pages\":\"1-14\"},\"PeriodicalIF\":1.9000,\"publicationDate\":\"2025-09-15\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Cancer Investigation\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1080/07357907.2025.2559103\",\"RegionNum\":4,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"ONCOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Cancer Investigation","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1080/07357907.2025.2559103","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ONCOLOGY","Score":null,"Total":0}
Histopathological Image Analysis and Enhanced Diagnostic Accuracy Explainability for Oral Cancer Detection.
Deep learning (DL) has transformed medical imaging, particularly in the realm of Oral Cancer (OC) diagnosis using histopathological images. Timely detection of OC is essential for enhancing precision medicine and saving lives. However, incorrect diagnosis may impede effective treatment. In this study, we have proposed a DL model for OC classification, enhanced diagnosis decision-making, and interpretability. We achieve this by starting with color normalization of histopathology images using the Vahadane Three-Stain Parameter Normalization and watershed segmentation method, followed by tiling and augmentation. Key features are selected using the Weighted Fisher Score (WFS) to address class imbalance. The U-Net classifier has been improved by using feature-based inputs instead of full images, reducing computational complexity and training time. The integration of Vahadane normalization for consistent preprocessing across samples, WFS, and Explainable Artificial Intelligence (XAI) addresses critical challenges in histopathological image analysis. The proposed model surpasses existing approaches with a classification accuracy of 99.54% and outperforms DenseNet201 and VGG10 in precision and reliability. The efficiency in handling imbalanced datasets and explainability features make it suitable for early precise OC detection, which can reduce diagnostic errors and enhance treatment outcomes..
期刊介绍:
Cancer Investigation is one of the most highly regarded and recognized journals in the field of basic and clinical oncology. It is designed to give physicians a comprehensive resource on the current state of progress in the cancer field as well as a broad background of reliable information necessary for effective decision making. In addition to presenting original papers of fundamental significance, it also publishes reviews, essays, specialized presentations of controversies, considerations of new technologies and their applications to specific laboratory problems, discussions of public issues, miniseries on major topics, new and experimental drugs and therapies, and an innovative letters to the editor section. One of the unique features of the journal is its departmentalized editorial sections reporting on more than 30 subject categories covering the broad spectrum of specialized areas that together comprise the field of oncology. Edited by leading physicians and research scientists, these sections make Cancer Investigation the prime resource for clinicians seeking to make sense of the sometimes-overwhelming amount of information available throughout the field. In addition to its peer-reviewed clinical research, the journal also features translational studies that bridge the gap between the laboratory and the clinic.