Sun Zhu, Huiyan Jiang, Zhaoshuo Diao, Qiu Luan, Yaming Li, Xuena Li, Yan Pei
{"title":"CoF-DResNet: Cancer Metastasis Recognition Network based on Dynamic\nCoordinated Metabolic Attention and Structural Attention","authors":"Sun Zhu, Huiyan Jiang, Zhaoshuo Diao, Qiu Luan, Yaming Li, Xuena Li, Yan Pei","doi":"10.2174/0113892010302534240530073118","DOIUrl":null,"url":null,"abstract":"\n\nCancer metastasis usually means that cancer cells spread to other tissues\nor organs, and the condition worsens. Identifying whether cancer has metastasized can help\ndoctors infer the progression of a patient's condition and is an essential prerequisite for devising\ntreatment plans. Fluorine 18 fluorodeoxyglucose positron emission tomography/computed tomography\n(18F -FDG PET/CT) is an advanced cancer diagnostic imaging technique that provides\nboth metabolic and structural information.\n\n\n\nIn cancer metastasis recognition tasks, effectively integrating metabolic and structural\ninformation stands as a key technology to enhance feature representation and recognition performance.\nThis paper proposes a cancer metastasis identification network based on dynamic\ncoordinated metabolic attention and structural attention to address these challenges. Specifically,\nmetabolic and structural features are extracted by incorporating a dynamic coordinated attention\nmodule (DCAM) into two branches of ResNet networks, thereby amalgamating high metabolic\nspatial information from PET images with texture structure information from CT images, and\ndynamically adjusting this process through iterations.\n\n\n\nNext, to improve the efficacy of feature expression, a multi-receptive field feature\nfusion module (MRFM) is included in order to execute multi-receptive field fusion of semantic\nfeatures.\n\n\n\nTo validate the effectiveness of our proposed model, experiments were conducted on\nboth a private lung lymph nodes dataset and a public soft tissue sarcomas dataset\n\n\n\nThe accuracy of our method reached 76.0% and 75.1% for the two datasets, respectively,\ndemonstrating an improvement of 6.8% and 5.6% compared to ResNet, thus affirming the\nefficacy of our method.\n","PeriodicalId":10881,"journal":{"name":"Current pharmaceutical biotechnology","volume":null,"pages":null},"PeriodicalIF":2.2000,"publicationDate":"2024-06-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Current pharmaceutical biotechnology","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.2174/0113892010302534240530073118","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"BIOCHEMISTRY & MOLECULAR BIOLOGY","Score":null,"Total":0}
引用次数: 0
Abstract
Cancer metastasis usually means that cancer cells spread to other tissues
or organs, and the condition worsens. Identifying whether cancer has metastasized can help
doctors infer the progression of a patient's condition and is an essential prerequisite for devising
treatment plans. Fluorine 18 fluorodeoxyglucose positron emission tomography/computed tomography
(18F -FDG PET/CT) is an advanced cancer diagnostic imaging technique that provides
both metabolic and structural information.
In cancer metastasis recognition tasks, effectively integrating metabolic and structural
information stands as a key technology to enhance feature representation and recognition performance.
This paper proposes a cancer metastasis identification network based on dynamic
coordinated metabolic attention and structural attention to address these challenges. Specifically,
metabolic and structural features are extracted by incorporating a dynamic coordinated attention
module (DCAM) into two branches of ResNet networks, thereby amalgamating high metabolic
spatial information from PET images with texture structure information from CT images, and
dynamically adjusting this process through iterations.
Next, to improve the efficacy of feature expression, a multi-receptive field feature
fusion module (MRFM) is included in order to execute multi-receptive field fusion of semantic
features.
To validate the effectiveness of our proposed model, experiments were conducted on
both a private lung lymph nodes dataset and a public soft tissue sarcomas dataset
The accuracy of our method reached 76.0% and 75.1% for the two datasets, respectively,
demonstrating an improvement of 6.8% and 5.6% compared to ResNet, thus affirming the
efficacy of our method.
期刊介绍:
Current Pharmaceutical Biotechnology aims to cover all the latest and outstanding developments in Pharmaceutical Biotechnology. Each issue of the journal includes timely in-depth reviews, original research articles and letters written by leaders in the field, covering a range of current topics in scientific areas of Pharmaceutical Biotechnology. Invited and unsolicited review articles are welcome. The journal encourages contributions describing research at the interface of drug discovery and pharmacological applications, involving in vitro investigations and pre-clinical or clinical studies. Scientific areas within the scope of the journal include pharmaceutical chemistry, biochemistry and genetics, molecular and cellular biology, and polymer and materials sciences as they relate to pharmaceutical science and biotechnology. In addition, the journal also considers comprehensive studies and research advances pertaining food chemistry with pharmaceutical implication. Areas of interest include:
DNA/protein engineering and processing
Synthetic biotechnology
Omics (genomics, proteomics, metabolomics and systems biology)
Therapeutic biotechnology (gene therapy, peptide inhibitors, enzymes)
Drug delivery and targeting
Nanobiotechnology
Molecular pharmaceutics and molecular pharmacology
Analytical biotechnology (biosensing, advanced technology for detection of bioanalytes)
Pharmacokinetics and pharmacodynamics
Applied Microbiology
Bioinformatics (computational biopharmaceutics and modeling)
Environmental biotechnology
Regenerative medicine (stem cells, tissue engineering and biomaterials)
Translational immunology (cell therapies, antibody engineering, xenotransplantation)
Industrial bioprocesses for drug production and development
Biosafety
Biotech ethics
Special Issues devoted to crucial topics, providing the latest comprehensive information on cutting-edge areas of research and technological advances, are welcome.
Current Pharmaceutical Biotechnology is an essential journal for academic, clinical, government and pharmaceutical scientists who wish to be kept informed and up-to-date with the latest and most important developments.