Identification of lesion location and discrimination between benign and malignant findings in thyroid ultrasound imaging
Abstract Thyroid nodules are a common thyroid disorder, and ultrasound imaging, as the primary diagnostic tool, is susceptible to variations based on the physician’s experience, leading to misdiagnosis. This paper constructs an end-to-end thyroid nodule detection framework based on YOLOv8, enabling...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-12-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-024-83888-1 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841559458343288832 |
|---|---|
| author | Xu Yang Hongliang Geng Xue Wang Lingxiao Li Xiaofeng An Zhibin Cong |
| author_facet | Xu Yang Hongliang Geng Xue Wang Lingxiao Li Xiaofeng An Zhibin Cong |
| author_sort | Xu Yang |
| collection | DOAJ |
| description | Abstract Thyroid nodules are a common thyroid disorder, and ultrasound imaging, as the primary diagnostic tool, is susceptible to variations based on the physician’s experience, leading to misdiagnosis. This paper constructs an end-to-end thyroid nodule detection framework based on YOLOv8, enabling automatic detection and classification of nodules by extracting grayscale and elastic features from ultrasound images. First, an attention-weighted DCN is introduced to enhance superficial feature extraction and capture local information. Next, the CPCA mechanism is employed to reduce the interference of redundant information. Finally, a feature fusion network based on an aggregation-distribution mechanism is utilized to improve the learning capability of fine-grained features, enhancing the performance of early nodule detection. Experimental results demonstrate that our method is accurate and effective for thyroid nodule detection, achieving diagnostic rates of 89.3% for benign and 90.4% for malignant nodules based on tests conducted on 611 clinical ultrasound images, with a mean Average Precision at IoU = 0.5 (mAP@50) of 95.5%, representing a 6.6% improvement over baseline models. |
| format | Article |
| id | doaj-art-a8e6aba85507415897ba60578c61b967 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-a8e6aba85507415897ba60578c61b9672025-01-05T12:28:04ZengNature PortfolioScientific Reports2045-23222024-12-0114111410.1038/s41598-024-83888-1Identification of lesion location and discrimination between benign and malignant findings in thyroid ultrasound imagingXu Yang0Hongliang Geng1Xue Wang2Lingxiao Li3Xiaofeng An4Zhibin Cong5School of Electronic Information Engineering, Changchun University of Science and TechnologySchool of Electronic Information Engineering, Changchun University of Science and TechnologyDepartment of Electrodiagnosis, The Affiliated Hospital to Changchun University of Traditional Chinese MedicineHuman Resources Department, The Third Affiliated Hospital of C.C.U.C.MEducation Quality Monitoring Center, Jilin Engineering Normal UniversityDepartment of Electrodiagnosis, The Affiliated Hospital to Changchun University of Traditional Chinese MedicineAbstract Thyroid nodules are a common thyroid disorder, and ultrasound imaging, as the primary diagnostic tool, is susceptible to variations based on the physician’s experience, leading to misdiagnosis. This paper constructs an end-to-end thyroid nodule detection framework based on YOLOv8, enabling automatic detection and classification of nodules by extracting grayscale and elastic features from ultrasound images. First, an attention-weighted DCN is introduced to enhance superficial feature extraction and capture local information. Next, the CPCA mechanism is employed to reduce the interference of redundant information. Finally, a feature fusion network based on an aggregation-distribution mechanism is utilized to improve the learning capability of fine-grained features, enhancing the performance of early nodule detection. Experimental results demonstrate that our method is accurate and effective for thyroid nodule detection, achieving diagnostic rates of 89.3% for benign and 90.4% for malignant nodules based on tests conducted on 611 clinical ultrasound images, with a mean Average Precision at IoU = 0.5 (mAP@50) of 95.5%, representing a 6.6% improvement over baseline models.https://doi.org/10.1038/s41598-024-83888-1Thyroid nodulesUltrasound imagingLesion classificationYOLOv8DCNCPCA mechanism |
| spellingShingle | Xu Yang Hongliang Geng Xue Wang Lingxiao Li Xiaofeng An Zhibin Cong Identification of lesion location and discrimination between benign and malignant findings in thyroid ultrasound imaging Scientific Reports Thyroid nodules Ultrasound imaging Lesion classification YOLOv8 DCN CPCA mechanism |
| title | Identification of lesion location and discrimination between benign and malignant findings in thyroid ultrasound imaging |
| title_full | Identification of lesion location and discrimination between benign and malignant findings in thyroid ultrasound imaging |
| title_fullStr | Identification of lesion location and discrimination between benign and malignant findings in thyroid ultrasound imaging |
| title_full_unstemmed | Identification of lesion location and discrimination between benign and malignant findings in thyroid ultrasound imaging |
| title_short | Identification of lesion location and discrimination between benign and malignant findings in thyroid ultrasound imaging |
| title_sort | identification of lesion location and discrimination between benign and malignant findings in thyroid ultrasound imaging |
| topic | Thyroid nodules Ultrasound imaging Lesion classification YOLOv8 DCN CPCA mechanism |
| url | https://doi.org/10.1038/s41598-024-83888-1 |
| work_keys_str_mv | AT xuyang identificationoflesionlocationanddiscriminationbetweenbenignandmalignantfindingsinthyroidultrasoundimaging AT honglianggeng identificationoflesionlocationanddiscriminationbetweenbenignandmalignantfindingsinthyroidultrasoundimaging AT xuewang identificationoflesionlocationanddiscriminationbetweenbenignandmalignantfindingsinthyroidultrasoundimaging AT lingxiaoli identificationoflesionlocationanddiscriminationbetweenbenignandmalignantfindingsinthyroidultrasoundimaging AT xiaofengan identificationoflesionlocationanddiscriminationbetweenbenignandmalignantfindingsinthyroidultrasoundimaging AT zhibincong identificationoflesionlocationanddiscriminationbetweenbenignandmalignantfindingsinthyroidultrasoundimaging |