Deep learning models of ultrasonography significantly improved the differential diagnosis performance for superficial soft-tissue masses: a retrospective multicenter study
Abstract Background Most of superficial soft-tissue masses are benign tumors, and very few are malignant tumors. However, persistent growth, of both benign and malignant tumors, can be painful and even life-threatening. It is necessary to improve the differential diagnosis performance for superficia...
Main Authors: | , , , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
BMC
2023-10-01
|
Series: | BMC Medicine |
Subjects: | |
Online Access: | https://doi.org/10.1186/s12916-023-03099-9 |
_version_ | 1797559097008062464 |
---|---|
author | Bin Long Haoyan Zhang Han Zhang Wen Chen Yang Sun Rui Tang Yuxuan Lin Qiang Fu Xin Yang Ligang Cui Kun Wang |
author_facet | Bin Long Haoyan Zhang Han Zhang Wen Chen Yang Sun Rui Tang Yuxuan Lin Qiang Fu Xin Yang Ligang Cui Kun Wang |
author_sort | Bin Long |
collection | DOAJ |
description | Abstract Background Most of superficial soft-tissue masses are benign tumors, and very few are malignant tumors. However, persistent growth, of both benign and malignant tumors, can be painful and even life-threatening. It is necessary to improve the differential diagnosis performance for superficial soft-tissue masses by using deep learning models. This study aimed to propose a new ultrasonic deep learning model (DLM) system for the differential diagnosis of superficial soft-tissue masses. Methods Between January 2015 and December 2022, data for 1615 patients with superficial soft-tissue masses were retrospectively collected. Two experienced radiologists (radiologists 1 and 2 with 8 and 30 years’ experience, respectively) analyzed the ultrasound images of each superficial soft-tissue mass and made a diagnosis of malignant mass or one of the five most common benign masses. After referring to the DLM results, they re-evaluated the diagnoses. The diagnostic performance and concerns of the radiologists were analyzed before and after referring to the results of the DLM results. Results In the validation cohort, DLM-1 was trained to distinguish between benign and malignant masses, with an AUC of 0.992 (95% CI: 0.980, 1.0) and an ACC of 0.987 (95% CI: 0.968, 1.0). DLM-2 was trained to classify the five most common benign masses (lipomyoma, hemangioma, neurinoma, epidermal cyst, and calcifying epithelioma) with AUCs of 0.986, 0.993, 0.944, 0.973, and 0.903, respectively. In addition, under the condition of the DLM-assisted diagnosis, the radiologists greatly improved their accuracy of differential diagnosis between benign and malignant tumors. Conclusions The proposed DLM system has high clinical application value in the differential diagnosis of superficial soft-tissue masses. |
first_indexed | 2024-03-10T17:40:47Z |
format | Article |
id | doaj.art-6533251ea2f247bc91f0b4786d7bec49 |
institution | Directory Open Access Journal |
issn | 1741-7015 |
language | English |
last_indexed | 2024-03-10T17:40:47Z |
publishDate | 2023-10-01 |
publisher | BMC |
record_format | Article |
series | BMC Medicine |
spelling | doaj.art-6533251ea2f247bc91f0b4786d7bec492023-11-20T09:41:26ZengBMCBMC Medicine1741-70152023-10-0121111110.1186/s12916-023-03099-9Deep learning models of ultrasonography significantly improved the differential diagnosis performance for superficial soft-tissue masses: a retrospective multicenter studyBin Long0Haoyan Zhang1Han Zhang2Wen Chen3Yang Sun4Rui Tang5Yuxuan Lin6Qiang Fu7Xin Yang8Ligang Cui9Kun Wang10Institute of Medical Technology, Peking University Health Science CenterCAS Key Laboratory of Molecular Imaging, Institute of Automation, Chinese Academy of SciencesDepartment of Ultrasound, The Second Hospital of Hebei Medical UniversityDepartment of Diagnostic Ultrasound, Peking University Third HospitalDepartment of Diagnostic Ultrasound, Peking University Third HospitalInstitute of Medical Technology, Peking University Health Science CenterDepartment of Ultrasound, Beijing Friendship Hospital, Capital Medical UniversityDepartment of Ultrasound, Beijing Civil Aviation General HospitalCAS Key Laboratory of Molecular Imaging, Institute of Automation, Chinese Academy of SciencesInstitute of Medical Technology, Peking University Health Science CenterCAS Key Laboratory of Molecular Imaging, Institute of Automation, Chinese Academy of SciencesAbstract Background Most of superficial soft-tissue masses are benign tumors, and very few are malignant tumors. However, persistent growth, of both benign and malignant tumors, can be painful and even life-threatening. It is necessary to improve the differential diagnosis performance for superficial soft-tissue masses by using deep learning models. This study aimed to propose a new ultrasonic deep learning model (DLM) system for the differential diagnosis of superficial soft-tissue masses. Methods Between January 2015 and December 2022, data for 1615 patients with superficial soft-tissue masses were retrospectively collected. Two experienced radiologists (radiologists 1 and 2 with 8 and 30 years’ experience, respectively) analyzed the ultrasound images of each superficial soft-tissue mass and made a diagnosis of malignant mass or one of the five most common benign masses. After referring to the DLM results, they re-evaluated the diagnoses. The diagnostic performance and concerns of the radiologists were analyzed before and after referring to the results of the DLM results. Results In the validation cohort, DLM-1 was trained to distinguish between benign and malignant masses, with an AUC of 0.992 (95% CI: 0.980, 1.0) and an ACC of 0.987 (95% CI: 0.968, 1.0). DLM-2 was trained to classify the five most common benign masses (lipomyoma, hemangioma, neurinoma, epidermal cyst, and calcifying epithelioma) with AUCs of 0.986, 0.993, 0.944, 0.973, and 0.903, respectively. In addition, under the condition of the DLM-assisted diagnosis, the radiologists greatly improved their accuracy of differential diagnosis between benign and malignant tumors. Conclusions The proposed DLM system has high clinical application value in the differential diagnosis of superficial soft-tissue masses.https://doi.org/10.1186/s12916-023-03099-9Superficial soft-tissue massesDeep learning modelUltrasoundDiagnosisComputer-assisted diagnosis |
spellingShingle | Bin Long Haoyan Zhang Han Zhang Wen Chen Yang Sun Rui Tang Yuxuan Lin Qiang Fu Xin Yang Ligang Cui Kun Wang Deep learning models of ultrasonography significantly improved the differential diagnosis performance for superficial soft-tissue masses: a retrospective multicenter study BMC Medicine Superficial soft-tissue masses Deep learning model Ultrasound Diagnosis Computer-assisted diagnosis |
title | Deep learning models of ultrasonography significantly improved the differential diagnosis performance for superficial soft-tissue masses: a retrospective multicenter study |
title_full | Deep learning models of ultrasonography significantly improved the differential diagnosis performance for superficial soft-tissue masses: a retrospective multicenter study |
title_fullStr | Deep learning models of ultrasonography significantly improved the differential diagnosis performance for superficial soft-tissue masses: a retrospective multicenter study |
title_full_unstemmed | Deep learning models of ultrasonography significantly improved the differential diagnosis performance for superficial soft-tissue masses: a retrospective multicenter study |
title_short | Deep learning models of ultrasonography significantly improved the differential diagnosis performance for superficial soft-tissue masses: a retrospective multicenter study |
title_sort | deep learning models of ultrasonography significantly improved the differential diagnosis performance for superficial soft tissue masses a retrospective multicenter study |
topic | Superficial soft-tissue masses Deep learning model Ultrasound Diagnosis Computer-assisted diagnosis |
url | https://doi.org/10.1186/s12916-023-03099-9 |
work_keys_str_mv | AT binlong deeplearningmodelsofultrasonographysignificantlyimprovedthedifferentialdiagnosisperformanceforsuperficialsofttissuemassesaretrospectivemulticenterstudy AT haoyanzhang deeplearningmodelsofultrasonographysignificantlyimprovedthedifferentialdiagnosisperformanceforsuperficialsofttissuemassesaretrospectivemulticenterstudy AT hanzhang deeplearningmodelsofultrasonographysignificantlyimprovedthedifferentialdiagnosisperformanceforsuperficialsofttissuemassesaretrospectivemulticenterstudy AT wenchen deeplearningmodelsofultrasonographysignificantlyimprovedthedifferentialdiagnosisperformanceforsuperficialsofttissuemassesaretrospectivemulticenterstudy AT yangsun deeplearningmodelsofultrasonographysignificantlyimprovedthedifferentialdiagnosisperformanceforsuperficialsofttissuemassesaretrospectivemulticenterstudy AT ruitang deeplearningmodelsofultrasonographysignificantlyimprovedthedifferentialdiagnosisperformanceforsuperficialsofttissuemassesaretrospectivemulticenterstudy AT yuxuanlin deeplearningmodelsofultrasonographysignificantlyimprovedthedifferentialdiagnosisperformanceforsuperficialsofttissuemassesaretrospectivemulticenterstudy AT qiangfu deeplearningmodelsofultrasonographysignificantlyimprovedthedifferentialdiagnosisperformanceforsuperficialsofttissuemassesaretrospectivemulticenterstudy AT xinyang deeplearningmodelsofultrasonographysignificantlyimprovedthedifferentialdiagnosisperformanceforsuperficialsofttissuemassesaretrospectivemulticenterstudy AT ligangcui deeplearningmodelsofultrasonographysignificantlyimprovedthedifferentialdiagnosisperformanceforsuperficialsofttissuemassesaretrospectivemulticenterstudy AT kunwang deeplearningmodelsofultrasonographysignificantlyimprovedthedifferentialdiagnosisperformanceforsuperficialsofttissuemassesaretrospectivemulticenterstudy |