Ultrasound Radiomics for Predicting Breast Cancer and Axillary Lymph Node Metastasis

Sponsor

The Affiliated Nanjing Drum Tower Hospital of Nanjing University Medical School (Other)

Overall Status

Not yet recruiting

CT.gov ID

NCT05768451

Collaborator

(none)

200

Enrollment

Anticipated Duration (Months)

Study Details

Study Description

Brief Summary

The project is proposed based on multimodal ultrasonic imaging omics building used for accurate prediction of the breast cancer and axillary lymph node metastasis load artificial intelligence forecasting model, this method can dig the hidden features of ultrasonic image is not visible to the naked eye, make up the subjectivity in the process of clinical doctors in diagnosis and treatment, provide accurate, objective basis for clinical decision making.

Condition or Disease	Intervention/Treatment	Phase
Breast Cancer Axillary Lymph Node Metastasis	Other: Ultrasonic image analysis

Study Design

Study Type:

Observational

Anticipated Enrollment :

200 participants

Observational Model:

Case-Control

Time Perspective:

Retrospective

Official Title:

Accurate Prediction of Breast Cancer and Axillary Lymph Node Metastasis Burden by Multi-modal Ultrasound-based Radiomics

Anticipated Study Start Date :

Apr 1, 2023

Anticipated Primary Completion Date :

Dec 31, 2024

Anticipated Study Completion Date :

Aug 31, 2025

Arms and Interventions

Arm	Intervention/Treatment
benign	Other: Ultrasonic image analysis Preoperative conventional ultrasound (US), elastic ultrasound (UE) and contrast-enhanced ultrasound (CEUS) images were analyzed. Histopathological results were used as the gold standard. The cases were randomly divided into training set and test set with a ratio of 7:3. The US image, UE image and CEUS image of the maximum long-axis section of each lesion were selected, and the region of interest (ROI) of the lesion was manually delineated.
malignant	Other: Ultrasonic image analysis Preoperative conventional ultrasound (US), elastic ultrasound (UE) and contrast-enhanced ultrasound (CEUS) images were analyzed. Histopathological results were used as the gold standard. The cases were randomly divided into training set and test set with a ratio of 7:3. The US image, UE image and CEUS image of the maximum long-axis section of each lesion were selected, and the region of interest (ROI) of the lesion was manually delineated.

Arm

Intervention/Treatment

benign

Other: Ultrasonic image analysis

Preoperative conventional ultrasound (US), elastic ultrasound (UE) and contrast-enhanced ultrasound (CEUS) images were analyzed. Histopathological results were used as the gold standard. The cases were randomly divided into training set and test set with a ratio of 7:3. The US image, UE image and CEUS image of the maximum long-axis section of each lesion were selected, and the region of interest (ROI) of the lesion was manually delineated.

malignant

Other: Ultrasonic image analysis

Outcome Measures

Primary Outcome Measures

Accuracy of differential diagnosis between benign and malignant [2022.12.15-2023.12.30]
ROC curve, sensitivity, specificity, accuracy, decision curve

Eligibility Criteria

Criteria

Ages Eligible for Study:

18 Years to 90 Years

Sexes Eligible for Study:

Female

Accepts Healthy Volunteers:

Inclusion Criteria:

patients who have not received breast puncture biopsy or neoadjuvant chemotherapy, etc. prior to ultrasonography;
patients who are not allergic to the ultrasound contrast agent (SonoVue) and have signed an informed consent form for ultrasonography
patients with breast tumors who have undergone surgical treatment at our hospital and have complete case histories.

Exclusion Criteria:

patients with incomplete case data;
cases with large breast masses that do not completely show the region of interest (ROI);
patients who have undergone breast aspiration biopsy or other treatments such as neoadjuvant chemotherapy or endocrine therapy before ultrasound examination.