Utility of Ultrasound Imaging for Diagnosis of Focal Liver Lesions: A Radiomics Analysis

NCT03871140 | Unknown

• 1 other identifier: 301jrcsk3
• Study Type: observational
• Target: 10,000
• Locations: 1 country
• Sites: 1

Brief Summary

Ultrasound (US) as first-line imaging technology in detecting focal liver lesions,also plays a crucial role in evaluating image and guiding ablation which is the main treatment for liver lesions. However, the effect of US in diagnosing liver lesions is challenged by several factors including being highly dependent on doctor's experience, low signal-to-noise ratio, low resolution for lesion feature,large error from thermal field evaluation during the process of ablation and so on. Therefore, it is of great significance to construct an intelligent US analysis system depending on the digital information technology. Basing on these problems,the following research will be involved in our project: 1) US database of liver lesions with seamless connection to Picture Archiving and Communication Systems (PACS) will be developed, with the aim to provide standard data for intelligent US analysis. 2) Deep learning model for accurate segmentation, detection and classification of liver lesions on US images will be studied. Then automatic extraction, selection and analysis of liver lesion ultrasound features and the intelligent US diagnosis for liver lesions will be realized. 3) Proposing a clustering model with deep image features, and depicting the similarity measurement of liver cancer, which can be furthered used to link the liver cancer feature to optimal ablation parameters. The intelligent decision-making system for quantifying thermal ablation will be established. 4) Regression algorithm and Generative Adversarial Nets will be developed to extract the image features of liver cancer which will predict risk factors after US-guided thermal ablation.Based on the above researches, it is of great value to establish an intelligent focal liver lesion US diagnosis system involving intelligent diagnosis,personalized ablation strategy and accurate prognosis evaluation, improving the level of accurate diagnosis and treatment of liver lesions.

Conditions

Focal Liver Lesions; Ultrasonics; Intelligent Diagnosis; Ablation; Radiomics

Outcome Measures

Primary Outcomes (3)

• AUC value

  Area under the receiver operating characteristic (ROC) curve (AUC）

  through study completion, an average of 3 year

• specificity

  diagnosis specificity of intelligent ultrasound analysis

  through study completion, an average of 3 year

• sensitivity

  diagnosis sensitivity of intelligent ultrasound analysis

  through study completion, an average of 3 year

Interventions

diagnosis OTHER

therre is no intervention diagnosis or treatment for patients

Eligibility Criteria

• Sex: all
• Healthy Volunteers: Yes
• Age Groups: Child (0-17), Adult (18-64), Older Adult (65+)
• Sampling Method: Probability Sample

Study Population

patients with focal liver lesions

You may qualify if:

• clear ultrasound imaging of focal liver lesions including malignant liver tumors such as hepatocellular carcinoma, metastatic liver cancer and benigh liver tumors such as hemangioma and focal nodular hyperplasia and so on can be acquired.
• clear ultrasound imaging of liver tissues backgroud without lesions can be acquired.
• disease history and pathological diagnosis of the lesions can be acquired.

You may not qualify if:

• patients unsuitable for ultrasound san
• patients counldn't provide disease history such as hepatitis, alcohol intake and so on
• patients without pathological results

Sponsors & Collaborators

• Chinese PLA General Hospital: lead

Study Sites (1)

Chinese PLA General Hospital

Beijing, Beijing Municipality, 100853, China

RECRUITING

Related Publications (2)

• Du Z, Fan F, Ma J, Liu J, Yan X, Chen X, Dong Y, Wu J, Ding W, Zhao Q, Wang Y, Zhang G, Yu J, Liang P. Development and validation of an ultrasound-based interpretable machine learning model for the classification of </=3 cm hepatocellular carcinoma: a multicentre retrospective diagnostic study. EClinicalMedicine. 2025 Feb 13;81:103098. doi: 10.1016/j.eclinm.2025.103098. eCollection 2025 Mar.

  PMID: 40034568 DERIVED

• Yang Y, Cairang Y, Jiang T, Zhou J, Zhang L, Qi B, Ma S, Tang L, Xu D, Bu L, Bu R, Jing X, Wang H, Zhou Z, Zhao C, Luo B, Liu L, Guo J, Nima Y, Hua G, Wa Z, Zhang Y, Zhou G, Jiang W, Wang C, De Y, Yu X, Cheng Z, Han Z, Liu F, Dou J, Feng H, Wu C, Wang R, Hu J, Yang Q, Luo Y, Wu J, Fan H, Liang P, Yu J. Ultrasound identification of hepatic echinococcosis using a deep convolutional neural network model in China: a retrospective, large-scale, multicentre, diagnostic accuracy study. Lancet Digit Health. 2023 Aug;5(8):e503-e514. doi: 10.1016/S2589-7500(23)00091-2.

  PMID: 37507196 DERIVED

Central Study Contacts

Jie Yu, Dr

CONTACT

15901417963; jiemi301@163.com

Study Design

• Study Type: observational
• Observational Model: COHORT
• Time Perspective: PROSPECTIVE
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Prof

Study Record Dates

• First Submitted: April 29, 2018
• First Posted: March 12, 2019
• Study Start: January 1, 2017
• Primary Completion: December 30, 2020
• Study Completion: December 30, 2021
• Last Updated: March 12, 2019

Record last verified: 2018-04

Locations

CN (1)