Potential of Deep Learning in Assessing Pneumoconiosis Depicted on Digital Chest Radiography

NCT04963348 | Completed

• 1 other identifier: M2019467
• Study Type: observational
• Target: 1,881
• Locations: 0 countries
• Sites: N/A

Brief Summary

Pneumoconiosis is relatively prevalent in low/middle-income countries, and it remains a challenging task to accurately and reliably diagnose pneumoconiosis. The investigators implemented a deep learning solution and clarified the potential of deep learning in pneumoconiosis diagnosis by comparing its performance with two certified radiologists. The deep learning demonstrated a unique potential in classifying pneumoconiosis.

Conditions

Pneumoconiosis

Outcome Measures

Primary Outcomes (1)

• the diagnosis of pneumoconiosis

  The diagnosis and staging of pneumoconiosis were made by an expert panel consisting of certified radiologists and occupational physicians. The diagnosis of pneumoconiosis was confirmed by medical history and previous medical records(chest X-rays and pulmonary function testing).

  up to 6 months

Study Arms (1)

convolutional neural network (CNN)

a classical deep convolutional neural network (CNN) called Inception-V3 was applied to the image sets and validated the classification performance of the trained models

Other: convolutional neural networks (CNNs)

Interventions

convolutional neural networks (CNNs) OTHER

CNN architecture named U-Net architecture

Also known as: deep learning technology

convolutional neural network (CNN)

Eligibility Criteria

• Sex: all
• Healthy Volunteers: No
• Age Groups: Child (0-17), Adult (18-64), Older Adult (65+)
• Sampling Method: Non-Probability Sample

Study Population

Of these subjects, 923 were diagnosed with pneumoconiosis, 958 were normal. Among these subjects, 163 were females.

You may qualify if:

• industrial workers with a history of exposure to dust and underwent DR screening of pneumoconiosis from 2015 to 2018

You may not qualify if:

• patients with poor image quality
• patients with incomplete clinical data

Sponsors & Collaborators

• Peking University Third Hospital: lead

MeSH Terms

Conditions

Pneumoconiosis

Interventions

Convolutional Neural Networks

Condition Hierarchy (Ancestors)

Lung Diseases, Interstitial; Lung Diseases; Respiratory Tract Diseases; Lung Injury; Occupational Diseases

Intervention Hierarchy (Ancestors)

Neural Networks, Computer; Mathematical Concepts

Study Officials

• Xiaohua Wang

  Peking University Third Hospital

  STUDY CHAIR

Study Design

• Study Type: observational
• Observational Model: CASE ONLY
• Time Perspective: RETROSPECTIVE
• Sponsor Type: OTHER
• Responsible Party: SPONSOR

Study Record Dates

• First Submitted: June 28, 2021
• First Posted: July 15, 2021
• Study Start: January 1, 2015
• Primary Completion: December 31, 2018
• Study Completion: December 31, 2019
• Last Updated: July 15, 2021

Record last verified: 2021-06

Data Sharing

• IPD Sharing: Will not share