D-Lung: An Analytics Platform for Lung Cancer Based on Deep Learning Technology

D-Lung: An Analytics Platform for Primary Lung Cancer Screening, Diagnosis and Management Based on Deep Learning Technology

Lung cancer is one of main cause of cancer death in worldwide, characterized of low 5-year survival rate of less than 20%. Pulmonary nodule is considered as the typical imaging manifestation in early stage of lung cancer. The National Lung Screen Trial has demonstrated that the mortality rates could decline greatly, by the utility of low-dose helical computed tomography for screen of pulmonary nodules. Thus, automatic detection, diagnosis and management of pulmonary nodules, play the vital roles in computer-aided lung cancer screening and early intervention.

Study Overview

• Status: Completed
• Conditions: Lung Cancer
• Intervention / Treatment: Radiation: computed tomography

• Study Type: Observational
• Enrollment (Actual): 130

Contacts and Locations

Study Locations

• Hong Kong
  • Shatin
    • Hong Kong, Shatin, Hong Kong
      • The Chinese University of Hong Kong, Prince of Wale Hospital

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: ADULT; OLDER_ADULT; CHILD
• Accepts Healthy Volunteers: N/A
• Genders Eligible for Study: All

• Sampling Method: Probability Sample

Study Population

This is a single institutional retrospective cohort study of patients within hospitals in Hong Kong, who had undergone thoracic CT for suspicious lung nodules.

Description

Inclusion Criteria:

• Subjects with suspicious lung nodules.
• Thin-layer thoracic CT and pathology examination have been performed for suspicious lung nodules.

Exclusion Criteria:

• Subjects with accompanied lesions on CT images that may interfere to lung nodules analysis

Study Plan

How is the study designed?

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
accuracy	proportion of true results(both true positives and true negatives) among whole instances	2 years
sensitivity	true positive rate in percentage(%) derived by ROC analysis	2 years
specificity	true negative rate in percentage (%) derived by ROC analysis	2 years
area under curve (AUC)	area under ROC curve in percentage (%)	2 years

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
average number of false positives per scan (FPs/scan)	FPs/scan in number (N) based on free-response receiver operating characteristic (FROC) analysis	2 years
competition performance metric (CPM)	Competitive performance metric (CPM) is a criterion used for CAD system evaluation. Based on FROC paradigm, CPM score is computed as an average sensitivity at seven predefined average false positive rates. CPM score ranges from 0 to 1, with higher CPM score indicating better CAD performance.	2 years

Collaborators and Investigators

• Sponsor: Chinese University of Hong Kong
• Collaborators: Department of Computer Science & Engineering, CUHK

Study record dates

Study Major Dates

• Study Start (ACTUAL): July 1, 2018
• Primary Completion (ACTUAL): June 30, 2020
• Study Completion (ACTUAL): June 30, 2020

Study Registration Dates

• First Submitted: July 26, 2019
• First Submitted That Met QC Criteria: July 26, 2019
• First Posted (ACTUAL): July 30, 2019

Study Record Updates

• Last Update Posted (ACTUAL): February 8, 2023
• Last Update Submitted That Met QC Criteria: February 7, 2023
• Last Verified: February 1, 2023

More Information

Terms related to this study

• Keywords: Deep learning; Lung nodules
• Additional Relevant MeSH Terms: Respiratory Tract Diseases; Neoplasms; Lung Diseases; Neoplasms by Site; Respiratory Tract Neoplasms; Thoracic Neoplasms; Lung Neoplasms
• Other Study ID Numbers: 2019.306

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No