MedPath

Effectiveness of Ultra-low-dose Chest CT With AI Based Denoising Solution

Not Applicable
Conditions
Lung Diseases
Interventions
Radiation: Low radiation dose CT
Radiation: Underwent ultra dose chest CT
Other: Artificial Intelligence based model
Registration Number
NCT05398887
Lead Sponsor
Intermed Hospital
Brief Summary

The main objective of the study is to evaluate the detection rate of pulmonary conditions, percentage of ionizing radiation dose reduction, and state of image quality of ULDCT coupling with innovative vendor-neutral CT denoising solution based on deep learning technology.

Detailed Description

Considering lung cancer-related public health challenges, a reliable lung cancer screening method for high-risk cohorts in Mongolia is needed. Thus, our study aims to assess the detection rate of pulmonary conditions, percentage of ionizing radiation dose reduction, and state of image quality of ULDCT coupling with artificial intelligence based CT denoising technique among various patient groups.

Recruitment & Eligibility

Status
UNKNOWN
Sex
All
Target Recruitment
200
Inclusion Criteria
  • Patients aged over 18-year-old
  • Patients undergoing CT Chest for all purpose
Exclusion Criteria
  • Age less than 18 years
  • Any suspicion of pregnancy
  • History of thoracic surgery or placement of the metallic device in the thorax
  • An inability to hold respiration during CT

Study & Design

Study Type
INTERVENTIONAL
Study Design
PARALLEL
Arm && Interventions
GroupInterventionDescription
Low dose Chest CT scanLow radiation dose CTUnderwent low dose chest CT with 30% lower radiation dose Interventions: Radiation: Low radiation dose CT Other: Image quality analysis
Ultra low dose CT scan with Artificial IntelligenceUnderwent ultra dose chest CTInterventions: Radiation: Low radiation dose CT Image quality Other: Deep-learning based contrast boosting algorithms
Ultra low dose CT scan with Artificial IntelligenceArtificial Intelligence based modelInterventions: Radiation: Low radiation dose CT Image quality Other: Deep-learning based contrast boosting algorithms
Primary Outcome Measures
NameTimeMethod
Detection rate of pulmonary conditionsWithin 2 weeks after data collection

Pulmonary condition detection rate on low dose chest CT and ultra dose chest CT with artificial intelligence-based CT denoising solution by blinded reviewers

Contrast media doseWithin 2 weeks after data collection

Administered contrast media dose in each patient

Secondary Outcome Measures
NameTimeMethod
Image contrastWithin 2 weeks after data collection

Signal to Noise, Noise and Edge-rise-distance on a five-point scale (1-5) with a higher score indicates better conspicuity.

© Copyright 2025. All Rights Reserved by MedPath