AI-based System for Assessing Suspected Viral Pneumonia Related Lung Changes

Launched by RESEARCH AND PRACTICAL CLINICAL CENTER FOR DIAGNOSTICS AND TELEMEDICINE TECHNOLOGIES OF THE MOSCOW HEALTH CARE DEPARTMENT · Jul 9, 2024

Keywords

ClinConnect Summary

This clinical trial is studying an AI-based system designed to help doctors analyze chest CT scans for signs of viral pneumonia, particularly related to COVID-19. The goal is to see how well this system can detect abnormal patterns in the lungs and provide useful information to doctors. By comparing the AI's results to standard measures of accuracy, the researchers hope to confirm that this technology works effectively in identifying different levels of pneumonia severity.

To be eligible for this trial, participants must be at least 18 years old and have undergone a specific type of chest CT scan without contrast. The scan must meet certain quality standards to ensure accurate results. Patients who have clear signs of COVID-19 pneumonia on their scans, as well as those without any lung issues related to COVID-19, may be included. Participants will not only contribute to important research but also receive insights into their lung health as part of the study. If you're considering joining, it's a chance to be part of an innovative project that could improve how pneumonia is diagnosed in the future.

Gender

ALL

Eligibility criteria

• Inclusion Criteria:
• • 1. General
• • 1. Patients over 18 years old;
• • 2. Patients who underwent CT without contrast enhancement;
• • 3. Patients who underwent a CT scan according to a standardized scanning protocol: 120 kilovolts, slice thickness max. 2 mm, rigid "lung" filter (kernel) reconstruction;
• • 4. Patients whose studies should be of acceptable quality, performed with breath-holding, without technical artifacts, and respiratory and motor artifacts;
• • 5. Patients whose studies must contain DICOM tags responsible for the orientation and position of the patient in the images during the study, as well as DICOM tags responsible for the size of the scans and image parameters;
• • 6. Patients in whom the localization of changes is predominantly bilateral, in the basal and subpleural parts of the lungs, may be located peribronchial;
• • 2. For group Normal
• • a. Patients who do not contain COVID-19-related CT patterns;
• • 3. For groups Mild, Moderate, Severe, and Critical
• • 1. Patients who contain COVID-19-related CT pattern: ground glass opacities (mild, moderate, and higher intensity);
• • 2. Patients who contain COVID-19-related CT pattern: pulmonary consolidation;
• • 3. Patients who contain COVID-19-related CT pattern: cobblestone infiltration of the lung parenchyma;
• • 4. Patients who contain COVID-19-related CT pattern: hydrothorax;
• • 5. Patients who contain a combination of one or more patterns.
• Exclusion Criteria:
• • Patients whose studies contain images with unreported CT patterns;
• • Patients whose examinations do not conform to DICOM format;
• • Patients whose examinations do not contain imaging of the lung region
• • Patients whose examinations contain technical artifacts caused by malfunctions or features of CT scanners;
• • Patients whose examinations contain improper patient positioning;
• • Patients whose examinations contain studies with deleted DICOM tags responsible for scan size and image parameters;
• • Patients whose examinations contain metal artifacts on the patient's body and clothing;
• • Patients whose examinations contain the presence of other pathologic changes of lungs in patients - neoplastic, tuberculosis process, bacterial pneumonia, etc.;
• • Patients under 18 years old.