Lung Nodules
Conditions
Keywords
Generative AI, Lung puncture, Cone beam CT, DSA
Brief summary
The goal of this clinical trial is to learn if Generative AI Based Puncture Surgery Navigation System (GPS) can guide lung puncture in adults better. It will also learn about the quality improvement of cone beam CT (CBCT) by GPS. The main questions it aims to answer are: * Does GPS lower the number of punctures, radiation dose, and complications of participants undergo percutaneous lung puncture? * Does GPS improve the quality of CBCT images? Researchers will compare GPS to a conventional CBCT guided percutaneous lung puncture to see if GPS can improve the efficacy of lung puncture. Participants will: * Take the percutaneous lung puncture by the guidance of GPS or conventional CBCT (placebo) * The number of punctures, the success rate of the procedure, the radiation dose, intraoperative complications and postoperative complications at 7 days will be recorded * The images of CBCT will be collected
Interventions
COMBINATION_PRODUCTGenerative AI Based Puncture Surgery Navigation System
Participants will receive percutaneous lung puncture under the guidance of GPS.
DEVICECone beam CT
Participants will receive percutaneous lung puncture under the guidance of cone beam CT.
Sponsors
Union Hospital, Tongji Medical College, Huazhong University of Science and Technology
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
TREATMENT
Masking
TRIPLE (Subject, Investigator, Outcomes Assessor)
Eligibility
Sex/Gender
ALL
Age
18 Years to 85 Years
Healthy volunteers
No
Inclusion criteria
* Participants who require CBCT-guided precutaneous lung puncture (PLP) and meet the clinical indications for the procedure. * Participants with pre-operation CT images available. * Participants whose physical condition is suitable for PLP.
Exclusion criteria
* Participants have metallic implants in the body, which severely affects the image quality. * Participants with a history of allergy or serious adverse reactions to iodine contrast media or other related drugs. * Participants are pregnant or breastfeeding. * Participants are unwilling or unable to sign informed consent.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Number of puncture needles of participants for GPS and CBCT-guided lung puncture procedures. | From enrollment to the end of the lung puncture procedure. | The number of punctures was defined as the number of punctures performed throughout the percutaneous lung puncture procedure. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Radiation dose for GPS and CBCT-guided lung puncture procedures. | From enrollment to the end of the lung puncture procedure. | Radiation dose was defined as the radiation dose generated throughout the percutaneous lung puncture procedure. |
| Puncture time for GPS and CBCT-guided lung puncture procedures. | From enrollment to the end of the lung puncture procedure. | Puncture time was defined as the duration of the entire percutaneous lung puncture procedure. |
| Intraoperative and postoperative complications of participants for GPS and CBCT-guided lung puncture procedures. | From enrollment to the end of the lung puncture procedure at 1 week. | Intraoperative and postoperative complications were defined as those arising in parricipants during percutaneous lung puncture and within 7 days after puncture. |
Other
| Measure | Time frame | Description |
|---|---|---|
| Doctors' scoring of image and lesion quality | Through study completion, an average of 3 months. | Doctors' scoring of image and lesion quality were got based on the 5-point scale Likert (1-5 points). Rated 5 out of 5 for very good images, 4 out of 5 for good, 3 out of 5 for moderate, 2 out of 5 for poor, 1 out of 5 for very poor. |
| Algorithmic performance (peak signal-to-noise ratio) | Through study completion, an average of three months. | peak signal-to-noise ratio (PSNR) is a metric used to measure the quality of an image or video, assessing the degree of distortion by comparing the peak signal power to the mean square error (MSE) between the original signal and the compressed or processed signal. PSNR is measured in decibels (dB), with higher values indicating less distortion and better image or video quality. |
| Algorithmic performance (structural similarity) | Through study completion, an average of 3 months. | Structural similarity (SSIM) is a metric for evaluating the similarity of two images. SSIM value is between 0 and 1, the larger the value, the more similar the images are. |
Countries
China
Outcome results
None listed