Lung Disease, Lung Ultrasound
Conditions
Keywords
Lung Ultrasound, Artificial Intelligence
Brief summary
This is a prospective, non-randomized clinical validation research study. Subjects will consent and have two ultrasounds as part of the study.
Detailed description
This study is to collect data to validate the performance of Caption Lung AI in enabling trained healthcare professionals to perform LUS to obtain diagnostic-quality Lung Ultrasound Studies.
Interventions
DEVICELung ultrasound
Lung ultrasound software.
Sponsors
Northwestern University
Rush University Medical Center
Yale University
GE Healthcare
Study design
Allocation
NON_RANDOMIZED
Intervention model
PARALLEL
Primary purpose
HEALTH_SERVICES_RESEARCH
Masking
NONE
Eligibility
Sex/Gender
ALL
Age
22 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* 1\. Patients over the age of 22; AND * 2\. Patients with known lung pathology or patients with no known lung pathology in the hospital or presenting to the hospital or outpatient setting with shortness of breath.
Exclusion criteria
* 1\. Patients in extremis and/or patients in whom a lung ultrasound would not normally be performed; * 2\. Patients who are unable to consent; * 3\. Patients who are prisoners; AND * 4\. Patients who do not speak English fluently.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Cohort 1: % of patient studies composed of images acquired by the THCPs with sufficient quality to make a clinical assessment | 7 months | The primary endpoint will be based on the majority assessment of the independent expert readers qualitatively assessment of the LUS exam (study level) as being of diagnostic quality. |
| Cohort 2: % of patient studies composed of images acquired by the Qualified user to make a clinical assessment | 7 months | The primary endpoint will be based on the majority assessment of the independent expert readers qualitatively assessment of the LUS exam (study level) as being of diagnostic quality. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Cohort 1: | 7 months | Sub-group analyses will be performed on the diagnostic image quality performance, with descriptive statistics reported based on age (\< 65, ≥ 65), BMI group (\< 25, 25 ≤ BMI \< 30, ≥ 30), gender (Male / Female), site location, THCP, and zone of the image (1 - 8). The statistical analysis procedure and all applicable performance metrics that were used on the whole sample (i.e., the Primary Outcome Measures) will be repeated for each sub-group level. This will be conducted to ensure that there is no inherent bias where a specific sub-group level has a greater impact on the overall whole-sample performance compared to other levels. |
| Cohort 2: | 7 months | Sub-group analyses will be performed on the diagnostic image quality performance, with descriptive statistics reported based on age (\< 65, ≥ 65), BMI group (\< 25, 25 ≤ BMI \< 30, ≥ 30), gender (Male / Female), site location, THCP, and zone of the image (1 -12). The statistical analysis procedure and all applicable performance metrics that were used on the whole sample (i.e., the Primary Outcome Measures) will be repeated for each sub-group level. This will be conducted to ensure that there is no inherent bias where a specific sub-group level has a greater impact on the overall whole-sample performance compared to other levels. |
| Cohort 1 and 2: | 7 months | Recording mode (% of clips acquired with auto-capture rate versus save best clip). |
Countries
United States
Outcome results
None listed