Kosmos Anatomical Object Labeling and View Identification Pivotal Study

This is a multi-site, multi-group study that obtained two sets of ultrasound scans from all enrolled participants. Participants underwent limited ultrasound scanning by both cardiac sonographers and abdominal sonographers. Algorithm's performance on the view identification task and the object labeling task is computed.

Completed

Looking for future studies?

Notify Me

Key information

Conditions

Cardiac Disorder Cardiovascular Diseases Digestive System Diseases Female Urogenital Diseases Female Urogenital Diseases and Pregnancy Complications GYN Disorders Genital Diseases Genital Diseases, Female Heart Diseases Hemic and Lymphatic Diseases Hepatic Disorder Hepatobiliary Disorders Kidney Diseases Kidney Disorder Lung Diseases Lymphatic Diseases Male Urogenital Diseases Pulmonary Disorder Renal Disorder Respiratory Tract Diseases Spleen Disorder Splenic Diseases Urogenital Diseases Urologic Diseases

Age range

18 year–89 year

Sex eligibility

All sexes

Study type

Interventional

Phase

Not applicable

Primary location

REVIVAL research institute, LLC

Sherman, Texas, 75092, United States

Who can participate

Healthy volunteers accepted: Yes

Only the study team can determine whether someone qualifies for participation.

Inclusion criteria

Persons able to:
Read and sign an English consent form.
Give consent for participation.
Able and willing to comply with study requirements.
Those aged 18 years through 89 years, healthy as well as individuals with pathology, including but not limited to:
Persons with heart-related conditions such as myocardial or pericardial disease.
Persons with lung-related conditions such as: asthma, COPD, bronchitis, pulmonary fibrosis, sarcoidosis, pulmonary hypertension, pulmonary hypertension, bronchiectasis, lung cancer, pneumonia, pulmonary edema, pulmonary embolism
Persons with abdomen-related conditions such as: pancreatitis, gastroesophageal reflux disease, irritable bowel syndrome, colitis, gastroenteritis, ulcers, abdominal aortic aneurysm, splenomegaly, liver disease, kidney disease.

Exclusion criteria

Children (minors) under 18 years old
Adults over 89 years old
Pregnant individuals
Those who cannot or refuse to sign their consent
Those who cannot provide informed consent
Those who cannot speak or read English
Vulnerable populations, i.e., individuals whose willingness to participate in the study could be unduly influenced by the expectation of medical benefits associated with participation
Individuals with mobility issues that prevent them from having an echo examination
Individuals with severe chest deformities, or other critical clinical situations, e.g., life critical urgent situations where there is no time for anything else than what is needed for patient care, that prevent them from having an echo examination.

Treatment and study plan

Kosmos Anatomical Object Labeling and View Identification Algorithms

Device

Participants underwent limited ultrasound scanning by both cardiac sonographers and abdominal sonographers. Algorithm's performance on the view identification task and the object labeling task is compared to majority opinion of 5 radiologists.

Primary outcomes

False discovery rate of the anatomical object labeling function on expert-obtained images

Time frame: Post data acquisition (3 months)

The objective of this study is to compare the average frame-level disagreement of object labeling by majority opinion of five expert reviewers. The frame-level disagreement is defined as False Discovery Rate as in (1) FDR = 1-Precision=FP/(FP+TP) (1) Where TP≡True Positive and FP≡False Positive. A True Positive is counted when the majority opinion of five expert reviewers agrees with the location of predicted labels in the image. A False Positive counted when majority opinion of five expert reviewers disagrees with the location of one or more predicted labels.

False Discovery Rate (FDR) is measured for a particular image frame between predicted objects and the majority opinion of five reviewers. The frame-level FDR threshold for this prospective study is 20%. Therefore, for a successful outcome the frame-level FDR must be less than 20%.
False discovery rate of the anatomical view identification function on expert-obtained images

Time frame: Post data acquisition (3 months)

The objective of this study is to compare the average frame-level disagreement of object labeling by majority opinion of five expert reviewers. The frame-level disagreement is defined as False Discovery Rate as in (1) FDR = 1-Precision=FP/(FP+TP) (1) Where TP≡True Positive and FP≡False Positive. A True Positive is counted when the majority opinion of five expert reviewers agrees with the location of predicted labels in the image. A False Positive counted when majority opinion of five expert reviewers disagrees with the location of one or more predicted labels.

False Discovery Rate (FDR) is measured for a particular image frame between predicted objects and the majority opinion of five reviewers. The frame-level FDR threshold for this prospective study is 20%. Therefore, for a successful outcome the frame-level FDR must be less than 20%.

Sponsors and collaborators

Lead sponsor

EchoNous Inc.

Industry

Registry information

Official study title

ECHO-009 Kosmos Anatomical Object Labeling and View Identification Pivotal Study

Important dates

Study start: 2024
Primary completion: 2024
Study completion: 2024
First posted: Jun 21, 2024
Registry last updated: Jun 21, 2024

OpenTrials presents study information sourced from ClinicalTrials.gov. The official registry record should be consulted for the latest information.

View the official ClinicalTrials.gov record (opens in a new tab)

This listing is for discovery and informational purposes only. It is not medical advice, does not guarantee that a study is recruiting, and does not determine eligibility. Contact the study team and a qualified healthcare professional when considering participation.

Published trials that share one or more normalized conditions with this study.

Completed