Feasibility of AI-based Classification of Normal, Wheeze and Crackle Sounds From Stethoscope in Clinical Settings
Primary Purpose
Respiratory, Lung
Status
Completed
Phase
Not Applicable
Locations
Pakistan
Study Type
Interventional
Intervention
Artificial Intelligence Algorithm
Sponsored by
About this trial
This is an interventional diagnostic trial for Respiratory
Eligibility Criteria
Inclusion Criteria:
- Ages all
- Written consent provided
Exclusion Criteria:
- Subject condition unstable
- Chest wall deformity or wounds in adhesive application areas
- Written consent not provided
Sites / Locations
- Lady Reading Hospital, Pakistan
Outcomes
Primary Outcome Measures
Testing the accuracy of artificial intelligence models for detection of wheeze, crackles, and normal lung sounds by measuring the sensitivity and specificity
Artificial intelligence models are trained on lung sounds collected from three different digital stethoscopes named NoaScope, eSteth, and Littmann individually. Data from all three digital stethoscopes is also merged to train separate AI models. These trained AI models will be evaluated based on sensitivity which is the ability to correctly identify wheezes and crackles, and specificity which is the ability to correctly identify normal lung sounds. True positive (TP), true negative (TN), false positive (FP), and false-negative (FN) values will be used to calculate sensitivity & specificity using the following expressions.
Sensitivity: TP/TP+FN Specificity: TN/TN+FP
Clinical validation of AI models for detection of wheeze, crackles, and normal lung sounds by comparison with gold standard
AI models will be tested for their clinical feasibility through comparison of results obtained from AI models with that of the gold standard by measuring positive and negative agreement (NPA & PPA). The gold standard is the label given to each lung sound recording by an experienced consultant pulmonologist. The AI model is blinded to these labels and is tested independently for detection of normal lung sounds, wheezes, and crackles
Secondary Outcome Measures
Performance analysis of three digital stethoscopes: Littmann, NoaScope, and eSteth
Performance analysis of three digital stethoscopes NoaScope, eSteth, and Littmann will be evaluated using the sensitivity and specificity achieved by each stethoscope. True positive (TP), true negative (TN), false positive (FP), and false-negative (FN) values will be used to calculate sensitivity & specificity using the following expressions.
Sensitivity: TP/TP+FN Specificity: TN/TN+FP
Full Information
NCT ID
NCT05268263
First Posted
January 8, 2022
Last Updated
April 4, 2023
Sponsor
Innova Smart Technologies (Pvt.) Ltd
Collaborators
Lady Reading Hospital, Pakistan, NOABIO LLC
1. Study Identification
Unique Protocol Identification Number
NCT05268263
Brief Title
Feasibility of AI-based Classification of Normal, Wheeze and Crackle Sounds From Stethoscope in Clinical Settings
Official Title
Evaluating the Feasibility of Artificial Intelligence Algorithms in Clinical Settings for Classification of Normal, Wheeze and Crackle Sounds Acquired From a Digital Stethoscope
Study Type
Interventional
2. Study Status
Record Verification Date
April 2023
Overall Recruitment Status
Completed
Study Start Date
January 6, 2022 (Actual)
Primary Completion Date
February 22, 2022 (Actual)
Study Completion Date
February 22, 2022 (Actual)
3. Sponsor/Collaborators
Responsible Party, by Official Title
Sponsor
Name of the Sponsor
Innova Smart Technologies (Pvt.) Ltd
Collaborators
Lady Reading Hospital, Pakistan, NOABIO LLC
4. Oversight
Studies a U.S. FDA-regulated Drug Product
No
Studies a U.S. FDA-regulated Device Product
Yes
Product Manufactured in and Exported from the U.S.
Yes
5. Study Description
Brief Summary
Assessing the feasibility and testing the accuracy of the developed artificial intelligence algorithms for detection of wheezes and crackles in patients with lung pathologies in clinical settings on unseen local patient data acquired through three digital stethoscopes.
6. Conditions and Keywords
Primary Disease or Condition Being Studied in the Trial, or the Focus of the Study
Respiratory, Lung
7. Study Design
Primary Purpose
Diagnostic
Study Phase
Not Applicable
Interventional Study Model
Single Group Assignment
Masking
None (Open Label)
Allocation
N/A
Enrollment
60 (Actual)
8. Arms, Groups, and Interventions
Intervention Type
Device
Intervention Name(s)
Artificial Intelligence Algorithm
Intervention Description
The enrolled population will include patients with a history of lung pathologies. Artificial intelligence-based models are developed for classification of wheezes, crackles and normal lung sounds. These AI models will be tested and assessed on local lung sounds clinical data.
Primary Outcome Measure Information:
Title
Testing the accuracy of artificial intelligence models for detection of wheeze, crackles, and normal lung sounds by measuring the sensitivity and specificity
Description
Artificial intelligence models are trained on lung sounds collected from three different digital stethoscopes named NoaScope, eSteth, and Littmann individually. Data from all three digital stethoscopes is also merged to train separate AI models. These trained AI models will be evaluated based on sensitivity which is the ability to correctly identify wheezes and crackles, and specificity which is the ability to correctly identify normal lung sounds. True positive (TP), true negative (TN), false positive (FP), and false-negative (FN) values will be used to calculate sensitivity & specificity using the following expressions.
Sensitivity: TP/TP+FN Specificity: TN/TN+FP
Time Frame
2 months
Title
Clinical validation of AI models for detection of wheeze, crackles, and normal lung sounds by comparison with gold standard
Description
AI models will be tested for their clinical feasibility through comparison of results obtained from AI models with that of the gold standard by measuring positive and negative agreement (NPA & PPA). The gold standard is the label given to each lung sound recording by an experienced consultant pulmonologist. The AI model is blinded to these labels and is tested independently for detection of normal lung sounds, wheezes, and crackles
Time Frame
2 months
Secondary Outcome Measure Information:
Title
Performance analysis of three digital stethoscopes: Littmann, NoaScope, and eSteth
Description
Performance analysis of three digital stethoscopes NoaScope, eSteth, and Littmann will be evaluated using the sensitivity and specificity achieved by each stethoscope. True positive (TP), true negative (TN), false positive (FP), and false-negative (FN) values will be used to calculate sensitivity & specificity using the following expressions.
Sensitivity: TP/TP+FN Specificity: TN/TN+FP
Time Frame
2 months
10. Eligibility
Sex
All
Accepts Healthy Volunteers
Accepts Healthy Volunteers
Eligibility Criteria
Inclusion Criteria:
Ages all
Written consent provided
Exclusion Criteria:
Subject condition unstable
Chest wall deformity or wounds in adhesive application areas
Written consent not provided
Facility Information:
Facility Name
Lady Reading Hospital, Pakistan
City
Peshawar
ZIP/Postal Code
25000
Country
Pakistan
12. IPD Sharing Statement
Plan to Share IPD
No
Learn more about this trial
Feasibility of AI-based Classification of Normal, Wheeze and Crackle Sounds From Stethoscope in Clinical Settings
We'll reach out to this number within 24 hrs