Search Results

The Butterfly iQ+ Ultrasound System is indicated for use by trained healthcare professionals in environments where healthcare is provided to enable diagnostic ultrasound imaging and measurement of anatomical structures and fluids of adult and pediatric patients for the following clinical applications: Peripheral Vessel (including carotid, deep ven thrombosis and arterial studies), Procedural Guidance, Small Organs (including thyroid, scrotum and breast), Cardiac, Abdominal, Lung, Urology, Fetal/Obstetric, Gynecological, Musculoskeletal (conventional), Musculoskeletal (superficial) and Ophthalmic. Modes of operation include B-mode + M-mode, B-mode + Color Doppler, B-mode + Power Doppler, Spectral Pulsed Wave Doppler.

The Butterfly iQ/Butterfly iQ+ Ultrasound System is a hand-held general-purpose diagnostic imaging system for use by trained healthcare professionals in environments where healthcare is provided to enable visualization and measurement of anatomical structures and fluid of adult and pediatric patients. The system consists of a single transducer with broad imaging capabilities connected to a standard handheld commercial off the shelf (COTS) mobile device compatible with the Butterfly iO/iO+ mobile application (app). The subject device introduces the Auto B-line Counter, a software application backed by an image analysis algorithm. The purpose of the Auto B-line Counter is to provide automated detection and automatic calculation of the number of B-lines to a user in a given rib space and also provides the users the capabilities of reviewing the detected B-lines (via visual overlays). The overlay of B-lines does not mark images for detection of specific pathologies. The Auto B-line Counter enables the automated identification and count of B-lines during a lung scan and is integrated into the existing Butterfly iQ/iQ+ mobile application for use with the Butterfly iQ or iQ+ transducers.

The provided text describes the Butterfly iQ/iQ+ Ultrasound System, which introduces an "Auto B-line Counter" software application. The information below summarizes the acceptance criteria and the study that proves the device meets these criteria.

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria for the Auto B-line Counter algorithm's performance are primarily established through analytical validation and clinical performance evaluation.

• Intraclass Correlation Coefficient (ICC) between annotators for Quality Indicator.
• Dice Coefficient Score (DSC) for conformance of automatic B-line segmentation to ground truth.
  (Specific numerical thresholds for ICC and DSC for acceptance are not provided in the text, but the claim is that criteria were met.) |
  | Clinical Performance Evaluation | Algorithm performance non-inferior to clinician annotator ground truth | Demonstrated non-inferiority. Performance assessed by calculating the Intraclass Correlation Coefficient (ICC) between the tool and the ground truth. Algorithm's performance was consistent among clinically meaningful subgroups: age, gender, and BMI. (Specific numerical thresholds for ICC for acceptance are not provided, but the claim is that non-inferiority was shown). |

2. Sample Sizes Used for the Test Set and Data Provenance

• Analytical Testing Test Set:
  • Sample Size: 6000 de-identified cines.
  • Data Provenance: Acquired from 253 sites. The datasets spanned many demographic variables including gender (male, female, and unidentified), age (20-90 years), and ethnicity via collection from a multitude of clinical sites with diverse and distinct racial patient populations. The data included various clinical subgroups and confounders such as congestive heart failure, heart failure with reduced ejection fraction, diabetes (with and without chronic complications), myocardial infarction, peripheral vascular disease, and renal disease. This suggests a retrospective, multi-center, multi-national (implied by "diverse and distinct racial patient populations") data provenance, although explicit country of origin is not stated.
• Clinical Performance Evaluation Test Set:
  • Sample Size: 99 subjects.
  • Data Provenance: Not explicitly detailed beyond being used for clinical performance evaluation. Given the context, it is likely also retrospective data from a similar pool as the analytical test set, or specifically collected for this evaluation.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications

• Number of Experts: The text refers to "expert annotators" (plural) for determining ground truth locations of B-lines and for establishing ICC. While a specific number isn't provided, the use of "annotators" and "experts" in plural implies a group.
• Qualifications of Experts: The text states, "The ground truthing for B-line counts was determined by the ICC among expert annotators presented with lung cines and instructions to determine the maximum number of B-Lines using the instant percent method. The ground truth locations of B-lines were then determined by expert annotator segmentations." The exact qualifications (e.g., number of years of experience, specialty like radiologist or emergency physician) are not detailed.

4. Adjudication Method for the Test Set

• Adjudication Method: The ground truthing involved assessing the "Intraclass Correlation Coefficient (ICC) among expert annotators." This suggests that multiple experts independently provided assessments, and their agreement (measured by ICC) was used to establish the ground truth or validate its reliability. It doesn't explicitly state a specific adjudication method like 2+1 or 3+1 for resolving discrepancies, but rather implies consensus or high agreement as the basis for ground truth.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

• The document does not describe a Multi-Reader Multi-Case (MRMC) comparative effectiveness study where human readers' performance with and without AI assistance is compared. The studies described focus on the standalone performance of the AI algorithm against human annotator ground truth.

6. Standalone (Algorithm Only Without Human-in-the-Loop Performance) Study

• Yes, standalone performance was done. The analytical validation and clinical performance evaluation sections explicitly describe testing the "Auto B-line Counter algorithm performance" as being "non-inferior to clinician annotators (Ground Truth)" and calculating the ICC between "the tool and the ground truth." This indicates a direct comparison of the algorithm's output against established ground truth, representing standalone performance.

7. Type of Ground Truth Used

• Expert Consensus / Expert Annotation: The ground truth for B-line counts was determined by the "ICC among expert annotators" and by "expert annotator segmentations" for the locations of B-lines. This strongly indicates an expert consensus or expert annotation approach, where human experts interpret the images to establish the reference standard.

8. Sample Size for the Training Set

• The document does not specify the sample size for the training set. It only states that the "data used for verification is completely distinct from that used during the training process and there is no overlap between the two."

9. How the Ground Truth for the Training Set Was Established

• The document does not explicitly describe how the ground truth for the training set was established. It only ensures the independence of training and testing data and mentions the process for establishing ground truth for the verification/test sets.

Ask a specific question about this device