Triton Canister System

• K Number: K142801
• Device Name: Triton Canister System
• Manufacturer: Gauss Surgical, Inc.
• Date Cleared: 2015-03-12 ( 164 days)
• Product Code: PBZ
• Regulation Number: 880.2750
• Type: Traditional
• Panel: SU
• Reference & Predicate Devices: N/A

Intended Use

The Triton Canister System is a software application intended to be used as an adjunct in the estimation of blood loss.

The Triton Canister System is intended to be used with blood bearing canisters, software, hardware, and accessory items which have been validated for use with the Triton Canister System to estimate the hemoglobin (Hb) mass contained within canisters with the input of the total volume in each canister. The Triton Canister System is also intended to calculate an estimate of blood volume in blood bearing canisters from the estimated Hb mass and a user-entered patient serum Hb value. The validated canister types, hardware, software, accessory devices, and Hb mass ranges are listed in the Instructions for Use.

Device Description

The Gauss Surgical Triton Canister System is an image processing system to estimate the external blood lost from patients into a suction canister used to collect fluid during surgical procedures. The system is comprised of the software which runs on a mobile platform (Apple® iPad®) and two accessories provided by Gauss Surgical, a Canister Type Specific Insert and a Canister Scanning Label. The Insert and Label ensure variables associated with imaging are standardized.

AI/ML Overview

Here's a breakdown of the acceptance criteria and study information for the Triton Canister System:

Acceptance Criteria and Device Performance

The general acceptance criteria for "Non-clinical performance data" is that the device performs as intended under anticipated conditions of use, with a comparison to a scientifically valid alternative method for measuring deposited hemoglobin mass. Specific conditions tested include:

• Lighting conditions
• Range of expected hemoglobin concentrations (and hemolysis levels)
• Range of expected blood volume absorption in canisters
• Presence of other non-sanguineous fluids (e.g., saline irrigation fluid)

The study employed Bland-Altman analysis to establish the bias and limits of agreement between the device's estimated hemoglobin mass (Triton cHbL) and the pre-measured hemoglobin mass (Assay cHbL).

Acceptance Criteria Category	Specific Criteria	Reported Device Performance and Methodology
Non-clinical performance data (Accuracy)	Device performs as intended under anticipated conditions of use, showing a comparison to a scientifically valid alternative method for measuring deposited hemoglobin mass. This includes testing:
A. Lighting conditions
B. Range of expected hemoglobin concentrations
C. Range of expected blood volume absorption
D. Presence of other non-sanguineous fluids (e.g., saline irrigation fluid)	Verification and Validation Bench Testing:

• Deposited known quantities of blood volume and Hb mass into canisters.
• Canister fluid samples represented clinically-expected ranges and distributions of fluid volume, dilution (by saline), Hb mass, hemolysis levels, ambient light illuminance, and serum patient Hb.
• Triton Canister App used to capture scans under three different ambient lighting conditions.
• User-entered volumes were recorded.
• Images and volumes transferred to server-based software to calculate Triton cHbL.
• cHbL compared to pre-measured Hb mass (Assay cHbL).
• Result: Strong positive linear correlation between Triton cHbL and Assay cHbL across tested conditions.
• Result: Bias and outer 95% CIs of the Bland-Altman Limits of Agreement fell within pre-determined acceptance criteria.
• This testing informed look-up tables and "error estimate" values displayed on the user interface. |
  | Software display of cumulative error | Software display must include an estimate of the cumulative error associated with estimated blood loss values. | Met: An estimate of the cumulative error associated with blood loss values is displayed to the user with each estimated hemoglobin mass and blood loss value. The results of verification testing and Bland-Altman methods inform these displayed values. |
  | Human factors testing and analysis | Validate that the device design and labeling are sufficient for appropriate use by intended users of the device. | Met: A usability study was conducted in a simulated setting with personnel who track blood loss during surgical procedures. Both quantitative and qualitative survey data were collected. All users successfully completed tasks per protocol pass/fail criteria. |
  | Electromagnetic Compatibility (EMC) and wireless performance | Appropriate analysis and non-clinical testing must validate the EMC and wireless performance of the device. | Met: EMC and wireless coexistence testing completed for the predicate device (Pixel 3 System) using the same iPad 2 was reviewed. The iPad 2 was found to be EMC compatible with the operating room environment (Class B requirements of IEC 60601-1-2:2007) and maintained essential wireless functionality under noisy conditions. |
  | Software verification, validation, and hazard analysis | Appropriate software verification, validation and hazard analysis must be performed. | Met: Software is considered a moderate level of concern. All elements of software information for moderate LOC devices (per FDA Guidance May 11, 2005) were provided, including development program, hazard analysis (patient's and user's standpoint), validation process, system-level test protocols, pass/fail criteria, results, and cybersecurity risk mitigation. Testing demonstrated software performs as intended and risks are mitigated. |
  | Labeling | Labeling must include:
  A. Warnings, cautions, and limitations for safe use;
  B. Detailed summary of performance testing pertinent to use, including bias and variance;
  C. Validated surgical materials, range of hemoglobin mass, software, hardware, and accessories;
  D. EMC and wireless technology instructions and information. | Met: Labeling includes all required details from the special controls. |

---

Study Details for Performance Testing Bench

1. Sample Size used for the test set and the data provenance:

   • The document states, "Whole blood samples of known Hb concentration and various pre-specified volumes were reconstituted from units of human packed red blood cells and plasma." It also mentions, "Serial dilution yielded canister samples reconstituted to ranges of fluid volume, dilution, hemolysis levels and Hb mass."
   • The exact numerical sample size for the test set (number of canisters/samples) is not explicitly stated in the provided text.
   • Data Provenance: The blood samples were reconstituted from units of human packed red blood cells and plasma. The testing was retrospective in nature, as it involved preparing samples with known characteristics in a lab setting ("bench testing") rather than collecting data from live surgical procedures. The country of origin for the data is not specified, but given it's an FDA submission, it's highly likely to be within the United States.

2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g., radiologist with 10 years of experience):

   • The ground truth for the test set was established by "pre-measured Hb mass (Assay cHbL)" of the reconstituted samples. This implies a laboratory assay was used, which would typically be performed by trained lab technicians or scientists.
   • The document does not specify the number or qualifications of individuals who established this ground truth, beyond referencing the "scientifically valid alternative method" of laboratory assay.

3. Adjudication method (e.g., 2+1, 3+1, none) for the test set:

   • The ground truth was established by direct measurement (Assay cHbL) rather than expert consensus on interpretive data. Therefore, an adjudication method for human readers is not applicable in this context.

4. If a multi-reader multi-case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance:

   • No, an MRMC comparative effectiveness study was not done. This study focuses on the standalone performance of the device in estimating hemoglobin mass in canisters, not on human reader performance with or without AI assistance. The device is described as "an adjunct in the estimation of blood loss," implying it provides data to clinicians, rather than directly assisting in the interpretation of images by human readers in a diagnostic setting.

5. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done:

   • Yes, a standalone performance study was done for the algorithm. The "Performance Testing Bench" described the "ability of the Triton Canister System (including algorithm, app, server and accessories – Insert and Scanning Label) to estimate canister hemoglobin mass loss and canister blood volume loss in comparison to a scientifically valid method of estimating hemoglobin mass and blood volume contained in canisters." The algorithm calculated the "Triton cHbL" which was then compared to the "Assay cHbL." While a user-entered volume is an input, the core estimation of Hb mass from the image is an algorithmic, standalone function.

6. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):

   • The ground truth used was laboratory assay / direct measurement of hemoglobin mass ("Assay cHbL") in the prepared samples. This is a highly objective, quantitative form of ground truth.

7. The sample size for the training set:

   • The document does not explicitly state the sample size used for the training set for the Triton Canister System's algorithm. It describes the testing of the performance of the algorithm.

8. 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. However, given the nature of the device and the performance testing, it is highly probable that the training data would also utilize laboratory-measured hemoglobin mass in similar prepared samples, just like the test set.

Regulation Number and Section

§ 880.2750 Image processing device for estimation of external blood loss.

(a)
Identification. An image processing device for estimation of external blood loss is a device to be used as an aid in estimation of patient external blood loss. The device may include software and/or hardware that is used to process images capturing externally lost blood to estimate the hemoglobin mass and/or the blood volume present in the images.(b)
Classification. Class II (special controls). The special controls for this device are:(1) Non-clinical performance data must demonstrate that the device performs as intended under anticipated conditions of use. Demonstration of the performance characteristics must include a comparison to a scientifically valid alternative method for measuring deposited hemoglobin mass. The following use conditions must be tested:
(i) Lighting conditions;
(ii) Range of expected hemoglobin concentrations;
(iii) Range of expected blood volume absorption; and
(iv) Presence of other non-sanguineous fluids (
e.g., saline irrigation fluid).(2) Human factors testing and analysis must validate that the device design and labeling are sufficient for appropriate use by intended users of the device.
(3) Appropriate analysis and non-clinical testing must validate the electromagnetic compatibility (EMC) and wireless performance of the device.
(4) Appropriate software verification, validation, and hazard analysis must be performed.
(5) Software display must include an estimate of the cumulative error associated with estimated blood loss values.
(6) Labeling must include:
(i) Warnings, cautions, and limitations needed for safe use of the device;
(ii) A detailed summary of the performance testing pertinent to use of the device, including a description of the bias and variance the device exhibited during testing;
(iii) The validated surgical materials, range of hemoglobin mass, software, hardware, and accessories that the device is intended to be used with; and
(iv) EMC and wireless technology instructions and information.