Search Results

The Confocal Microprobe Imaging System can enter the human body cavity or surgical channel through an endoscope, allowing confocal laser imaging of the microstructure of tissues, including but not limited to the identification of cells, vessels and their organization or architecture.

The working principle of Confocal Microprobe Imaging System is based on probe-based confocal laser endomicroscopy technology (pCLE). The system combines confocal technology and fiber beam imaging technology. The fiber Optic Microprobe can enter the human cavity through the endoscopic working channel and contact the tissue cells through the object lens at the front end of the Fiber Optic Microprobe. The imaging principle of the device is as follows:

The laser scanning beam emitted by the laser in the Laser Scanning System forms a light source through the grating pinhole and is transmitted to the focal plane of the fluorescent labeled tissue cells through the Fiber Optic Microprobe. The fluorescent substance in the measured tissue cell emits fluorescence under the excitation of the laser. The fluorescence signal is collected by Fiber Optic Microprobe front end object lens and transmitted through the fiber beam microprobe to the detecting hole and then is transmitted to the photomultiplier tube (PMT) of the Photoelectric detector and then to the host for signal analysis and processing. Finally, the image is formed on the computer monitoring screen after software processing.

Light emitted at the top and bottom of the focal plane of the tested tissue produces a large diameter spot (much larger than that of the detecting hole) at the detecting hole, thus only a very small part of the light can be received by the detector through the detecting hole. Moreover, the larger the distance from the focal plane of the object lens, the larger the diffuse spot produced by the non-targeted tissue in the detecting hole and the lesser energy passes through the detecting hole (from 10% to 1%, slowly close to 0%), thus the weaker the unwanted signal is generated on the detector, and smaller the impact is caused by non-targeted tissue. Because confocal microscopy only images the focal plane of the target tissue, it effectively avoids the interference of diffracting light and scattered light, so that it has a higher resolution than ordinary microscopy and has been widely used in biology.

It appears that the provided FDA 510(k) Clearance Letter does not contain detailed information about a clinical study involving human readers or a specific "acceptance criteria" table with reported performance metrics for an AI component.

The document discusses the Confocal Microprobe Imaging System, which is a hardware device for imaging tissues. While it mentions "software" and "cybersecurity," these sections focus on general software validation and cybersecurity considerations, not the performance evaluation of an AI algorithm designed to interpret or analyze the images beyond the device's basic function.

The "Performance Testing" section states that "Performance Verification Test has been conducted in accordance with the internal performance requirements stated in the Performance Validation Scheme (HRD0003932 & HRD0004124)" and lists technical performance requirements like "Field of view, Horizontal resolution, Depth of Observation, Frame rate." These relate to the imaging system's hardware performance, not an AI's diagnostic accuracy.

Therefore,Based on the provided FDA 510(k) clearance letter, I cannot fulfill your request for detailed information regarding acceptance criteria for an AI component and the study that proves the device meets those criteria. The letter primarily addresses the clearance of a Confocal Microprobe Imaging System (hardware), focusing on its substantial equivalence to predicate devices based on technological characteristics, biocompatibility, reprocessing, electrical safety, and general software/cybersecurity validation.

There is no mention of an AI-specific component, its performance criteria, or any clinical studies (e.g., MRMC studies) pertaining to AI algorithm performance within this document. The "Performance Testing" section refers to the optical and functional performance of the imaging system itself, not the diagnostic performance of an AI that might interpret the images generated.

If such an AI component exists, its performance evaluation would typically be described in a separate section with specific metrics like sensitivity, specificity, or AUC, and details about the study design (test set, ground truth, expert adjudication, etc.). This information is absent in the provided text.

Ask a specific question about this device

Upon intravenous administration of ICG(Indocyanine green for injection) consistent with its approved label, the OPTOVISION provides real-time endoscopic visible and near-infrared fluorescence imaging. The OPTOVISION enable surgeons to perform minimally invasive surgery using standard endoscope visual light as well as visual assessment of vessels, blood flow and related tissue perfusion, and at least one of the major extra-hepatic bile ducts (cystic duct, common bile duct and common hepatic duct), using near-infrared imaging. Fluorescence imaging of biliary ducts with the OPTOVISION is intended for use with standard-of-care white light and, when indicated, intraoperative cholangioaraphy. The device is not intended for standalone use for biliary duct visualization.

Upon interstitial administration and use of ICG consistent with its approved label, the OPTOVISION is used to perform intraoperative fluorescence imaging and visualization of the lymphatic system, including lymphatic vessels and lymph nodes.

OPTOVISION is an endoscopic light source that enables real-time endoscopic visible and nearinfrared fluorescence imaging minimally invasive surgical procedures. Near-infrared illumination is used for fluorescence imaging using indocyanine green (ICG). This device is largely composed of a power circuit, a control circuit, and a lighting lamp (light source). A power circuit provides power to the lighting lamp and cuts off the power in an emergency, a control circuit controls the light output (intensity)/other settings of light rays, and a light source unit irradiates light directly to the Light Guide Cable (K111342) for observation of the affected area. With the configuration above, the power input from the initial power source is transmitted to the light source unit and the main body, and rays of white light and near-infrared light are directly irradiated through the Light Guide Cable to the affected area for observation. This device is designed for use in the following applications: standard endoscopic visible-light imaging and near-infrared fluorescence imaging.

The provided FDA 510(k) summary (K243591) for the "OPTOVISION Endoscopic Light Source Unit" describes the device and its comparison to a predicate device, but it does not present an acceptance criteria table or report specific device performance metrics in the way typically seen for AI/ML device studies (e.g., sensitivity, specificity, AUC).

Instead, this submission is for an endoscopic light source, and its "performance" is demonstrated through compliance with established electrical, electromagnetic, usability, photobiological safety, and software validation standards, as well as a general "Performance - Bench" test. This type of device, an endoscopic light source, does not inherently involve an AI algorithm with output metrics like those you've requested. Thus, many of the specific questions about sample sizes, expert ground truth, adjudication, MRMC studies, and standalone performance are not applicable to the information provided.

However, I can extract and present the available information in a structured format as best as possible, noting where the requested details are not present in the document.

Explanation regarding the absence of typical AI/ML study information:

The "OPTOVISION Endoscopic Light Source Unit" is classified as an "Endoscope And Accessories" (21 CFR 876.1500) with product code OWN, which refers to "Confocal Optical Imaging". This device is a light source that enables real-time endoscopic visible and near-infrared fluorescence imaging using Indocyanine green (ICG). While it uses near-infrared light for fluorescence imaging, the document does not describe it as an AI/ML device that performs diagnosis or prediction, hence the absence of metrics like sensitivity, specificity, and the related study design details. Its "performance" is primarily assessed against safety and engineering standards.

Acceptance Criteria and Device Performance (Based on available information)

Since this is an endoscopic light source, the "acceptance criteria" are compliance with relevant safety and performance standards for medical electrical equipment and usability. The "reported device performance" is the successful passing of these tests.

Study Details (Based on available information)

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

   • Test Set Sample Size: Not applicable. The "performance data" describes compliance with engineering and safety standards, not a clinical study on a population of patients or data sets in the context of AI/ML.
   • Data Provenance: Not applicable. The "tests" are engineering and safety evaluations of the device itself.

2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

   • Not applicable. This device is a light source, not an AI/ML algorithm requiring expert-established ground truth for its output. The "ground truth" for these tests would be the specifications and requirements of the standards themselves.

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

   • Not applicable. This type of adjudication is for disagreements in expert readings/labeling, which does not apply to compliance testing against engineering standards.

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 MRMC study was done, as this device is a light source and not an AI assistant intended to improve human reader performance.

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

   • Not applicable. This is not an AI algorithm performing a standalone task. Its function is to provide illumination for human visualization.

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

   • Not applicable. The "ground truth" for the performance data are the requirements and specifications defined by the referenced international and national standards (e.g., IEC, ES standards) for medical device safety, EMC, usability, photobiological safety, and software.

7. The sample size for the training set:

   • Not applicable. The device is not an AI/ML model that requires a training set.

8. How the ground truth for the training set was established:

   • Not applicable. As above, there is no AI/ML training set.

Ask a specific question about this device

The Histolog® Scanner is a confocal laser system intended to allow imaging of the internal microstructure of tissues including, but not limited to, the identification of cells, vessels and their organization or architecture.

The Histolog® Scanner is a digital microscopy scanner for use on excised human tissue. Its operating principle is based on confocal fluorescence microscopy and uses non-ionizing, lowpower optical radiation (Class 1 laser product as per IEC 60825-1:2014-05). The Histolog® Scanner acquires digital images with high, micrometer-range resolution and enables the visualization of tissue microstructures down to the cellular level.

The Histolog® Scanner is based on a massively parallel signal acquisition and processing technology providing fast digital imaging over large areas. Image reconstruction does not involve any image stitching or any other similar image blending algorithms. Each pixel in the image is assigned an intensity value based on the light intensity collected by the detector for this particular position in the scan pattern.

The provided text does not contain the detailed information required to fulfill all aspects of the request regarding the device's acceptance criteria and the study proving it meets them. The document focuses on the regulatory submission and comparison to a predicate device, rather than a detailed clinical performance study.

Here's a breakdown of what can and cannot be extracted from the provided text:

What can be extracted:

• Acceptance Criteria for Non-Clinical Tests: The document lists acceptance criteria for various non-clinical performance and safety tests.
• Results for Non-Clinical Tests: The document states "PASS" for all listed internal validation tests.

What cannot be extracted (critical for a clinical performance study):

• Table of Acceptance Criteria and Reported Device Performance for Imaging Quality (Clinical): While "Imaging Quality" is listed as a test, the specific acceptance criteria (e.g., sensitivity, specificity, accuracy for a specific diagnostic task) and the actual reported performance values are not provided. The text only says "Histolog® Scanner system imaging requirements verification protocols. All requirements met." This is insufficient for a clinical performance study.
• Sample size used for the test set and data provenance: No information on the number of images/patients, or whether the data was retrospective/prospective or its origin.
• Number of experts used to establish ground truth and qualifications: No mention of experts or their qualifications.
• Adjudication method for the test set: No information.
• MRMC comparative effectiveness study details: No mention of human readers or AI assistance in a comparative study.
• Standalone (algorithm only) performance: While the device images tissue, there's no mention of an algorithm being evaluated in a standalone capacity against a ground truth. The device itself is the "scanner."
• Type of ground truth used: No mention of ground truth (e.g., pathology, outcomes).
• Sample size for the training set: The document discusses validation, not training.
• How the ground truth for the training set was established: Not applicable, as training data and ground truth establishment for AI are not mentioned.

Based on the provided text, here is the information that can be extracted and a clear indication of what is missing:

1. Table of Acceptance Criteria and Reported Device Performance

For non-clinical safety and performance tests:

Missing: Specific quantitative acceptance criteria (e.g., sensitivity, specificity, accuracy) and corresponding reported performance metrics for "Imaging Quality" related to the device's diagnostic capabilities or image interpretation for identifying specific microstructures (cells, vessels, organization). The document confirms internal verification protocols were met, but doesn't detail these protocols or their outcomes for clinical relevance.

2. Sample size used for the test set and the data provenance

Missing: No information regarding the sample size of any test set (e.g., number of tissue samples, patients, or images) used for evaluating the device's clinical performance or imaging quality related to microstructure identification. Data provenance (country of origin, retrospective/prospective) is also not provided.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts

Missing: The document does not describe any expert involvement in establishing ground truth for a test set.

4. Adjudication method for the test set

Missing: No information on an adjudication method.

5. 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

Missing: No MRMC study is described. The device is a scanner intended for imaging the microstructure of tissues, not explicitly an AI-assisted diagnostic tool for human readers based on this submission.

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

Missing: While the device performs standalone imaging, the document doesn't describe a separate "algorithm only" performance evaluation that would assess, for example, automated detection or classification capabilities without human interpretation of the images. The device itself is the imaging system.

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

Missing: No information on the type of ground truth used for any clinical performance or imaging quality assessment.

8. The sample size for the training set

Missing: No information on a training set, as the document focuses on device performance validation rather than machine learning model development.

9. How the ground truth for the training set was established

Missing: Not applicable, as detailed training data and its ground truth establishment are not mentioned.

Conclusion: The provided FDA submission letter and summary focus explicitly on demonstrating substantial equivalence to a predicate device primarily through non-clinical performance and safety data, and a high-level statement about meeting "imaging quality" requirements. It does not present a clinical performance study with detailed acceptance criteria, sample sizes, ground truth establishment, or human reader performance metrics that would be typical for an AI/CADe device or a device requiring such detailed clinical validation.

Ask a specific question about this device

KARL STORZ ICG Imaging System
The KARL STORZ ICG Imaging System is intended to provide real-time visible (VIS) and near-infrared (NIR) fluorescence imaging.

Endoscopic ICG System
Upon intravenous administration and use of ICG consistent with its approved label, the KARL STORZ Endoscopic ICG System enables surgeons to perform minimally invasive surgery using standard endoscopic visible light as well as visual assessment of vessels, blood flow and related tissue perfusion in adults and pediatric patients ≥1 month of age, and at least one of the major extrahepatic bile duct, common bile duct and common hepatic duct) in adults and pediatric patients ≥ 12 years of age, using near infrared imaging in accordance with the appropriately indicated endoscope. Fluorescence imaging of biliary ducts with the KARL STORZ Endoscopic ICG System is intended for use with standard of care white light and, when indicated, intraoperative cholangiography. The device is not intended for standalone use for biliary duct visualization.

Additionally, the KARL STORZ Endoscopic ICG System enables surgeon to perform minimally invasive cranial neurosurgery in adults and pediatrics and endonasal skull base surgery in adults and pediatrics > 6 years of age using standard endoscopic visible light as well as visual assessment of vessels, blood flow and related tissue perfusion using near infrared imaging.

Upon interstitial administration and use of ICG consistent with its approved label, the KARL STORZ Endoscopic ICG System is used to perform intraoperative fluorescence imaging and visualization of the lymphatic system, including lymphatic vessels and lymph nodes.

VITOM II ICG System
The KARL STORZ VITOM II ICG System is intended for capturing fluorescent mages for the visual assessment of blood flow, as an adjunctive method for the evaluation of tissue perfusion, and related tissue-transfer circulation in tissue and free flaps used in plastic, micro- and reconstructive surgical procedures. The VITOM II ICG System is intended to provide a magnified view of the surgical field in standard white light.

KARL STORZ Image1 S CCU
The Imagel S camera control unit (CU) in combination with either a compatible camera head or an appropriately indicated video endoscope is intended for real-time visualization, image recording and documentation during general endoscopic and microscopic procedures in adults and pediatrics.

KARL STORZ ICG Imaging System
The KARL STORZ ICG Imaging System is intended to provide real-time visible (VIS) and near-infrared (NIR) fluorescence imaging for general surgical sites including the abdomen, bile ducts, brain/skull base, and the lymph nodes/lymphatic vessels. Components of the system include:
Scopes:
3D TIPCAM®1 Rubina videoendoscope
26006ACA/BCA, 26616ACA/BCA Rigid Endoscope
26003ACA/ARA/BCA/BRA/FCA/FRA/FCEA/FREA
26046ACA/ARA/BCA/BRA/FCA/FRA
28164AC/BC/FC VITOM II ICG/NIR Telescope
20916025AGA

Light Source:
Power LED Rubina (TL400) Foot Switch (UF101) Fiber Optic Light Cable (495TIP/NCSC/NAC)

Camera Head:
Image1 S 4U Rubina (TH121)

Camera Control Unit (CCU):
Image1 S Connect II (TC201US) Image1 S 4U-Link (TC304US)

KARL STORZ Image1 S CCU
The KARL STORZ IMAGE1 S Camera Control Unit (CCU) is a modular CCU that consists of Image1 S Connect and Connect II modules and the link modules. The Connect modules can be connected to minimum of one and a maximum of three links modules. The modularity enables customers to customize their Image1 S system to their specific video needs.

The Image1 S includes, but not limited to, the following features:
Brightness control Enhancement Control Automatic Light Source Control Shutter Control Image/Video Capture
Seven increments of zoom from 1-2.5x and adaptive zoom
Modules of the Image1 S CCU include: Image1 S Connect (TC200US) Image1 S Connect II (TC201US) Image1 S H3-Link (TC300US) Image1 S X-Link (TC301US) Image1 S D3-Link (TC302US) Image1 S 4U-Link (TC304US)
Accessories to the Image1 S CCU include: Microscope Footswitch (TC019) Image1 S Pilot (TC014) LINK Cable (TC011, TC012)

The provided text describes the regulatory submission for the KARL STORZ ICG Imaging System and KARL STORZ Image 1S Camera Control Unit.

However, the document explicitly states that "Clinical testing was not required to demonstrate substantial equivalence to the predicate devices." This means that the submission does not contain information about a study proving the device meets acceptance criteria based on clinical performance metrics (like sensitivity, specificity, accuracy, or human reader improvement with AI assistance).

The acceptance criteria and performance data mentioned in the document are non-clinical performance data, specifically related to electrical safety, electromagnetic compatibility, and software verification and validation. This type of information is usually presented as compliance with established standards rather than a clinical study with a test set, ground truth, or expert readers.

Therefore, most of the requested information regarding clinical study design (sample size, data provenance, expert ground truth, adjudication, MRMC studies, standalone performance, training set details) cannot be extracted from this document because such a clinical study was not required or provided for this specific submission as per the FDA's determination of substantial equivalence to predicate devices (K212695 and K201135).

Here's what can be extracted and inferred from the document:

1. A table of acceptance criteria and the reported device performance

Based on the "Non-Clinical Performance Data" section, the acceptance criteria are compliance with relevant safety and software standards.

2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)

• Not applicable / not provided. The document states "Clinical testing was not required." The "performance data" refers to non-clinical engineering and software testing, which does not involve a "test set" of clinical cases in the sense of imaging data for diagnostic performance evaluation.

3. 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)

• Not applicable / not provided. No clinical ground truth was established from experts as clinical testing was not required.

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

• Not applicable / not provided. No clinical ground truth was established, therefore no adjudication method was used for clinical interpretation.

5. 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

• Not applicable / not provided. No MRMC study was performed as clinical testing was not required for this submission. The device is an imaging system, not an AI-based diagnostic aid that would assist human readers in interpretation.

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

• Not applicable / not provided. The device is an imaging system, not a standalone diagnostic algorithm. No such performance study was conducted or required.

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

• Not applicable / not provided for clinical performance. For the non-clinical performance data, the "ground truth" or standard was compliance with specified international and FDA-recognized standards for electrical safety, EMC, and software validation.

8. The sample size for the training set

• Not applicable / not provided. As this is an imaging system and not an AI/ML-based diagnostic algorithm, there is no mention of a "training set" of data in the context of machine learning model development. The software testing mentioned refers to standard software verification and validation, not machine learning model training.

9. How the ground truth for the training set was established

• Not applicable / not provided. See above.

Ask a specific question about this device

The KARL STORZ ICG Imaging System is intended to provide real-time visible (VIS) and near-infrared (NIR) fluorescence imaging.

Upon intravenous administration and use of ICG consistent with its approved label, the KARL STORZ Endoscopic ICG System enables surgeons to perform minimally invasive surgery using standard endoscopic visible light as well as visual assessment of vessels, blood flow and related tissue perfusion, and at least one of the major extra-hepatic bile ducts (cystic duct, common bile duct and common hepatic duct), using near infrared imaging of biliary ducts with the KARL STORZ Endoscopic ICG System is intended for use with standard of care white light and, when indicated, intraoperative cholangiography. The device is not intended for biliary duct visualization.

Additionally, the KARL STORZ Endoscopic ICG System enables surgeon to perform minimally invasive cranial neurosurgery in adults and pediatrics and endonasal skull base surgery in adults and pediatrics > 6 years of age using standard endoscopic visible light as well as visual assessment of vessels, blood flow and related tissue perfusion using near infrared imaging.

The KARL STORZ VITOM ICG System is intended for capturing fluorescent images for the visual assessment of blood flow, as an adjunctive method for the evaluation of tissue perfusion, and related tissue-transfer circulation in tissue and free flaps used in plastic, micro- and reconstructive surgical procedures in adults and pediatrics >Imonth of age. The VITOM ICG System is intended to provide a magnified view of the surgical field.

Upon interstitial administration and use of ICG consistent with its approved label, the KARL STORZ Endoscopic ICG System is used to perform intraoperative fluorescence imaging and visualization of the lymphatic system, including lymphatic vessels and lymph nodes.

The subject device KARL STORZ ICG System includes the following components:

1. VITOM EAGLE (TH201): a 3D video exoscope with 4K resolution used during open procedures for the evaluation of tissue perfusion, related tissue-transfer circulation in tissue and free flaps used in plastic, micro and reconstructive surgical procedures. The subject device VITOM EAGLE System is being indicated for use in in adults and pediatrics >1month of age.
2. Fiber Light Cable (495VTE): used to transmit visible and NIR light from the Power LED Rubina light source to the VITOM Eagle.
3. IMAGE1 Pilot (TC014): used to control the optical functions of the VITOM EAGLE.
4. Microscope Footswitch (TC019): alternatively used control the optical functions of the VITOM EAGLE
5. The Power LED Rubina light source (TL400) along with the footswitch (UF101): previously cleared in K201399, K202925 and K212695.
6. Imagel S Camera Control Unit (TC201US, TC304US): previously cleared in K201399, K202925 and K212695.

The provided text describes the KARL STORZ ICG Imaging System and its acceptance criteria, along with a summary of the non-clinical performance data used to demonstrate substantial equivalence to a predicate device. However, it does not describe a study involving an AI algorithm. The device is an imaging system that uses Indocyanine Green (ICG) fluorescence for various surgical visualizations.

Here's a breakdown of the requested information based only on the provided text, heavily noting limitations due to the absence of AI-specific study details:

1. Table of Acceptance Criteria and Reported Device Performance

Since this is a non-AI imaging system without specific AI performance metrics, the acceptance criteria are generally related to the technical performance of the imaging capabilities. The document states that the KARL STORZ ICG Imaging System (subject device) was compared to the predicate VITOM II ICG/NIR telescope of the KARL STORZ ICG Imaging System (K212695). The performance was demonstrated by testing for:

2. Sample size used for the test set and the data provenance

The document does not specify a "test set" in the context of an AI algorithm or patient data. The performance evaluation was based on non-clinical bench testing comparing the subject device's imaging capabilities to a predicate device. Therefore, there's no mention of sample size in terms of patient data or data provenance (country of origin, retrospective/prospective). The "sample" here would refer to the physical devices and controlled test scenarios.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts

Not applicable. As this is a non-AI imaging system being evaluated via bench testing, there are no "experts" establishing ground truth for a test set of images or patient data. The ground truth for the technical performance criteria would be established by validated measurement techniques and instrumentation during the bench tests.

4. Adjudication method for the test set

Not applicable, as there is no "test set" in the context of human expert review or an AI algorithm's output. The evaluation was based on objective technical measurements.

5. 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. The document explicitly states "Clinical testing was not required to demonstrate the substantial equivalence to the predicate devices. Non-clinical bench testing was sufficient to establish the substantial equivalence of the modifications." Furthermore, this is not an AI-assisted device, so MRMC studies on AI assistance would not be relevant.

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

Not applicable. The device is an imaging system, not a standalone AI algorithm. It produces images for human surgeons to interpret.

7. The type of ground truth used

For the non-clinical performance data, the ground truth was based on objective measurements from bench testing (e.g., measuring spatial resolution, signal-to-noise ratio, etc.) against established technical specifications or a predicate device's performance.

8. The sample size for the training set

Not applicable. The device does not involve an AI algorithm that requires a training set.

9. How the ground truth for the training set was established

Not applicable, as there is no AI algorithm or training set.

Summary of AI-related information (or lack thereof):

The provided text describes a submission for a KARL STORZ ICG Imaging System, which is a medical device for real-time visible and near-infrared fluorescence imaging during surgery. The entire document focuses on demonstrating the substantial equivalence of this updated imaging system to a previously cleared predicate device through non-clinical bench testing. There is no mention of any artificial intelligence (AI) component, machine learning model, or any studies related to AI performance, human-in-the-loop improvements with AI, or standalone algorithm performance. Therefore, most of the questions regarding AI-specific criteria cannot be answered from the provided text.

Ask a specific question about this device

The 1788 4K Camera System with Advanced Imaging Modality is indicated for use in general laparoscopy, nasopharyngoscopy, ear endoscopy, sinuscopy, neurosurgery and plastic surgery whenever a laparoscope/ endoscope/ arthroscope/ sinuscope is indicated for use. The 1788 4K Camera System with Advanced Imaging Modality is indicated for use in adults and pediatric patients.

A few examples of the more common endoscope surgeries are Laparoscopic cholecystectomy, Laparoscopic hernia repar, Laparoscopic appendectomy, Laparoscopic pelvic lymph node detection, Laparoscopically assisted hysterectomy, Laparoscopic and thorascopic anterior spinal fusion, Anterior cruciate ligament reconstruction, Knee arthroscopy, Small joint arthroscopy, Decompression fixation, Wedge resection, Lung biopsy, Dorsal sympathectomy, Pleurodesis, Internal mammary artery dissection for coronary artery bypass, Coronary artery bypass grafting where endoscopic visualization is indicated and Examination of the evacuated cardiac chamber during performance of valve replacement.

The users of the 1788 4K Camera System with Advanced Imaging Modality are general and pediatric surgeons, gynecologists, cardiac surgeons, thoracic surgeons, orthopedic surgeons, ENT/heurosurgeons and urologists.

Upon intravenous administration of SPY AGENT GREEN (indocyanine green for injection, USP), the L12 LED Light Source with Advanced Imaging Modality and SafeLight Cable are used with SPY AGENT GREEN to provide real-time endoscopic visible and near infrared fluorescence imaging. The L12 Light Source with Advanced Imaging Modality and SafeLight Cable enable surgeons to perform minimally invasive surgery using standard endoscopic visible light as well as visual assessment of vessels, blood flow and related tissue perfusion in adults and pediatric patients aged one month and older, and visualization of at least one of the major extra-hepatic bile duct, common bile duct and common hepatic duct) in adults and pediatric patients 12 to 17 years of age, using near-infrared imaging.

Fluorescence imaging of biliary ducts with the L12 LED Light Source with Advanced Imaging Modality and SafeLight Cable is intended for use with standard-of-care white light and, when indicated, intraoperative cholangiography. The devices are not intended for standalone use for biliary duct visualization.

Additionally, the L12 LED Light Source with Advanced Imaging Modality and SafeLight Cable enable surgeons to perform minimally invasive cranial neurosurgery in adults and pediatric patients and endonasal skull base surgery in adults and pediatric patients > 6 years of age using standard endoscopic visible light as well as visual assessment of vessels, blood flow and related tissue perfusion using near infrared imaging.

Upon interstitial administration of SPY AGENT GREEN, the L12 LED Light Source with Advanced Imaging Modality and SafeLight Cable are used to perform intraoperative fluorescence imaging and visualization of the lymphatic system, including lymphatic vessels and lymph nodes.

Upon administration and use of pafolacianine consistent with its approved label, the L12 LED Light Source with Advanced Imaging Modality and SafeLight™ Cable are used to perform intraoperative fluorescence imaging of tissues that have taken up the drug.

The L12 LED Light Source with Advanced Imaging Modality is also intended to transilluminate the ureter during open or laparoscopic surgical procedures.

Stryker's Advanced Imaging Modality (AIM) System is an endoscopic real-time 4K visible white light, near-infrared illumination and transillumination, and cyan spectral imaging system. Near-infrared illumination is used for fluorescence imaging using SPY AGENT™ GREEN (indocyanine green for injection, USP) or CYTALUX™ (pafalocianine) injection. Near-infrared illumination is also intended for use during transillumination of the ureters using the IRIS Ureteral Kit during minimally invasive and open surgical procedures. Cyan Spectral Imaging is intended as an alternative mode of visualization that uses narrow bands for illumination during endoscopic surgery to provide greater visualization of surface structures and fine capillary patterns of the mucosal membrane.

Here's an analysis of the provided text regarding the acceptance criteria and supporting studies for the Stryker 1788 4K Camera System with Advanced Imaging Modality and L12 LED Light Source with AIM:

A table of acceptance criteria and the reported device performance cannot be fully constructed from the provided document. The document primarily lists regulatory standards and general performance categories, indicating that the device "passed" these tests. It does not provide specific quantitative acceptance criteria or detailed reported performance values for each criterion.

However, based on the Performance Data section (page 9), we can infer the types of tests conducted and their general outcome:

Inferred Acceptance Criteria and Reported Device Performance

Here's the requested information based on the provided document:

1. A table of acceptance criteria and the reported device performance

   • (See table above)

2. Sample size used for the test set and the data provenance (e.g., country of origin of the data, retrospective or prospective)

   • The document states: "NOTE: The Advanced Imaging Modality (AIM) System does not require clinical studies to support the determination of substantial equivalence." This implies that no clinical test set (patient data) was used in the assessment for substantial equivalence. The testing described appears to be primarily bench and laboratory-based. Therefore, sample size and data provenance in the context of clinical data are not applicable based on this document.

3. 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)

   • Since no clinical studies were performed, there's no mention of experts establishing ground truth for a clinical test set. The "Pass" results are based on compliance with engineering standards and comparisons to predicate devices in bench testing.

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

   • Not applicable. No clinical test set demanding adjudication was used.

5. 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 MRMC comparative effectiveness study was done. The document explicitly states: "The Advanced Imaging Modality (AIM) System does not require clinical studies to support the determination of substantial equivalence." There is also no mention of AI assistance in the context of human readers for this device. The "Advanced Imaging Modality" refers to specialized illumination and visualization techniques, not AI-powered image analysis for diagnosis.

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

   • Not applicable. This device is an imaging system (camera and light source) that provides visual information to a surgeon. It is not an AI algorithm that operates standalone without human interpretation or intervention for diagnostic or therapeutic decisions.

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

   • For the performance testing, the implied "ground truth" or reference was primarily device input specifications and the performance of currently legally marketed predicate devices. For regulatory compliance, the ground truth was adherence to FDA-recognized voluntary consensus standards (e.g., IEC standards for electrical safety, usability, software, etc.).

8. The sample size for the training set

   • Not applicable. This document describes a medical imaging hardware system, not a machine learning model that requires a training set.

9. How the ground truth for the training set was established

   • Not applicable. As there is no machine learning component requiring a training set, the establishment of ground truth for such a set is not mentioned.

Ask a specific question about this device

The ActivSight Intraoperative Imaging System (ActivSight) is intended to provide real-time endoscopic fluorescence and near infrared imaging. ActivSight enables surgeons to visually assess vessels, blood flow, and related tissue perfusion using fluorescence and near infrared imaging, and at least one of the major bile ducts (cystic duct, or common hepatic duct) using fluorescence, all during minimally invasive surgery.

Fluorescence imaging of biliary ducts with ActivSight is intended for use with standard of care white light, and when indicated, intraoperative cholangiography. The device is not intended for stand-alone use for biliary duct visualization.

The ActivSight Intraoperative Imaging System (ActivSight) is an accessory to existing commercial surgical laparoscope systems, including cameras and video processor units. ActivSight provides real-time endoscopic fluorescence and nearinfrared imaging. These imaging features allow surgeons to visually assess vessels, blood flow, and tissue perfusion (using fluorescence and near-infrared imaging), and to visually assess at least one of the major bile ducts (cystic duct, common bile duct, or common hepatic duct) using fluorescence. Fluorescence imaging is enabled through use of any commercially available Indocyanine Green (ICG). These visualization features are available for surgeons to use during minimally invasive surgery. ActivSight is intended to be used in a surgical environment.

This report confirms that the ActivSight Intraoperative Imaging System (K231344) has been found substantially equivalent to its predicate device (K203550) by the FDA. This determination is primarily based on non-clinical testing related to a new sterilization method, as the device itself is identical in intended use and technology to the predicate. Therefore, detailed information regarding acceptance criteria and performance of an AI/human-in-the-loop system, as typically required for novel AI-powered medical devices, is largely not applicable to this specific submission.

However, based on the limited information provided, we can infer some aspects and highlight what is not present given the nature of the submission. Since the provided text is a 510(k) clearance letter and summary for a modification (specifically, a new sterilization method) to an existing device, it does not contain the detailed clinical study data typically found in original submissions for AI-powered devices.

Here's an analysis based on the context:

1. A table of acceptance criteria and the reported device performance

The provided documentation does not specify acceptance criteria and reported device performance related to diagnostic accuracy or clinical effectiveness, as this was a 510(k) submission for a modification (sterilization) rather than a de novo AI device.

The non-clinical testing performed focused on:

2. Sample size used for the test set and the data provenance

For the sterilization efficacy testing:

• Sample Size: Not explicitly stated, but typically involves a sufficient number of test articles (e.g., medical devices or biological indicators) to statistically validate a SAL of 10^-6.
• Data Provenance: Not specified, but likely laboratory testing conducted under controlled conditions. This is not clinical data from patients.

For functional testing after reprocessing:

• Sample Size: Not explicitly stated, but usually involves testing multiple units through a predefined number of reprocessing cycles.
• Data Provenance: Not specified, but likely laboratory testing.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts

This is not applicable to this type of submission. Ground truth, in the context of diagnostic accuracy, would involve expert radiologists or pathologists interpreting images or samples. The testing here is for sterility and functional reliability, not diagnostic performance.

4. Adjudication method for the test set

This is not applicable as there are no expert interpretations requiring adjudication.

5. 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 MRMC study was performed or required for this 510(k) submission, as it relates to a modification of an existing imaging system's sterilization method, not its diagnostic or assistive AI capabilities. The device is an intraoperative imaging system that provides fluorescence and near-infrared imaging for surgeons, not an AI-powered diagnostic tool for human readers.

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

Not applicable. The ActivSight system is described as an "accessory to existing commercial surgical laparoscope systems" and "enables surgeons to visually assess," indicating it's a visualization tool used by surgeons, not a standalone AI algorithm producing interpretations.

7. The type of ground truth used

For sterilization testing, the ground truth is established by microbiological methods (e.g., culturing biological indicators after sterilization to confirm kill rates). For functional testing, the ground truth is the device's operational specifications and performance metrics (e.g., image quality, light output, mechanical integrity).

8. The sample size for the training set

Not applicable. This device is not an AI algorithm that requires a training set. Even if the original predicate device (K203550) involved some form of image processing or enhancement, the provided document does not indicate that it is an AI/ML-based device that would undergo a training phase.

9. How the ground truth for the training set was established

Not applicable for the same reason as point 8.

In summary:

This 510(k) clearance is for a change to an already cleared medical device (specifically, its sterilization method). Therefore, the detailed performance data, acceptance criteria, and study designs typically associated with new AI/ML-driven diagnostic or treatment devices are not present in this document. The "study" mentioned here refers to the non-clinical validation of the new sterilization processes and confirmation of the device's continued functional reliability after these processes.

Ask a specific question about this device

Upon intravenous administration and use of an ICG consistent with its approved labeling, the FloNavi Open Field Fluorescence Imaging System is used to perform intraoperative fluorescence angiography.

The FloNavi Open Field Fluorescence Imaging System is indicated for fluorescence imaging of blood flow and tissue perfusion before, during, and after: vascular, gastrointestinal, organ transplant, and plastic, micro- and reconstructive surgical procedures.

Upon interstitial administration and use of an ICG consistent with its approved labeling, the FloNavi Open Field Fluorescence Imaging System is used to perform intraoperative fluorescence imaging and visualization of the lymphatic system, including lymphatic vessels and lymph nodes.

The FloNavi Open Field Fluorescence Imaging System is an imaging system used in hospitals for capturing and viewing fluorescent images for the visual assessment of blood flow, as an adjunctive method for the evaluation of tissue perfusion, and related tissue transfer circulation for use in imaging during various surgical procedures.

The FloNavi Open Field Fluorescence Imaging System has two system models: HD system and 4K system. Each system consists of the following main components: Imaging Head, Image Processing Unit, power supply cord and video cables.

The Imaging Head may be either handheld or attached to a mechanical arm and provides illumination of the regions of a patient's body to be observed with near infrared light to excite ICG fluorescence. Alternatively, the Imaging Head provides white light illumination of the regions of a patient's body to be observed for color imaging. The cameras in the Imaging Head capture the fluorescent image under near infrared illumination or a color image under white light illumination. The Image Processing Unit receives the video signal from the Imaging Head and processes and outputs the video image to a medical grade video monitor and/or video recorder. Adjustments to the operation of the FloNavi Open Field Fluorescence Imaging System are possible through switches at either the Imaging Head or the Image Processing Unit.

The FloNavi Open Field Fluorescence Imaging System is a reusable device and provided nonsterile. Its components should be cleaned and low-level disinfected prior to the first use and after every subsequent use.

The FloNavi Open Field Fluorescence Imaging System is indicated for fluorescence imaging of blood flow and tissue perfusion and for visualization of the lymphatic system. It is a Class II device.

Here's an analysis of the acceptance criteria and supporting study details based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

The provided document describes the device as being "Substantially Equivalent (SE) to the predicate device" (SPY Portable Handheld Imaging (SPY-PHI) System, K200737). The acceptance criteria are implicitly linked to the predicate device's performance and the subject device's ability to meet its own design specifications and relevant safety standards.

2. Sample Size and Data Provenance for Test Set

The document mentions "Performance testing and animal testing" were conducted.

• Sample Size:
  • Performance Testing: Not specified.
  • Animal Testing: Not specified.
• Data Provenance: Not specified (e.g., country of origin, retrospective or prospective).

3. Number of Experts and Qualifications for Ground Truth - Test Set

Not applicable/Not specified. The document does not describe a clinical study involving human readers or expert-established ground truth for the "test set" in the context of diagnostic accuracy. The testing primarily focuses on the device's functional and safety performance, as well as animal studies.

4. Adjudication Method for Test Set

Not applicable/Not specified. No such adjudication method is mentioned as the studies described appear to be non-clinical performance and animal studies, not reader studies for diagnostic accuracy.

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

No. The document does not mention a MRMC comparative effectiveness study where human readers' performance with and without AI assistance was evaluated. The device appears to be an imaging system, not an AI-powered diagnostic algorithm enhancing human reading.

6. Standalone (Algorithm Only) Performance Study

No. The document presents the FloNavi Open Field Fluorescence Imaging System as a standalone imaging device, not an AI algorithm. The performance testing and animal testing evaluate the device itself.

7. Type of Ground Truth Used

For the performance and animal testing mentioned, the "ground truth" would likely be based on:

• Engineering Specifications: For the technical performance aspects (e.g., image resolution, light output).
• Physiological Response: In animal testing, the "ground truth" for fluorescence imaging would be the actual blood flow, tissue perfusion, or lymphatic system visualization as observed and validated within the animal model, potentially through direct observation or comparison with established methods.

8. Sample Size for Training Set

Not applicable. This device is presented as an imaging system, not an AI/machine learning algorithm requiring a separate training set.

9. How Ground Truth for Training Set was Established

Not applicable, as there is no mention of an AI/machine learning component requiring a training set.

Ask a specific question about this device

Upon intravenous administration of SPY AGENT GREEN (indocyanine green for injection, USP), the L12 LED Light Source with AIM and SafeLight Cable are used with SPY AGENT GREEN to provide real-time endoscopic visible and near infrared fluorescence imaging. The L12 Light Source with AIM and SafeLight Cable enable surgeons to perform minimally invasive surgery using standard endoscopic visible light as well as visual assessment of vessels, blood flow and related tissue perfusion in adults and pediatric patients aged one month and older, and visualization of at least one of the major extra-hepatic bile duct, common bile duct and common hepatic duct) in adults and pediatric patients 12 to 17 years of age, using near-infrared imaging.

Fluorescence imaging of biliary ducts with the L12 Light Source with AIM and SafeLight Cable is intended for use with standard-of-care white light and, when indicated, intraoperative cholangiography. The devices are not intended for standalone use for biliary duct visualization.

Additionally, the L12 LED Light Source with AIM and SafeLight Cable enable surgeons to perform minimally invasive cranial neurosurgery in adults and pediatric patients and endonasal skull base surgery in adults and pediatric patients > 6 years of age using standard endoscopic visible light as well as visual assessment of vessels, blood flow and related tissue perfusion using near infrared imaging.

Upon interstitial administration of SPY AGENT GREEN, the L12 LED Light Source with AIM and SafeLight Cable are used to perform intraoperative fluorescence imaging and visualization of the lymphatic system, including lymphatic vessels and lymph nodes.

Upon administration and use of pafolacianine consistent with its approved label, the L12 LED Light Source with AIM and SafeLight™ Cable are used to perform intraoperative fluorescence imaging of tissues that have taken up the drug.

The L12 LED Light Source with AIM is also intended to transilluminate the ureter during open or laparoscopic surgical procedures.

The L12 LED Light Source with AIM is part of the Advanced Imaging Modality (AIM) System. The system is an endoscopic real-time 4K visible white light and near-infrared illumination and transillumination imaging system. Near-infrared illumination is used for fluorescence imaging using indocyanine green and pafolacianine injection. Near-infrared illumination is also intended for use during transillumination of the ureters using the IRIS Ureteral Kit during minimally invasive and open surgical procedures. The L12 LED Light Source is a light-generating until designed to illuminate surgical sites in the following applications: visible light, near-infrared fluorescence, and near-infrared transillumination.

This document describes the Stryker L12 LED Light Source with AIM, which is an endoscopic imaging system used to provide real-time visible and near-infrared fluorescence imaging during various surgical procedures. The device received 510(k) clearance (K230754) based on substantial equivalence to predicate devices, primarily the 780 nm L11 LED Light Source with AIM (K221611).

The provided text does not contain details about acceptance criteria or a study proving the device meets those criteria, as typically seen for AI/ML-enabled devices involving diagnostic accuracy or similar performance metrics. This is because the L12 LED Light Source with AIM is a hardware component (a light source) that enables certain imaging modalities, rather than a diagnostic algorithm itself. As such, its performance evaluation focuses on functional aspects and safety rather than diagnostic accuracy or human reader improvement.

Here's an analysis based on the information provided, specifically highlighting why certain requested sections are not applicable:

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria here are fundamental engineering and safety standards, and the reported performance is a simple "PASS" for these tests. There are no specific quantitative performance metrics (e.g., sensitivity, specificity, accuracy) provided as would be expected for an AI/ML-driven diagnostic device.

2. Sample size used for the test set and the data provenance

Not applicable. The device is a hardware component (light source). Performance testing was conducted in a bench setting on the device itself, not on a dataset of patient images or patient data.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts

Not applicable. This information is relevant for studies evaluating human perceptual or cognitive performance, typically in conjunction with AI/ML algorithms. The provided document describes safety and functional testing for a light source.

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

Not applicable. Adjudication methods are used to establish ground truth in image datasets, often by resolving discrepancies among expert readers. This device underwent bench and software functional testing.

5. 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

Not applicable. This device is not an AI/ML algorithm designed to assist human readers in image interpretation. It is a light source that enables certain imaging modalities. The document explicitly states: "The L12 LED Light Source does not require clinical studies to support the determination of substantial equivalence."

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

Not applicable. This is a hardware component, not a standalone AI algorithm.

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

Not applicable in the conventional sense for AI/ML evaluation. The "ground truth" for the tests performed was defined by engineering specifications and expected functional outputs of the light source (e.g., specific light output, proper communication, correct timing).

8. The sample size for the training set

Not applicable. This device does not use a training set as it's not an AI/ML algorithm.

9. How the ground truth for the training set was established

Not applicable. As above, there is no training set for this device.

Ask a specific question about this device

The SmartSurgN Visualization System is intended to provide realtime endoscopic visible (VIS) and real-time nearinfrared (NIR) fluorescence imaging. Upon intravenous administration and use of an ICC consistent with its approved label the SmartSurgN Visualization System enables surgeons to perform minimally invasive surgery using standard endoscopic visible light as well as visual assessment of vessels, blood flow and related tissue perfusion, and at least one of the major extra-hepatic bile duct, common bile duct and common hepatic duct) using near-infrared imaging. Fluorescence imaging of biliary ducts with the SmartSurgN Visualization System is intended for use with standard of care white light and, when indicated, intraoperative cholangiography. The device is not intended for standalone use for biliary duct visualization.

The SmartSurgN Visualization System is designed to provide real-time endoscopic visible (VIS) and real-time near-infrared (NIR) fluorescence imaging during minimally invasive surgery. The SmartSurgN Visualization System is comprised of the following main components: EyeRSurgn Console with Camera Head IRLightSurgN Light Source 10mm ICG Laparoscope, 0° or 30° The SmartSurgN Visualization System enables surgeons to perform minimally invasive surgery using standard endoscopic visible light as well as visual assessment of vessels, blood flow and related tissue perfusion, and at least one of the major extra-hepatic bile ducts (cystic duct, common bile duct and common hepatic duct), using near-infrared imaging. Fluorescence imaging of biliary ducts with the SmartSurgN Visualization System is intended for use with standard of care white light and, when indicated, intraoperative cholangiography. The device is not intended for standalone use for biliary duct visualization. During the use of the SmartSurgN Visualization System, the IRLightSurgN provides the light source for illumination of the surgical site. The IRLightSurgN is capable of outputting light in the visible light spectrum as well as in the near infrared spectrum. The user selects the image capture mode (Regular, IRMax, IRFlo, IRTrue) which determines the light spectrum used to capture imaging. The IRLightSurgN is connected to the SmartSurgN Laparoscope using a commercially available fiber optic light cable. Additionally, the SSN Laparoscope connects to the EyeRSurgN Camera Head. Images are acquired by the EyeRSurgN Camera Head and transmitted to the EyeRSurgN Console. Images are processed by the EyeRSurgN Console and outputted to a medical grade monitor. The SmartSurgN Visualization System can be used with any medical grade monitor with a HDMI or 3G-SDI input connection. The SmartSurgN Visualization System is intended to be used in conjunction with commercially available indocyanine green imaging (ICG) kits. ICG is a tricarbocyanine dye which fluoresces after excitation under near infrared light at 806 nm, permitting visualization of anatomical structures.

The provided document does not contain the detailed information required to fill out all the requested sections regarding acceptance criteria and a study proving device performance, especially in the context of an AI-based device and human reader improvement.

This document is a 510(k) summary for the SmartSurgN Visualization System, which is an endoscopic imaging system. It focuses on demonstrating substantial equivalence to predicate devices, not on evaluating an AI algorithm's performance or its impact on human readers.

Specifically, the document states: "Clinical testing was not required to demonstrate substantial equivalence to the predicate devices." This directly indicates that the type of clinical or performance study involving human readers and detailed performance metrics as requested (e.g., acceptance criteria table, MRMC study, ground truth establishment) was not conducted or reported in this 510(k) submission.

While the document mentions "Non-Clinical Performance Data" including a "benchmark study with the predicate devices to assess endoscopic video imaging in visible and near-infrared conditions" and assessment in "an animal model for simulated surgical environment feedback from clinicians," these are general performance tests and not a detailed clinical study demonstrating AI-assisted performance or human reader improvement.

Therefore, most of the requested fields cannot be answered based on the provided text.

Here's what can be extracted and what cannot:

1. A table of acceptance criteria and the reported device performance

• Cannot provide. The document does not specify quantitative acceptance criteria or detailed performance results in the format of a table for an AI algorithm. It mentions "performance" and "benchmark study" but no specific metrics or targets met.

2. Sample size used for the test set and the data provenance (e.g., country of origin of the data, retrospective or prospective)

• Cannot provide. No information about test set sample size or data provenance is given for a human-in-the-loop or AI performance study. "Clinical testing was not required." The "animal model" assessment is mentioned, but details are not provided.

3. 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)

• Cannot provide. Since no clinical study or test set requiring ground truth for an AI algorithm or human reader performance is detailed, there's no information on experts or their qualifications for establishing ground truth.

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

• Cannot provide. No information on adjudication methods is present.

5. 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. The document explicitly states: "Clinical testing was not required to demonstrate substantial equivalence to the predicate devices." Therefore, an MRMC study comparing human readers with and without AI assistance was not performed or reported here.

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

• Cannot determine. While the device itself is a "Visualization System" and not explicitly an "AI algorithm," its performance characteristics are assessed. The "benchmark study with the predicate devices to assess endoscopic video imaging" suggests standalone imaging performance was tested, but not necessarily an AI algorithm's specific performance. The document does not describe an AI algorithm or its standalone performance metrics.

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

• Cannot provide. No ground truth establishment is described for a performance study of an AI algorithm.

8. The sample size for the training set

• Cannot provide. The document does not describe the development or training of an AI algorithm, or any associated training set.

9. How the ground truth for the training set was established

• Cannot provide. As no AI training set is mentioned, there's no information on how its ground truth might have been established.

Ask a specific question about this device