Search Results

RESCAN 700 provides non-contact, high resolution optical coherence tomographic (OCT) and biomicroscopic imaging of the anterior and posterior segment of the eve via an ophthalmic surgical microscope. The RESCAN 700 is indicated for in vivo viewing, axial cross sectional, and three-dimensional imaging of posterior ocular structures, including retina, macula, and optic disc, as well as imaging of anterior ocular structures, including the cornea, lens and anterior chamber angle.

RESCAN 700 uses the assistance system (CALLISTO eye) that provides non-diagnostic video documentation and image capture for ophthalmic surgeries. The assistance system allows the remote control of RESCAN 700.

RESCAN 700 brings Spectral Domain OCT technology to the ZEISS ophthalmic surgical microscopes (e.g. ARTEVO 800). Used in conjunction with the assistance system, CALLISTO eye, OCT images taken intra-operatively are presented on the monitor and may also be seen within the surgeon's oculars using the surgical microscopes integrated data injection system (IDIS). OCT images may be stored for subsequent retrieval using CALLISTO eye's data management system. RESCAN 700 can be controlled via the touch panel of the assistance system or via the foot control panel of an ophthalmic surgical microscope

This document is a 510(k) summary for the RESCAN 700 (SW 3.0), a medical device used for ophthalmic imaging. It outlines the reasons for the 510(k) submission, primarily focusing on minor software and component changes, and argues for its substantial equivalence to a previously cleared predicate device (RESCAN 700 Software Version 2.0).

Based on the provided text, the device in question (RESCAN 700 SW 3.0) is an imaging device that captures OCT (Optical Coherence Tomography) and biomicroscopic images of the eye. It is not an AI/ML-driven diagnostic device that would have acceptance criteria based on diagnostic performance metrics like sensitivity, specificity, or AUC, or studies involving human readers improving with AI assistance.

The 510(k) submission is for a software and minor component update to an already cleared device, arguing for substantial equivalence. Therefore, the "acceptance criteria" discussed in this document are primarily related to software verification and validation, electrical safety, and electromagnetic compatibility (EMC), ensuring that the updated device remains safe and effective with its existing intended use. It is not a de novo submission for a novel AI/ML algorithm.

Given this context, I will address your prompt based on the information available within the document, explaining why some of your requested points are not applicable to this type of regulatory submission and focusing on the relevant criteria and studies mentioned.

Acceptance Criteria and Study to Prove Device Meets Acceptance Criteria for RESCAN 700 (SW 3.0)

Context: The RESCAN 700 (SW 3.0) is an updated version of an already cleared ophthalmic imaging device. The 510(k) submission primarily addresses minor software changes and component updates. This is a substantial equivalence claim, not a new AI/ML diagnostic or assistive device that would typically involve comparative effectiveness studies with human readers or complex ground truth establishment for novel algorithms. The "acceptance criteria" here refer to regulatory and engineering standards rather than clinical diagnostic performance metrics for an AI algorithm.

1. Table of Acceptance Criteria and Reported Device Performance

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

This is not applicable in the context of this 510(k) summary. The document describes a software and hardware update for an imaging device, not a diagnostic AI/ML algorithm requiring a clinical test set of patient data with a specific sample size for performance evaluation (e.g., sensitivity, specificity). The "testing" referred to is primarily engineering verification and validation (V&V) of the software and hardware changes (e.g., unit testing, integration testing, system testing, safety testing), not clinical performance testing on patient data.

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

This is not applicable. As explained in point 2, this submission is for an updated imaging device, not a new AI/ML diagnostic algorithm that would require expert-established ground truth for its performance evaluation on a clinical test set. The validation focuses on engineering and regulatory compliance.

4. Adjudication Method for the Test Set

This is not applicable for the same reasons as points 2 and 3. There is no mention of "adjudication" in the context of a clinical test set for this device.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done, and the effect size of how much human readers improve with AI vs without AI assistance

No, an MRMC comparative effectiveness study was not done. The RESCAN 700 is an imaging device, providing high-resolution OCT and biomicroscopic images. It is used for in-vivo viewing, axial cross-sectional, and three-dimensional imaging of ocular structures. It is not an AI-assisted diagnostic tool designed to improve human reader performance in interpreting images. The document explicitly states that the "assistance system (CALLISTO eye)" provides "non-diagnostic video documentation and image capture."

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

This is not applicable. The RESCAN 700 is an imaging system, not a standalone AI algorithm. Its function is to acquire images, not to provide automated diagnostic interpretations that would be evaluated for standalone performance.

7. The Type of Ground Truth Used

For software verification and validation, the "ground truth" implicitly used would be the functional and performance specifications of the software and hardware. For electrical safety and EMC, the "ground truth" is adherence to international consensus standards (IEC 60601-1-2). There is no patient-data derived "ground truth" (like expert consensus, pathology, or outcomes data) mentioned for the purpose of demonstrating the device's acceptable performance, as this is related to a software update for an imaging system, not a diagnostic AI.

8. The Sample Size for the Training Set

This is not applicable. The RESCAN 700 (SW 3.0) is not an AI/ML algorithm that requires a "training set" of data in the sense of machine learning model development. The software updates are described as "minor software changes" and "changes to components and parts," indicating traditional software development, testing, and hardware modifications, not the training of a learning algorithm.

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

This is not applicable, as there is no mention of a "training set" for an AI/ML algorithm. The "ground truth" for the device's functionality and safety would be established through engineering design specifications, adherence to quality system regulations, and compliance with relevant international standards.

Ask a specific question about this device

RESCAN 700 provides non-contact, high resolution, optical coherence tomographic (OCT) and biomicroscopic imaging of the anterior and posterior segment of the eye via an ophthalmic surgical microscope. The RESCAN 700 is indicated for in vivo viewing, axial cross sectional, and three-dimensional imaging of posterior ocular structures, including retina, macula, and optic disc, as well as imaging of anterior ocular structures, including the cornea, lens and anterior chamber angle.

RESCAN 700 uses the assistance system (CALLISTO eye) that provides non-diagnostic video documentation and image capture for ophthalmic surgeries. The assistance system allows the remote control of RESCAN 700.

RESCAN 700 brings Spectral Domain OCT technology to the Zeiss OPMI Lumera 700 ophthalmic surgical microscope. Used in conjunction with the assistance system, CALLISTO eye, OCT images taken intra-operatively are presented on the monitor and may also be seen within the surgeon's oculars using the OPMI Lumera 700's integrated data injection system (IDIS). OCT images may be stored for subsequent retrieval using CALLISTO eye's data management system. RESCAN 700 can be controlled via the touch panel of the assistance system or via the foot control panel of an ophthalmic surgical microscope.

The proposed RESCAN 700 device and its primary functionality remain unchanged from the previously cleared device while including the following modifications:
Improvement of OCT Visualization
Increased Power of OCT Beam
Increased Scan Depth

RESCAN 700 is used with CALLISTO eye Software version 3.6.

CALLISTO eye Software version 3.6 is an assistance, information, and documentation system to support ophthalmic surgical procedures. CALLISTO eye software operates as an adjunct to the OPMI Lumera family of ophthalmic surgical microscopes to process real-time videos and OCT data (B-Scan images). CALLISTO eye must be installed on a touchscreen Panel PC; the Panel PC is offered as an accessory.

In conjunction with RESCAN 700 the following modifications were added to CALLISTO eye Software version 3.6.
Ability to record surgical videos directly to a connected USB hard drive.
The OCT function was enhanced by adding OCT XY-tracking for the Scan Location Marker. The functionality is similar to the Z-tracking function but works in the XY-plane instead of the Z-axis.

Here's a breakdown of the acceptance criteria and study information for the RESCAN 700 device, based on the provided FDA 510(k) summary:

1. Table of Acceptance Criteria and Reported Device Performance

The FDA 510(k) summary doesn't explicitly define "acceptance criteria" in a quantitative, pass/fail manner for the changes made to RESCAN 700. Instead, it describes general improvements and then presents a clinical study to confirm the acceptability of these changes. The core "acceptance criteria" for this submission would be that the modifications do not raise new issues of safety or effectiveness and that the upgraded device performs at least as well as, or better than, the predicate device, especially regarding image quality at new scan depths.

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

• Sample Size (Clinical Study): 22 participants
  • 11 with corneal conditions
  • 7 with glaucoma
  • 4 with retina conditions
  • A total of 84 images were obtained.
• Data Provenance: The document does not explicitly state the country of origin or whether the study was retrospective or prospective. Given it's a "clinical study" and "participants were enrolled," it suggests a prospective study. The location is not specified, but the manufacturer is Carl Zeiss Meditec AG, Germany, with a US contact, so it could be in either region or internationally.

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

• Number of Experts: Three (3) independent, masked graders.
• Qualifications: The document does not specify the exact qualifications (e.g., "radiologist with 10 years of experience"). However, given the context of ocular surgery and OCT images, it's highly probable these were ophthalmologists or similarly qualified eye care specialists with experience in interpreting OCT images. The term "graders" typically implies expertise in evaluating clinical images.

4. Adjudication Method for the Test Set

• Adjudication Method: Not explicitly stated as a formal adjudication method (like 2+1, 3+1). However, the document mentions "fair to good concordance between graders," suggesting that while there were three independent graders, a formal resolution process for disagreements is not detailed. The results likely involved an analysis of the ratings from all three for each image.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done, and its effect size

• MRMC Study: Yes, a type of MRMC study was conducted. The study compared the visualization quality of OCT images from two RESCAN 700 systems (predicate vs. modified with new software/capabilities) using three independent graders across multiple cases (22 participants, 84 images).
• Effect Size of Human Readers' Improvement with AI vs. without AI Assistance: This specific information is not provided. The study aimed to compare image quality between different device versions and scan depths, not explicitly to measure the improvement of human readers with AI assistance versus without. The CALLISTO eye system provides "assistance" (non-diagnostic video documentation, image capture, remote control, and new "Cockpits" for information visualization), but the study's focus was on the inherent image quality differences produced by the hardware/software modifications, not on a human-AI interactive improvement metric.

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

• Standalone Performance: No, a standalone algorithm performance study was not explicitly described for diagnostic purposes. The CALLISTO eye is an "assistance system" that provides "non-diagnostic video documentation and image capture." The "improvement of OCT Visualization" involved optimized standard image processing methods for noise reduction, which could be considered an algorithmic enhancement, but its performance was evaluated via the clinical study where human graders assessed the resulting images. There is no mention of the device making automated diagnoses or measurements that were evaluated in a standalone manner against a ground truth.

7. The Type of Ground Truth Used

• Ground Truth: The clinical utility and quality of the images were assessed by expert consensus/opinion (three independent, masked graders). There's no mention of pathology, long-term outcomes data, or other objective measures being used for ground truth in this specific image quality study.

8. The Sample Size for the Training Set

• Training Set Sample Size: This information is not provided in the document. The general description focuses on the verification and validation of the modified device, not on the development or training of specific machine learning models within the device that would require a distinct training set. The "optimized image processing methods for noise reduction" likely involved internal development and testing, but details on a formal training set for this are not disclosed.

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

• Training Set Ground Truth: As the training set size is not provided, how its ground truth was established is also not described. If "optimized image processing methods" involved machine learning or AI, the ground truth for such internal training would have typically involved expert-labeled images or simulations. However, this level of detail is absent from the 510(k) summary.

Ask a specific question about this device

RESCAN 700 provides non-contact, high resolution, optical coherence tomographic (OCT) and biomicroscopic imaging of the anterior and posterior segment of the eye via an ophthalmic surgical microscope. The RESCAN 700 is indicated for in vivo viewing, axial cross sectional, and three-dimensional imaging of posterior ocular structures, including retina, macula, and optic disc, as well as imaging of anterior ocular structures, including the cornea, lens and anterior chamber angle.

RESCAN 700 uses an assistance system (CALLISTO eye) that provides non-diagnostic video documentation and image capture for ophthalmic surgeries. The assistance system allows the remote control of RESCAN 700.

RESCAN 700 brings Spectral Domain OCT technology to the Zeiss OPMI Lumera 700 ophthalmic surgical microscope. Used in conjunction with software version 3.2 of an assistance system, CALLISTO eye, OCT images taken intra operatively are presented on the monitor and may also be seen within the surgeon's oculars using the OPMI Lumera 700's integrated data injection system (IDIS). OCT images may be stored for subsequent retrieval using CALLISTO eye's data management system. RESCAN 700 can be controlled via the touch panel of the assistance system or via the foot control panel of an ophthalmic surgical microscope.

RESCAN 700 is used with CALLISTO eye software version 3.2.

I am sorry, but based on the provided text, there is no information about acceptance criteria or a study that proves the device meets those criteria. The document is a 510(k) summary for the RESCAN 700 device, which focuses on demonstrating substantial equivalence to predicate devices rather than providing detailed performance studies against specific acceptance criteria.

The "PERFORMANCE DATA" section (page 7) briefly mentions that:

• RESCAN 700 was designed and tested to applicable standards for electrical and optical safety.
• Performance testing was conducted, and the device was found to perform as intended.
• Each function/feature was tested by an appropriate test case or specification.
• Testing demonstrated conformance to IEC 60601-1 and IEC 60601-1-2 standards.
• Software verification activities included tests accompanying development, integration tests, and system verification.
• Validation and usability testing were conducted with RESCAN 700 in conjunction with CALLISTO eye to ensure it meets product and user requirements.

However, this section does not provide:

• A table of specific acceptance criteria.
• Reported device performance against numerical targets.
• Details about sample sizes, data provenance, ground truth establishment, or expert involvement for any specific performance study.
• Information on MRMC comparative effectiveness studies or standalone performance.

Therefore, I cannot fulfill your request for a table of acceptance criteria and the study proving the device meets them based on the text provided.

Ask a specific question about this device

ResScan is intended to augment the standard follow-up care of patients by providing transfer of machine and therapeutic information. This includes the ability to remotely change settings in non-life support devices only.

It is intended to be used by Clinicians in conjunction with ResMed compatible therapy devices, using ResMed's proprietary communications protocol.

The performance and functional characteristics of ResScan includes all the user friendly features of the predicate device.

ResScan allows the clinician to:

• Download and view patient and machine data from ResMed flow generators
• Store patient details
• Set machine parameters (Using Removal Media or PC direct connection), non-life support devices only.
• Create and print reports
• Uses Removal Media or PC direct connection as the interface between the flow generator and ResScan
• Support for Data Card Reader

The provided document describes a 510(k) submission for the ResScan device, which is a software intended to augment patient follow-up care by transferring machine and therapeutic information and allowing remote changes to settings in non-life support devices. This submission aims to demonstrate substantial equivalence to a predicate device, also named ResScan (K113815).

Here's an analysis of the acceptance criteria and study information:

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly present a table of quantitative acceptance criteria for specific performance metrics (e.g., accuracy, sensitivity, specificity) with corresponding reported device performance values. Instead, it discusses "predetermined acceptance criteria" in the context of verification testing.

The acceptance criteria appear to be qualitative and focused on successful communication and data transfer. The reported device performance is that these criteria were met.

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

The document describes "End-to-End bench testing" and "non-clinical verification activities." These refer to laboratory or bench testing rather than testing with patient data or in a clinical setting.

• Sample Size for Test Set: Not applicable in the context of patient data. The "test set" would consist of various configurations, data types, and scenarios used in the bench testing. No numerical sample size for these test cases is provided.
• Data Provenance: Not applicable as the testing was non-clinical/bench testing. The document states that ResScan "only obtains patient and machine information from therapeutic devices for which clinical trials have already been conducted, or compared with previous predicate comparison test results." This implies reliance on data generated by the therapeutic devices themselves, not data specifically collected for ResScan's performance evaluation.

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

• Number of Experts: Not applicable. Given that the testing was end-to-end bench testing of data transfer and settings changes, the "ground truth" would be established by the expected behavior of the system and direct observation/validation during engineering testing. There is no mention of human experts defining ground truth for this type of technical performance evaluation.

4. Adjudication Method for the Test Set

• Adjudication Method: Not applicable. The testing described is technical verification (e.g., "end-to-end testing to confirm that settings were successfully transferred," "data captured by the flow generator was sent"). Adjudication methods like 2+1 or 3+1 are typically used for clinical image interpretation or diagnostic tasks where expert consensus is needed to resolve discrepancies in human assessments of often ambiguous data. Here, the outcome of a test case (e.g., "was the setting changed correctly?") is expected to be binary and objectively verifiable.

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

• Was it done?: No. The document explicitly states: "Clinical testing was not deemed necessary as identified in the Risk Analysis, as ResScan only obtains patient and machine information from therapeutic devices for which clinical trials have already been conducted, or compared with previous predicate comparison test results. Accordingly no clinical testing is required."
• Effect size of human readers with vs. without AI assistance: Not applicable, as no MRMC study was conducted. ResScan is a data management and setting adjustment software, not an AI-assisted diagnostic or interpretation tool.

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

• Was it done?: Yes, in essence. The "end-to-end bench testing" described is a standalone performance test of the software's ability to correctly transfer data and settings. The performance evaluated is the software's internal functionality. While it is ultimately used by clinicians, the testing described focuses on the software's direct interaction with the flow generators and its own data processing capabilities, without evaluating a human-in-the-loop task.

7. Type of Ground Truth Used

• Type of Ground Truth: Expected System Behavior/Technical Specification Adherence.
  • For settings transfer: The ground truth is whether the new setting was correctly applied to the flow generator as initiated by ResScan.
  • For data capture and transmission: The ground truth is whether the data generated by the flow generator was accurately received and stored by ResScan.
  • This is established by comparing the software's output/actions against the design specifications and expected functional behavior, not against pathology, outcomes data, or expert consensus in a clinical diagnostic sense.

8. Sample Size for the Training Set

• Sample Size for Training Set: Not applicable. This document describes a software system that manages data and settings, and interfaces with medical devices. It is not an AI/machine learning model that typically requires a distinct "training set" for model development. The software capabilities are likely developed through traditional software engineering paradigms, not statistical learning from a large dataset.

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

• How Ground Truth for Training Set Was Established: Not applicable, as there isn't a "training set" in the machine learning sense. The "ground truth" for developing the software (e.g., for unit and integration testing) would be derived from functional requirements, design specifications, and communication protocols.

Ask a specific question about this device

The ResScan software is intended to be used by clinicians with ResMed flow generators that have software incorporating ResMed's proprietary communication protocol. ResScan is used to download and view therapy data, as well as store therapy information and print reports.

The ResScan software with its capabilities of data collection and display of patient information provides an adjunct for patient compliance management. ResScan allows the clinician to download and view data from a ResMed CPAP or Bilevel device. The data displayed will depend on the ventilator used. Parameters displayed may include one or more of the following: usage (hrs: min); respiratory rate (breaths/min); tidal volume (L/min); leak (L/min); minute ventilation (L/min); apnea and hypopnea index including apnea index (AI), hypopnea index (HI) and AHI (events/hr); oxygen saturation SpO2 (%); pulse rate (bpm);. The software allows the creation of reports on patient details, and downloaded treatment data. ResScan displays data as retrieved. Data processing in ResScan is limited to computing the statistics (e.g., median) on retrieved data. The software operates under Windows 98SE, Windows 2000 SP2, Windows NT4 SP6, Windows XP, and is available on CD-ROM.

Here's an analysis of the provided text, focusing on the acceptance criteria and study information for the ResScan device:

1. Table of Acceptance Criteria and Reported Device Performance:

The provided document does not explicitly list quantitative acceptance criteria for the ResScan software. Instead, the performance data section states that the software was tested in accordance with FDA guidance documents and "passed all tests." This implies that the acceptance criteria were met, but the specific metrics are not detailed.

Given the device's function as data download and display software, the implied "performance" likely relates to:

• Accuracy of Data Display: The displayed data (usage, respiratory rate, tidal volume, leak, minute ventilation, AHI, SpO2, pulse rate) should accurately reflect the data downloaded from the ResMed CPAP or Bilevel device.
• Functionality: The software should reliably perform its stated functions: downloading data, viewing data, storing data, and printing reports.
• Compatibility: The software should operate correctly on the specified Windows operating systems and be compatible with ResMed flow generators using their proprietary communication protocol.
• Security/Integrity: Data downloaded and stored should not be corrupted or altered by the software.

2. Sample Size Used for the Test Set and Data Provenance:

The document provides no specific information regarding the sample size used for the test set or the provenance of the data (e.g., country of origin, retrospective/prospective). The statement "The ResScan software passed all tests" is a high-level summary without granular details.

3. Number of Experts Used to Establish Ground Truth and Qualifications:

The document provides no information on the use of experts or the establishment of ground truth in the context of typical clinical validation studies for diagnostic or AI devices. Given that ResScan is primarily data download and display software, the "ground truth" would likely be the data as logged by the ventilator, and testing would focus on the software's ability to accurately retrieve and present that data, rather than clinical interpretation by experts.

4. Adjudication Method:

The document provides no information on any adjudication methods.

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

The document provides no information on an MRMC comparative effectiveness study, nor would it be expected for this type of device. The device's function is to display data, not to assist human readers in making a diagnosis or interpretation in a way that would require an MRMC study to measure improved effectiveness.

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

For the ResScan software, the concept of "standalone" performance is inherent to its function. The software is designed to operate without human intervention during the data download and display process itself. Tests would assess if the software accurately processes and presents the data as designed, without a human "in the loop" for that specific function. However, the document does not describe the methodology or results of such a standalone study in detail, beyond stating that "all tests" were passed and it's substantially equivalent.

7. Type of Ground Truth Used:

The fundamental "ground truth" for this device would be the raw therapy data generated and logged by the ResMed CPAP or Bilevel flow generators. The software's performance is measured against its ability to accurately download, display, and store this pre-existing data. It's not a diagnostic device that interprets signals to establish a medical ground truth (like a pathology report or clinical outcome).

8. Sample Size for the Training Set:

The concept of a "training set" is typically relevant for machine learning or AI-driven devices. The ResScan software, as described, appears to be a rule-based or algorithmic data processing and display tool, not a machine learning model. Therefore, no "training set" information is provided or would be expected.

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

As there is no mention or indication of a training set, the question of how ground truth for it was established is not applicable.

Ask a specific question about this device