Search Results

a. Endoscope Model EB-710P
Endoscope Model EB-710P is a bronchoscope intended for the observation, diagnosis and endoscopic treatment of the trachea and bronchus at medical facilities under the management of physicians. Never use this product for any other purposes.

b. Processor EP-8000
The EP-8000 is an endoscopic processor with an integrated light source that is intended to provide illumination, process electronic signals transmitted from a video endoscope and enable image recording.

This product can be used in combination with compatible medical endoscope, a monitor, a recorder and various peripherals.

It is used for endoscopic observation, diagnosis and treatment.

a. Endoscope Model EB-710P
FUJIFILM Endoscope Model EB-710P is comprised of three general sections: an insertion portion, a control portion, and a connector portion to the peripherals. The insertion portion is flexible and contains glass fiber bundles, several channels, and a complementary metal-oxide semiconductor (CMOS) image sensor in its distal end. The glass fiber bundles allow light to travel through the endoscope and emit light from the tip of the insertion portion to illuminate the body cavity. This provides enough light to the CMOS image sensor to capture an image and display it on the monitor. The channels in the insertion portion assist in delivering suction as well as endoscopic accessories. The control portion controls the angulation and rotation of the bending portion in the insertion portion. The connector portion consists of electronic components needed to operate the endoscope when connected to the video processor. The endoscopes are used in combination with FUJIFILM's video processors, light sources, and peripheral devices such as monitor, printer, foot switch, and cart.

b. Processor EP-8000
FUJIFILM Video Processor EP-8000 is intended to provide illumination, process electronic signals transmitted from a video endoscope and enable image recording.

FUJIFILM Video Processor EP-8000 relays the image from an endoscope to a video monitor. The projection can be either analog or digital at the user's preference. The processor employs fiber bundles to transmit light from four LED lamps (Violet, Blue, Green, and Amber), with a total power of 79.2W lamps, to the body cavity.

The device is AC operated at a power setting of 100-240V/50-60Hz/ 3.0-1.5A. The processor is housed in a steel-polycarbonate case measuring 395×210×515mm

The provided FDA 510(k) clearance letter pertains to the FUJIFILM Endoscope Model EB-710P and Processor EP-8000. It details the substantial equivalence of these devices to their predicates based on non-clinical testing.

However, the provided text does not contain information about acceptance criteria or a study that uses a test set to prove the device meets those criteria.

The document primarily focuses on:

• Product identification: Device names, regulation numbers, product codes.
• Regulatory details: FDA clearance status, general controls, and compliance requirements.
• Substantial equivalence justification: Comparison of intended use, technological characteristics, and principles of operation between the new devices and their predicates.
• Non-clinical testing: A list of engineering tests performed (e.g., electrical safety, software validation, color and optical performance, image quality assessments like reproduction, geometric distortion, resolution, depth of field, ISO-SNR, dynamic range, intensity uniformity, and field of view).

Therefore, I cannot fulfill the request to describe the acceptance criteria and the study that proves the device meets them, as this information is not present in the provided FDA clearance letter.

To answer your request, I would need a document (e.g., a summary of safety and effectiveness, or a clinical study report) that explicitly defines:

1. Acceptance criteria: Quantitative thresholds or qualitative statements that define successful device performance.
2. Reported device performance: The actual outcomes measured during the study.
3. Test set details: Sample size, data provenance, ground truth establishment (experts, adjudication, type of ground truth).
4. Training set details: Sample size, ground truth establishment.
5. MRMC study information: If applicable, whether human readers improved with AI assistance and by how much.
6. Stand-alone algorithm performance: If an algorithm-only study was conducted.

The provided document only states that "EP-8000 demonstrated substantial equivalence to VP-7000 and BL-7000 in Image performance and color reproduction" for the listed non-clinical tests, implying that the new device performed as well as the predicate for these specific engineering parameters, but it does not provide the specific performance values or the acceptance thresholds for these parameters. It also makes no mention of AI assistance or human reader studies.

Ask a specific question about this device

FUJIFILM Endoscope Models EG-840T and EG-840TP is intended for the visualization of the upper digestive tract, specifically for the observation, diagnosis, and endoscopic treatment of the esophagus, stomach, and duodenum.

FUJIFILM Endoscope Model EG-840N is intended for the visualization of the upper digestive tract, specifically for the observation, diagnosis, and endoscopic treatment of the esophagus, stomach, and duodenum. This product can be inserted orally or nasally.

Never use this product for any other purposes.

The insertion portion of the device has a mechanism (hereinafter "the bending portion") which bends the tip from right to left and up and down, and a flexible tube (hereinafter "the flexible portion") consists of the bending portion and operating portion with a knob which controls the bending portion. The forceps channel which runs through the operating portion to the tip is arranged inside the insertion portion for inserting the surgical instrument. The insertion portion of the endoscopes comes into contact with the mucosal membrane.

The tip of the insertion portion is called the "Distal end" which contains the Imaging section, Distal cap, Objective lens, Air/water nozzle, Water jet nozzle (Except EG-840N), Instrument channel outlet, and Light guide. The bending portion is controlled by knobs on the control portion/operation section to angulate the distal end to certain angles.

The Flexible portion refers to the long insertion area between the Bending portion and the Control portion (a part of Non-insertion portion). This portion contains light guides (glass fiber bundles), air/water channels, a forceps/suction channel, a CrOS image sensor, and cabling. The glass fiber bundles allow light to travel through the endoscope to illuminate the body cavity, thereby providing enough light to the CMOS image sensor to capture an image and display the image on a monitor. The forceps channel is used to introduce biopsy forceps and other endoscopic accessories, as well as providing suction.

The control portion/operating section provides a grip to grasp the endoscopes and contains mechanical parts to operate the endoscopes. This section includes a Forceps inlet, which allows endoscope accessories to be introduced. The Scope connector connects the endoscopes to the light source.

The provided FDA 510(k) clearance letter pertains to endoscopes, which are hardware devices, not AI/ML software. Therefore, the information requested regarding acceptance criteria, study details, and data provenance for an AI/ML device is not explicitly present in the provided text.

The document focuses on demonstrating substantial equivalence to a predicate device through various non-clinical tests (electrical safety, biocompatibility, endoscope-specific testing, software validation, and reprocessing validation) rather than a clinical effectiveness study involving human readers or AI performance metrics.

However, I can extract information about the device's performance specifications that were tested to prove its general functionality and safety, which can be interpreted as fulfilling certain "acceptance criteria" for a physical medical device.

Here's a breakdown of the available information based on your request, with the caveat that it does not directly address AI/ML performance:

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

The document lists various performance specifications that the device met. It does not provide specific numerical acceptance criteria alongside reported performance values in a table format. Instead, it states that "The subject device met performance specifications in the following additional testing." This implies that the device did meet predefined internal thresholds for these parameters.

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

• Sample Size: Not specified for any of the tests. The document mentions "testing" or "validation activities" without detailing the number of devices or trials involved.
• Data Provenance: Not specified. As this is a 510(k) for a physical medical device (endoscope), the testing would typically be conducted in a laboratory setting by the manufacturer, rather than involving patient data in the context of imaging performance.

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

Not applicable. The tests performed are engineering and safety validations. There is no mention of human expert involvement for establishing "ground truth" in the context of image interpretation or diagnosis, as this is not an AI/ML diagnostic tool.

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

Not applicable. This is a concept typically used in clinical studies involving interpretation by multiple readers, which is not described here.

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 is not an AI-assisted device. The document describes a traditional endoscope.

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

Not applicable. The device itself is an endoscope, not an algorithm. Software validation was conducted (IEC 62304:2015), but this relates to the software controlling the endoscope's functions, not a standalone diagnostic AI.

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

For the listed performance specifications (field of view, resolution, etc.), the "ground truth" would be established by objective physical measurements using calibrated equipment and engineering standards. For biocompatibility, it's based on biological response to materials, and for reprocessing, it's based on sterility and decontamination efficacy. There is no "ground truth" in the diagnostic sense as there would be for an AI/ML algorithm.

8. The sample size for the training set

Not applicable. This is not an AI/ML device that requires a training set.

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

Not applicable.

Ask a specific question about this device

This product is intended for the visualization of the lower digestive tract, specifically for the observation, diagnosis, and endoscopic treatment of the rectum and large intestine.

The insertion portion of the device has a bending mechanism and a flexible tube consisting of the bending portion and an operating portion with a knob to control the bending. A forceps channel runs through the operating portion to the tip for inserting surgical instruments. The insertion portion's tip, called the "Distal end," contains the Imaging section, Distal cap, Objective lens, Air/water nozzle, Water jet nozzle, Instrument channel outlet, and Light guide. The bending portion is controlled by knobs on the control portion/operation section. The flexible portion, between the bending and control portions, contains light guides (glass fiber bundles), air/water channels, a forceps/suction channel, a CMOS image sensor, and cabling. The control portion/operating section provides a grip and mechanical parts to operate the endoscopes, including a Forceps inlet. The Scope connector links the endoscopes to the light source and video processor.

The provided FDA 510(k) clearance letter and summary for the FUJIFILM Endoscope Models EC-860P/M, EC-860P/L, and EC-860S/L primarily focus on demonstrating substantial equivalence to predicate devices through bench testing and compliance with various consensus standards. It does not describe a clinical study in the traditional sense, where device performance is measured against specific acceptance criteria in a human subject population using metrics like sensitivity, specificity, or accuracy, often seen with AI or diagnostic imaging devices.

Instead, the submission relies on demonstrating that the new endoscope models meet established performance specifications and safety standards through non-clinical testing.

Here's an analysis based on the provided text, addressing your questions where information is available:

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly state numerical acceptance criteria for each item. Instead, it refers to the subject device meeting "performance specifications" or demonstrating "substantially equivalent in performance to the predicate devices" for various parameters.

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

No human or patient test set is described. The "test set" consists of the physical endoscope models themselves, subjected to various bench tests and standard compliance evaluations. The data provenance is derived from these non-clinical tests performed by the manufacturer, rather than 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 is not a study requiring expert readers to establish ground truth for clinical performance. The "ground truth" for the non-clinical tests is established by the specifications and standards themselves, and verified by testing personnel.

4. Adjudication method for the test set

Not applicable. There is no expert adjudication process described for clinical interpretation. The compliance with standards and performance specifications is determined through objective measurements and validated test methods.

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 an endoscope, not an AI-powered diagnostic system. No MRMC study or AI assistance is mentioned. The clearance is for the physical endoscope models.

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

Not applicable. There is no AI algorithm involved in this device submission.

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

For this submission, the "ground truth" is defined by:

• Consensus standards: e.g., electrical safety (ANSI/AAMI ES 60601-1), biocompatibility (ISO 10993-1), medical device software (IEC 62304), endoscope-specific standards (ISO 8600 series).
• Manufacturer's internal performance specifications: These are the benchmarks against which specific performance parameters (e.g., field of view, resolution, bending capability) are measured.
• Predicate device performance: The "bench testing data regarding 'Optical performance' demonstrated that the subject devices are substantially equivalent in performance to the predicate devices." This implies that the performance of the predicate devices serves as a comparative ground truth for equivalence.

8. The sample size for the training set

Not applicable. This is not a machine learning or AI device that requires a training set.

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

Not applicable. No training set is described.

Ask a specific question about this device

Synapse 3D Base Tools is medical imaging software that is intended to provide trained medical professionals with tools to aid them in reading, interpreting, reporting, and treatment planning. Synapse 3D Base Tools accepts DICOM compliant medical images acquired from a variety of imaging devices including, CT, MR, CR, US, NM, PT, and XA, etc.

This product is not intended for use with or for the primary diagnostic interpretation of Mammography images. Synapse 3D Base Tools provides several levels of tools to the user: Basic imaging tools for general images, including 2D viewing, volume rendering and 3D volume viewing, orthogonal / oblique / curved Multi-Planar Reconstructions (MPR), Maximum (MIP), Average (RaySum) and Minimum (MinIP) Intensity Projection, 4D volume viewing, image fusion, image subtraction, surface rendering, sector and rectangular shape MPR image viewing, MPR for dental images, creating and displaying multiple MPR images along an object, time-density distribution, basic image processing, noise reduction, CINE, measurements, annotations, reporting, printing, storing, distribution, and general image management and administration tools, etc.

• Tools for regional segmentation of anatomical structures within the image data, path definition through vascular and other tubular structures, and boundary detection.
• Image viewing tools for modality specific images, including CT PET fusion and ADC image viewing for MR studies.
• Imaging tools for CT images including virtual endoscopic viewing, dual energy image viewing.
• Imaging tools for MR images including delayed enhancement image viewing, diffusion-weighted MRI image viewing.

The intended patient population for all applications implemented as base tools is limited to adult population (over 22 years old).

The 3D image analysis software Synapse 3D Base Tools (V7.0) is medical application software running on Windows server/client configuration installed on commercial general-purpose Windows-compatible computers. It offers software tools which can be used by trained professionals to interpret medical images obtained from various medical devices, to create reports, or to develop treatment plans.

The provided text details the FDA 510(k) clearance for Synapse 3D Base Tools (V7.0). It primarily focuses on demonstrating substantial equivalence to a predicate device and includes information on nonclinical and certain clinical performance testing for newly added deep-learning-based organ segmentation features.

Here's an analysis of the acceptance criteria and study that proves the device meets them, based on the provided text:

Acceptance Criteria and Reported Device Performance

The core of acceptance criteria for this 510(k) submission appears to be demonstrating substantial equivalence to a predicate device (Synapse 3D Base Tools V6.6 K221677) and proving the safety and effectiveness of new features, particularly those utilizing deep learning for automatic or semi-automatic organ extraction.

While no explicit "acceptance criteria" table is provided in the document in terms of specific thresholds for the overall device functionality, the performance section for the deep learning models serves as such for those specific features. The acceptance criterion for these features is implicitly showing a high Dice Similarity Coefficient (DICE) score, indicating strong agreement between the automated segmentation and the ground truth.

Table of Acceptance Criteria and Reported Device Performance (for Deep Learning Segmentation)

Note: For items marked with an asterisk (*), the extraction is performed semi-automatically. All others are executed automatically. The acceptance criterion is "High DICE," as no specific quantitative threshold is given, but the reported values generally indicate good agreement. "Additional distance based metrics 95% Hausdorff Distance and Mean Surface Distance were also reported along with the subgroup analysis. Detailed results are reported in the labeling."

Study that Proves the Device Meets Acceptance Criteria

The study described is a performance testing for the new deep-learning-based automatic or semi-automatic organ extraction functions.

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

   • Sample Size: 1086 cases were collected for performance testing.
   • Data Provenance: The data was collected newly from US patient populations across various regions: US_East (295 cases), US_Midwest (175 cases), US_Southeast (185 cases), US_Southwest (73 cases), and US_Northwest (4 cases). This indicates a prospective data collection specifically for this testing, originating from the US. The text also mentions the test data is "independence from training data."
   • Demographics: The test set included 672 men, 414 women, and a range of ages from 22 to 120+ years old. Modalities covered CT and MRI from various major manufacturers (SIEMENS, GE, PHILIPS, CANON, FUJIFILM).

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

   • The document does not specify the number of experts or their qualifications used to establish the ground truth. It only states that the performance testing used an "average DICE" score, implying a comparison against some form of expertly derived ground truth.

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

   • The document does not specify any adjudication method for establishing the ground truth.

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

   • No, an MRMC comparative effectiveness study was not explicitly mentioned or performed as part of this submission for demonstrating substantial equivalence. The clinical testing mentioned focused on the standalone performance of the new deep learning features (i.e., automatic or semi-automatic segmentation accuracy) rather than human reader performance with or without AI assistance. The submission states, "The subject of this 510(k) notification, Synapse 3D Base Tools does not require clinical studies to support safety and effectiveness of the software."

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

   • Yes, a standalone performance evaluation was done for the automatic (and semi-automatic) deep learning segmentation functions. The Dice Similarity Coefficient (DICE) scores provided are a measure of the algorithm's performance in segmenting anatomical structures compared to a ground truth, without human intervention in the segmentation process itself, although some extractions are noted as "semi-automatic" where human interaction would refine the output.

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

   • The text implies the ground truth for the segmentation tasks was established by expert consensus/manual annotation (as DICE is a metric comparing algorithmic output to a reference segmentation, typically derived from expert outlines). However, the specific method (e.g., single expert, multi-expert consensus) is not detailed. It mentions "Additional distance based metrics 95% Hausdorff Distance and Mean Surface Distance were also reported," which are also used for comparing segmentation masks to a ground truth.

7. The sample size for the training set:

   • The document does not explicitly provide the sample size for the training set. It only states that the test data was "independence from training data," implying a separate training dataset was used.

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

   • The document does not provide details on how the ground truth for the training set was established. However, for deep learning segmentation, it is typically established through manual annotation by qualified experts.

Ask a specific question about this device

The SCENARIA View system is indicated to acquire axial volumes of the whole body including the head. Images can be acquired in axial, helical, or dynamic modes. The SCENARIA View system can also be used for interventional needle guidance. Volume datasets acquired by a SCENARIA View system can be post-processed in the SCENARIA View system to provide additional information. Post-processing capabilities of the SCENARIA View software include multi-planar reconstruction (MPR), and volume rendering. Volume datasets acquired by a SCENARIA View system can be transferred to external devices via a DICOM standard interface.

The Low Dose CT Lung Cancer Screening Option for the SCENARIA View system is indicated for using low dose CT for lung cancer screening. The screening must be conducted with the established program criteria and protocols that have been approved and published by a governmental body, a professional medical society, and/or FUJIFILM Corporation.

The SCENARIA View system is intended for general populations.

The subject device SCENARIA View is a multi-slice CT system consists of a gantry, operator's workstation, patient table, high-frequency X-ray generator, and accessories. The system performance is similar to the predicate device.

The SCENARIA View system uses 128-slice CT technology, where the X-ray tube and detector assemblies are mounted on a frame that rotates continuously around the patient using slip ring technology. The solid-state detector assembly design collects up to 64 slices of data simultaneously. The X-ray sub-system features a high frequency generator, X-ray tube, and collimation system that produces a fan beam X-ray output. The system can operate in a helical (spiral) scan mode where the patient table moves during scanning. As the X-ray tube/detector assembly rotates around the patient, data is collected at multiple angles.

The collected data is then reconstructed into cross-sectional images by a high-speed reconstruction sub-system. The images are displayed on a Computer Workstation, stored, printed, and archived as required. The workstation is based on current PC technology using the Windows™ operating system.

Compared to the predicate device referenced within this submission, the subject devices support the following modifications:

1. New features

• AutoPose is an AI-based function that recognizes a specific body part in an image of localization scan and then automatically sets the scan range and the image reconstruction range.
• RemoteRecon is a function of setting image reconstruction parameters that runs on the external personal computer (hereinafter referred to as "PC") connected to the CT system.

2. Modified features

• The maximum load capacity of patient table type has been increased from 250kg to 300 kg.
• Motion corrected reconstruction is an image reconstruction feature that reduces motion artifacts. The feature has been modified to include applicability for chest examinations, which is a non-gated scan.
• AutoPositioning is a feature that assist in positioning the patient by camera images. The feature has been modified to include additional 12 body parts (Head and Neck, Neck, C-spine, Heart, Chest-Abdomen, Chest-Upper Abdomen, Abdomen-Pelvis, Abdomen, Pelvis, T-spine, L-spine, T-L-spine), in addition to the 2 body parts (Head, Chest) of the predicate device, with scanogram ranges displayed according to the selected protocol.

The provided FDA 510(k) Clearance Letter for SCENARIA View Phase 5.0 primarily focuses on demonstrating substantial equivalence to a predicate device (SCENARIA View 4.2). The document outlines non-clinical and some clinical tests, but it does not present a formal "acceptance criteria" table with specific quantitative metrics for the device performance of the new AI features (AutoPose, Body Still Shot) in the same way one might find for a novel AI/CADe device.

The "acceptance criteria" for this submission appear to be centered around workflow improvement and sufficient image quality when compared to manual or predicate methods, rather than hard quantitative performance targets. The study designs are more akin to usability studies and qualitative image reviews.

Here's an attempt to extract and interpret the information based on your requested structure, acknowledging the limitations of the provided text in terms of explicit acceptance criteria and standalone performance metrics for the AI components.

Acceptance Criteria and Device Performance for SCENARIA View Phase 5.0 (AI Components)

The provided document describes the acceptance criteria and study results for the new features AutoPose and Body Still Shot introduced in the SCENARIA View Phase 5.0 system. The acceptance criteria are largely qualitative, focusing on workflow improvement and sufficient image quality.

1. Table of Acceptance Criteria and Reported Device Performance

Note on "Acceptance Criteria": The document does not explicitly list quantitative acceptance criteria in a table format for the AI features. The criteria listed above are inferred from the Methods and Results sections of the non-clinical and clinical tests described. The primary goal was to demonstrate workflow efficiency (AutoPose) and qualitative image improvement (Body Still Shot).

2. Sample Sizes and Data Provenance

• AutoPose (Clinical Test):

  • Total Cases: 50 (Head), 50 (Neck), 52 (Chest), 54 (Heart), 52 (Abdomen), 52 (Abdomen-Pelvis), 50 (Chest-Abdomen), 24 (T-L-Spine), 50 (C-Spine), 50 (T-Spine), 50 (L-Spine).
    • Note: The table layout in the original document makes it unclear if Chest-Upper Abdomen had cases listed, but it's empty. Assuming 50 for Chest-Abdomen and 0 for Chest-Upper Abdomen as it's not specified.
  • Total Sum (if all distinct): 50 + 50 + 52 + 54 + 52 + 52 + 50 + 50 + 50 + 50 + 24 = 504 cases.
  • Data Provenance: Clinical sites in the USA.
  • Retrospective/Prospective: Not explicitly stated, but the nature of evaluating steps in a procedure suggests it was likely a prospective workflow evaluation with certified technologists.

• Body Still Shot (Clinical Test):

  • Total Cases: Not specified.
  • Data Provenance: Not specified (only mentions "Japanese M.D." reviewers).
  • Retrospective/Prospective: Not specified.

• Training Set Sample Size:

  • Not disclosed in the provided document. The document primarily details the validation/test set.

3. Number of Experts and Qualifications for Ground Truth

• AutoPose (Clinical Test):

  • Number of Experts: Not explicitly stated how many "certified radiological technologists" performed the evaluations, only that they were certified.
  • Qualifications: "certified radiological technologists." No specific years of experience or other details are provided.
  • Role in Ground Truth: Their assessment of the number of steps and the "expected position" for manual adjustment served as the comparison for AutoPose's performance.

• Body Still Shot (Clinical Test):

  • Number of Experts: Not explicitly stated how many "Japanese M.D." (Medical Doctors) reviewed the images, only that they were "Japanese M.D."
  • Qualifications: "Japanese M.D." No specific specialty (e.g., radiologist), years of experience, or other details are provided.
  • Role in Ground Truth: Their qualitative review ("evaluated to be able to obtain images of sufficient quality") served as the ground truth for image quality.

4. Adjudication Method for the Test Set

• AutoPose: Not explicitly stated. The results imply a direct comparison of workflow steps, but it's not mentioned if multiple technologists evaluated the same cases or how discrepancies were handled.
• Body Still Shot: Not explicitly stated how reviews were conducted if multiple M.D.s were involved (e.g., consensus, majority vote).

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

• Was it done?: No, a formal MRMC comparative effectiveness study demonstrating how human readers improve with AI vs. without AI assistance was not performed or reported for either AutoPose or Body Still Shot.
• Effect Size: Not applicable, as no such study was presented. The studies were focused on workflow efficiency (AutoPose) and qualitative image quality (Body Still Shot).

6. Standalone (Algorithm Only) Performance

• Was it done?: The document does not describe a standalone AI performance study (e.g., precision, recall, F1-score for AutoPose's pose recognition; or a quantitative image quality metric for Body Still Shot). The evaluation of AutoPose was focused on the workflow impact, and Body Still Shot on perceived image quality by human reviewers.

7. Type of Ground Truth Used

• AutoPose:
  • For Scan Range Setting: The "ground truth" or reference for the AutoPose evaluation was the manual scan range setting process and the expected optimal scan position as determined by certified radiological technologists. The metric was a reduction in the number of workflow steps.
• Body Still Shot:
  • For Image Quality: The ground truth for image quality was based on the qualitative assessment and review by "Japanese M.D." to determine "sufficient quality." This is essentially expert consensus on image usability.

8. Sample Size for the Training Set

• The document does not provide information on the sample size used for training the AI models (AutoPose and Body Still Shot).

9. How Ground Truth for the Training Set was Established

• The document does not provide information on how the ground truth for the training set was established for the AI models.

Ask a specific question about this device

a. Processor EP-8000

• The EP-8000 is an endoscopic processor with an integrated light source that is intended to provide illumination, process electronic signals transmitted from a video endoscope and enable image recording.
• This product can be used in combination with compatible medical endoscope, a monitor, a recorder and various peripherals.
• It supplies air through the endoscope, for obtaining clear visualization and is used for endoscopic observation, diagnosis and treatment.
• BLI (Blue Light Imaging), LCI (Linked Color Imaging), ACI (Amber-red Color Imaging) and FICE (Flexible spectral-Imaging Color Enhancement) are adjunctive tools for gastrointestinal endoscopic examinations which can be used to supplement Fujifilm white light endoscopy. BLI, LCI, ACI and FICE are not intended to replace histopathological sampling as a means of diagnosis.

b. Endoscope Model EG-860R

This product is intended for the visualization of the upper digestive tract, specifically for the observation, diagnosis, and endoscopic treatment of the esophagus, stomach, and duodenum.

FUJIFILM Video Processor EP-8000 is intended to provide illumination, process electronic signals transmitted from a video endoscope and enable image recording.

FUJIFILM Video Processor EP-8000 relays the image from an endoscope to a video monitor. The projection can be either analog or digital at the user's preference. The processor employs fiber bundles to transmit light from four LED lamps (Violet, Blue, Green, and Amber), with a total power of 79.2W lamps, to the body cavity.

The EP-8000, like the VP-7000 and BL-7000, has additional image processing options called BLI, BLI-bright, and LCI that provide endoscopic assistance for white light imaging (WLI). There is also an additional image processing option called "ACI"(Amber-red Color Imaging).

ACI is an image processing function that simultaneously emphasizes the brightness and color difference of red information in endoscopic images and serves as an adjunct to white light imaging (WLI).

Compared to WLI mode, ACI relatively increases the ratio of amber red light and decreases the ratio of violet light.

Relatively high-saturation red information such as blood-like red in the image signal digitized by the camera unit is enhanced by signal processing.

The EP-8000 also has a Multi Observation option that allows endoscopic images to be displayed in the main screen area and sub-screen area by switching image processing options at every frame. This allows each image frame to be displayed in the main screen area and sub-screen area 1 with a different combination of image processing options applied [WLI+(LCI), LCI+(WLI), BLI+(WLI), WLI+(BLI)].

The device is AC operated at a power setting of 100-240V/50-60Hz/ 3.0-1.5A. The processor is housed in a steel-polycarbonate case measuring 395x210x515mm

The insertion portion of the device has a mechanism (hereinafter "the bending portion") which bends the tip from right to left and up and down, and a flexible tube (hereinafter "the flexible portion") consists of the bending portion and operating portion with a knob which controls the bending portion. The forceps channel which runs through the operating portion to the tip is arranged inside the insertion portion for inserting the surgical instrument.

The insertion portion of the endoscopes comes into contact with the mucosal membrane.

The tip of the insertion portion is called the "Distal end" which contains the Imaging section, Distal cap, Objective lens, Air/water nozzle, Water jet nozzle, Instrument channel outlet, Objective lens, and Light guide.

The bending portion is controlled by knobs on the control portion/operation section to angulate the distal end to certain angles.

The Flexible portion refers to the long insertion area between the Bending portion and the Control portion (a part of Non-insertion portion). This portion contains light guides (glass fiber bundles), air/water channels, a forceps/suction channel, a CMOS image sensor, and cabling. The glass fiber bundles allow light to travel through the endoscope to illuminate the body cavity, thereby providing enough light to the CMOS image sensor to capture an image and display the image on a monitor. The forceps channel is used to introduce biopsy forceps and other endoscopic accessories, as well as providing suction.

The control portion/operating section provides a grip to grasp the endoscopes and contains mechanical parts to operate the endoscopes. This section includes a Forceps inlet, which allows endoscope accessories to be introduced. The Scope connector connects the endoscopes to the light source and video processor, respectively.

The provided FDA 510(k) clearance letter and summary for FUJIFILM Processor EP-8000 and FUJIFILM Endoscope Model EG-860R focus on establishing substantial equivalence to predicate devices, primarily through engineering performance testing rather than clinical study data involving human readers or AI algorithms. The document explicitly states that the various imaging modes (BLI, LCI, ACI, FICE) are "adjunctive tools" and "not intended to replace histopathological sampling as a means of diagnosis." This indicates that the device operates as an image enhancement and visualization tool, not a diagnostic AI that makes independent claims.

Therefore, the study described in the document is a non-clinical engineering performance evaluation comparing the new device's image quality and functional parameters to those of existing predicate devices. It is not a clinical study involving an AI algorithm and human readers.

Here's an attempt to answer the questions based on the provided text, recognizing that many details typically requested for AI/human reader studies are not applicable or not provided in this type of 510(k) submission:

Acceptance Criteria and Device Performance

The document does not explicitly present a table of acceptance criteria with corresponding performance metrics in a pass/fail format typical of standalone AI performance studies. Instead, it states that "the devices met the pre-defined acceptance criteria for the test" for the EG-860R, and for the EP-8000, "EP-8000 demonstrated substantial equivalence to VP-7000 and BL-7000 in Image performance and color reproduction." The acceptance criteria were "engineering requirements listed in this section" and "identical to those assessed for the predicate devices."

The "performance (of) Image and the performance of the WLI, FICE, BLI, BLI-bright, LCI and ACI imaging modes" was evaluated for the EP-8000. For the EG-860R, a range of performance characteristics was evaluated.

Table of Performance Evaluation (Based on provided text, not explicit acceptance criteria):

Study Details:

1. Sample size used for the test set and the data provenance:
   This section describes engineering performance testing, not a clinical test set with patient data. The "test set" would refer to the physical devices and various test setups (e.g., optical phantoms, standardized targets) used to evaluate the specified engineering parameters. The document does not specify a "sample size" in terms of number of patient cases or images, as it is evaluating hardware and its image generation capabilities directly through engineering tests.

   • Provenance: Not applicable in the context of patient data. The tests were "conducted in combination with representative FUJIFILM gastroscopes."

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

   • Not applicable. Ground truth in this context is established by the engineering specifications and calibrated measurement equipment, not clinical expert consensus. The device produces images; it does not make a diagnosis that would require expert ground truth.

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

   • Not applicable. This relates to clinical interpretation and consensus, which is not part of this engineering performance evaluation.

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

   • No. An MRMC study was not performed. The submission describes engineering performance comparisons to predicate devices, not an evaluation of human reader performance with or without AI assistance. The new imaging modes (BLI, LCI, ACI, FICE) are explicitly stated as "adjunctive tools...not intended to replace histopathological sampling as a means of diagnosis."

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

   • No. This is not an AI algorithm making independent diagnostic claims. The performance evaluated is that of the hardware (processor and endoscope) and its image enhancement capabilities.

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

   • Engineering specifications and measurements. The "ground truth" for the performance tests (e.g., resolution, color reproduction, geometric distortion) would be derived from precisely known physical targets, measurement instruments, and established engineering standards. It is not clinical ground truth (e.g., pathology, clinical outcomes, or expert consensus) because the device's function is image generation and enhancement, not diagnostic interpretation.

7. The sample size for the training set:

   • Not applicable. This device is an endoscope and processor system, not a machine learning model that requires a training set in the conventional sense.

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

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

Ask a specific question about this device

This product is intended to be used in combination with an endoscope to assist endoscopic insertion into the body. Never use this product for any other purpose. This product is intended for use in medical facilities by medical professionals who are properly trained in using it as well as in endoscopic procedures and endoscopic treatments.

Not Found

The provided text is an FDA 510(k) clearance letter and its associated summary for a medical device called "Over-tube (TR-1108A)". This document describes the device, its intended use, and its substantial equivalence to a predicate device.

Crucially, this document is for an "Over-tube" which is a physical accessory used with an endoscope, not an AI/software-based medical device. Therefore, the concepts of acceptance criteria, study design (like MRMC, human-in-the-loop, standalone performance), data provenance, expert ground truth establishment for AI models, and training/test sets as commonly understood for AI/ML device clearances do not apply to this specific product.

The document discusses performance testing of the physical device to demonstrate that changes in size (outer diameter, maximum outer diameter) compared to the predicate device do not affect its safety or efficacy.

Here's how the information provided relates to the closest equivalent concepts for a physical device:

Acceptance Criteria and Reported Device Performance

The document states that "the subject device met performance specifications in the following additional testing." This implies that the 'acceptance criteria' were met if the device passed these tests. The reported performance is simply that it "passed all test objectives."

Table of "Acceptance Criteria" (Implied) and "Reported Device Performance" for an Over-tube (TR-1108A):

Note: For a physical device, these are typically engineering specifications and not statistical performance metrics like sensitivity, specificity, AUC as seen in AI/ML devices.

Study Details (as applicable to a physical device)

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

   • Sample Size: Not explicitly stated as a 'sample size' of cases/patients in the context of a typical AI study. For a physical device, this refers to the number of units tested or the number of movements/stresses performed during testing. This information is not detailed in the provided summary.
   • Data Provenance: Not applicable in the AI/ML sense. The testing is likely done in a lab or benchtop setting by the manufacturer, not with patient data from specific countries or in retrospective/prospective studies.

2. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications of those Experts:

   • Not Applicable. Ground truth, in the context of an AI/ML device, refers to a definitive diagnosis or annotation of medical images/data. For a physical device like an over-tube, performance is assessed against engineering specifications and functional requirements, not expert-labeled "ground truth."

3. Adjudication Method for the Test Set:

   • Not Applicable. Adjudication is typically for expert disagreements on ground truth labels in AI studies. For a physical device, test results are typically objective measurements against specifications.

4. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done:

   • No. MRMC studies are specific to evaluating how AI impacts human reader performance, primarily in image interpretation. This is a physical device, so such a study would not be performed.

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

   • Not Applicable. This is a physical device, not an algorithm.

6. The Type of Ground Truth Used:

   • Engineering Specifications / Functional Requirements. The "ground truth" for a physical device's performance is whether it meets its designed specifications (e.g., minimum flexibility, maximum diameter, insertion force within limits). This is determined by direct measurement and testing, not by expert consensus or pathology on patient data.

7. The Sample Size for the Training Set:

   • Not Applicable. "Training set" refers to data used to train an AI model. This is a physical device.

8. How the Ground Truth for the Training Set was Established:

   • Not Applicable. As there's no training set for a physical device.

In summary, the provided document is for a conventional hardware medical device, not an AI/ML software device. Therefore, many of the questions related to AI/ML study design and ground truth establishment are not applicable. The manufacturer demonstrated "substantial equivalence" by showing that the physical changes (size) did not negatively impact the device's ability to meet its performance specifications.

Ask a specific question about this device

This hood is intended to be used in combination with compatible endoscopes to maintain the field of view during endoscopic procedures such as mucosal resection.

This hood is intended to be used in compatible endoscopes to maintain the field of view during endoscopic procedures such as mucosal resection.

Principles of Operation: Align the objective lens of endoscope with the drain of the hood and attach the hood to the distal end of endoscope by pressing the hood until it stops. Securely attach the endoscope with sterile medical tape. Insert the endoscope equipped with the hood through the mouth. Perform the intended treatment. After completion of the examination, slowly withdraw the endoscope along with the attached hood. Peel the tape off the endoscope completely and remove the hood from the endoscope. Dispose of the hood and tape according to local laws and regulations.

The presented document is a 510(k) summary for the Hood (DH-083ST) by Fujifilm Corporation. It describes the device's intended use and performance testing conducted to demonstrate substantial equivalence to a predicate device.

Here's a breakdown of the requested information based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

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

The document does not specify the sample size used for the test set or the data provenance (e.g., country of origin, retrospective/prospective). It only states that "performance testing was conducted, and the passed all test objectives."

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

This information is not provided in the document. The testing described focuses on physical dimensions and compatibility with endoscopes, rather than a clinical evaluation requiring expert ground truth for interpretation.

4. Adjudication Method for the Test Set

This information is not provided in the document. Given the nature of the tests (dimensional measurements and compatibility), an adjudication method as typically used in clinical studies with human interpretation is unlikely and not described.

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

This information is not applicable and therefore not provided. The device is a physical hood for an endoscope, not an AI-powered diagnostic tool.

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

This information is not applicable and therefore not provided. The device is a physical component (an endoscope hood), not an algorithm.

7. The Type of Ground Truth Used

For the described tests, the "ground truth" would be the engineering specifications and measurements of the device's dimensions and its fit with compatible endoscopes, along with compliance with the ISO 8600-1:2015 standard. There is no mention of expert consensus, pathology, or outcomes data as these types of ground truth are more relevant for diagnostic or therapeutic devices with clinical interpretation.

8. The Sample Size for the Training Set

This information is not provided and not applicable as this is not an AI/machine learning device that requires a training set.

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

This information is not provided and not applicable as this is not an AI/machine learning device.

Ask a specific question about this device

FUJIFILM Endoscope Models EG-840T and EG-840TP is intended for the visualization of the upper digestive tract, specifically for the observation, diagnosis, and endoscopic treatment of the esophaqus, stomach, and duodenum.

FUJIFILM Endoscope Model EG-840N is intended for the visualization of the upper digestive tract, specifically for the observation, diagnosis, and endoscopic treatment of the esophagus, stomach, and duodenum. This product can be inserted orally or nasally. Never use this product for any other purposes.

The insertion portion of the device has a mechanism (hereinafter "the bending portion") which bends the tip from right to left and up and down, and a flexible tube (hereinafter "the flexible portion") consists of the bending portion with a knob which controls the bending portion. The forceps channel which runs through the the tip is arranged inside the insertion portion for inserting the surgical instrument. The insertion of the endoscopes comes into contact with the mucosal membrane.

The tip of the insertion portion is called the "Distal end" which contains the Imaging section, Distal cap, Objective lens, Air/water nozzle, Water jet nozzle (Except EG-840N), Instrument channel outlet, and Light guide. The bending portion is controlled by knobs on the control portion/operation section to angulate the distal end to certain angles.

The Flexible portion refers to the long insertion area between the Control portion (a part of Non-insertion portion). This portion contains light guides), air/water channels, a forceps/suction channel, a CMOS image sensor, and cabling. The class fiber bundles allow light to travel through the endoscope to illuminate the body cavity, thereby providing enough light to the CMOS image sensor to capture an image on a monitor. The forceps channel is used to introduce biopsy forceps and other endoscopic accessories, as well as providing suction.

The control portion/operating section provides a grip to grasp the endoscopes and contains mechanical parts to operate the endoscopes. This section includes a Forceps inlet, which allows endoscope accessories to be introduced. The Scope connector connects the endoscopes to the light source.

This is a 510(k) summary for medical devices (endoscopes), and it refers to "bench testing data" and "performance specifications" being met, rather than clinical study results against acceptance criteria in the manner of an AI/algorithm-driven device.

From the provided text, there is no information about acceptance criteria or a study proving that an AI-driven device meets those criteria for the following reasons:

• Device Type: The devices described (FUJIFILM Endoscope Model EG-840N; FUJIFILM Endoscope Model EG-840T; FUJIFILM Endoscope Model EG-840TP) are physical endoscopes for visualization, diagnosis, and treatment. They are not described as AI or software-as-a-medical-device (SaMD) products designed to process or interpret images/data using algorithms.
• Study Types Mentioned: The document refers to:
  • Electrical safety evaluations (standards: ANSI/AAMI ES 60601-1-2012, IEC 60601-1-6:2020, IEC 60601-2-18:2009)
  • Biocompatibility testing (standards: ISO 10993-5:2009, ISO 10993-10:2010, FDA guidance)
  • Endoscope-specific testing (standards: ISO 8600-3:1997, ISO 8600-4:2014)
  • Software-specific testing (standard: IEC 62304:2015, FDA Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices) - This refers to software within the device, not a standalone AI algorithm for medical image analysis.
  • Cleaning and disinfection validation (FDA guidance)
  • Additional testing for physical parameters like Field of view, Bending capability, Rate of suction, Working length, Diameter of forceps channel, Resolution, LG output, Uneven illumination, Viewing direction, Color reproducibility.

The phrase "Resolution" is listed twice, implying a focus on the optical performance of the endoscope itself.

Therefore, the information required for your request (acceptance criteria, study details for an AI-driven device) cannot be extracted from this document because it pertains to the clearance of a traditional medical device (an endoscope) and not an AI/algorithm-based diagnostic or treatment tool.

If this were an AI device, the document would typically contain sections explicitly detailing:

• The AI algorithm's intended use and function (e.g., detecting polyps, classifying lesions).
• Specific performance metrics (e.g., sensitivity, specificity, AUC) for the AI.
• The clinical study design, comparator (e.g., human experts), and statistical analysis.
• Details about the dataset used (test set, training set, ground truth derivation).

Ask a specific question about this device

This software is a computer-assisted reading tool designed to aid endoscopists in detecting colonic mucosal lesions (such as polyps and adenomas) in real time during standard endoscopy examinations of patients undergoing screening and surveillance endoscopic mucosal evaluations. This software is used with standard White Light Imaging (WLI) and Linked Color Imaging (LCI) endoscopy imaging. This software is not intended to replace clinical decision making.

The subject device represents application of AI technology to endoscopic images to assist in detecting the presence of potential lesions. This development greatly contributes to improving the quality of colonoscopy. In recent years, computer-aided diagnosis (CAD) systems employing AI technologies have been approved and marketed as radiological medical devices for use with computed tomography (CT), X-ray, magnetic resonance imaging (MRI), and mammogram diagnostic images. In endoscopy as well, many images for diagnosis are taken. Since increasing the polyp detection rate is also in demand, CAD systems for endoscopy are being actively developed. Against this background, the company has developed this software (EW10-EC02), a new AI-based CAD system, to support Health Care Provider (HCP) detection of large intestine polyps in colonoscopic images. EW10-EC02 detects suspected large intestine polyps in the endoscope video image in real-time.

Here's a breakdown of the acceptance criteria and study findings for the EW10-EC02 Endoscopy Support Program, based on the provided document:

1. Table of Acceptance Criteria and Reported Device Performance

The document describes two main types of studies: standalone performance testing (evaluating the algorithm only) and clinical testing (evaluating human-in-the-loop performance). The acceptance criteria for the standalone performance are explicitly stated and met, while the clinical study endpoints serve as the criteria for evaluating the device's clinical benefit when assisting human readers.

Standalone Performance Acceptance Criteria & Results:

Clinical Study Endpoints & Results (serving as criteria for human-in-the-loop):

Ask a specific question about this device