Search Results

The Single Use Aspiration Needle NA-201SX-4021 is intended to be used for ultrasound guided fine needle aspiration (FNA) of submucosal and extramural lesions of the tracheobronchial tree in combination with an ultrasound endoscope.

The Single Use Aspiration Needle NA-201SX-4022 is intended to be used for ultrasound guided fine needle aspiration (FNA) of submucosal and extramural lesions of the tracheobronchial tree in combination with an ultrasound endoscope.

The Olympus Single Use Aspiration Needle NA-201SX-4021 and NA-201SX-4022 are designed for use with ultrasound endoscopes for ultrasound-guided fine needle aspiration (FNA) of submucosal and extramural lesions of the tracheobronchial tree. These devices are single-use, disposable, sterile (Ethylene Oxide) needles available in two models with different specifications:

• NA-201SX-4021:

• Working Length: 700mm
• Maximum Insertion Portion Diameter: 1.9mm
• Needle Width: 21G
• Normal Needle Length: 20mm
• Maximum Needle Length: 40mm

• NA-201SX-4022:

• Working Length: 700mm
• Maximum Insertion Portion Diameter: 1.8mm
• Needle Width: 22G
• Normal Needle Length: 20mm
• Maximum Needle Length: 40mm

Both models feature an adjustable sheath length and are intended for single use.

This document is a 510(k) clearance letter for a physical medical device (Single Use Aspiration Needle), not an AI/Software-as-a-Medical-Device (SaMD). Therefore, the requested information pertaining to AI/SaMD performance evaluation criteria (such as test set size, data provenance, expert ground truth, MRMC studies, standalone performance, training set details, etc.) is not applicable and is not present in the provided document.

The document discusses the substantial equivalence of the new aspiration needles to a predicate device based on their physical and functional characteristics. The performance data section refers to bench testing of physical attributes, not computational performance or diagnostic accuracy.

Therefore, I cannot provide the requested information as it relates to AI/SaMD. I can, however, extract the acceptance criteria and reported device performance for the physical device based on the provided text.

Acceptance Criteria and Reported Device Performance (Physical Device)

This section summarizes the performance data for the Single Use Aspiration Needle (NA-201SX-4021 and NA-201SX-4022), based on the provided FDA 510(k) summary. These criteria relate to the physical and functional performance of the device, not an AI algorithm.

1. Table of Acceptance Criteria and Reported Device Performance

Regarding the AI/SaMD specific questions:

As stated previously, the provided document is for a physical medical device (Single Use Aspiration Needle), not an AI/SaMD device. Therefore, the following information is not applicable and not provided in the source text:

2. Sample size used for the test set and the data provenance: Not applicable. Performance testing was physical bench testing, not data-driven AI evaluation.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts: Not applicable. Ground truth for a physical device's performance is typically established by engineering specifications and direct measurement, not expert review of AI outputs.
4. Adjudication method (e.g., 2+1, 3+1, none) for the test set: Not applicable.
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 applies to AI-assisted diagnostic or interpretive systems.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done: Not applicable. This applies to AI algorithms.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.): Not applicable. The "ground truth" for a physical needle refers to its engineered specifications and expected mechanical/biological performance, verified through bench testing.
8. The sample size for the training set: Not applicable. This applies to machine learning models.
9. How the ground truth for the training set was established: Not applicable. This applies to machine learning models.

Ask a specific question about this device

The AIR/WATER VALVE MAJ-1444 and SUCTION VALVE MAJ-1443 have been designed for use with an Olympus ultrasound endoscope for the gastrointestinal (GI) tract, as follows.

Air/Water Valve (MAJ-1444)

• To feed air to remove any fluids or debris adhering to the objective lens.
• To feed water for lens washing.
• To fill the balloon with sterile water.

Suction Valve (MAJ-1443)

• To remove any fluids, debris, or air from the patient.
• To remove the water from the balloon.

The Suction Valve MAJ-1443 and Air/Water Valve MAJ-1444 have been designed for use with an Olympus ultrasound endoscope for the gastrointestinal (GI) tract. The subject devices are reusable. The subject devices are compatible with Olympus endoscopes GF-UCT180 and GF-UE160-AL5.

The Suction Valve MAJ-1443 is attached to the suction cylinder of a compatible endoscope to remove any fluids, debris, or air from the patient and to remove water from the balloon. Suction Valve MAJ-1443 has no patient-contacting components and is a reusable device.

The Air/Water Valve MAJ-1444 is attached to the air/water cylinder of a compatible endoscope to feed air to remove any fluid or debris adhering to the objective lens, to feed water for lens washing, and to fill the balloon with sterile water. MAJ-1444 has indirect patient-contacting components and is a reusable device.

The subject device has the same technological characteristics and similar design as the applicable predicate device.

The provided text is an FDA 510(k) clearance letter for two medical device accessories: the Suction Valve (MAJ-1443) and the Air/Water Valve (MAJ-1444). These devices are classified as endoscopic accessories.

Based on the information provided, the study conducted to prove the device meets acceptance criteria focused on bench testing to demonstrate the functionality of reusable accessories, rather than a clinical study involving human patients, AI integration, or multi-reader multi-case (MRMC) studies.

Here's an analysis of the provided text in relation to your request:

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

The document states: "The following performance bench tests were conducted. All test samples passed pre-defined acceptance criteria." However, it does not provide the specific numerical acceptance criteria or the quantitative reported performance for each test. It only lists the types of tests performed.

Therefore, while the document confirms that acceptance criteria were predefined and met, the specific details of these criteria and the quantitative performance results are not included in this FDA clearance letter. This information would typically be found in the full 510(k) submission.

Regarding the other points of your request, much of the requested information (especially concerning AI, human readers, ground truth establishment for AI models, and large-scale clinical studies) is not applicable to this particular device and its clearance.

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

• Sample Size for Test Set: The document simply states "Test samples were final, finished devices." It does not specify the number of samples tested for each performance bench test.
• Data Provenance: The tests were "bench testing" performed by Olympus. The data provenance is laboratory/bench data generated internally for regulatory submission. It is not patient data, and therefore, concepts like "country of origin of the data" or "retrospective/prospective" study design in a clinical sense do not apply.

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

• Not Applicable. For a device like an endoscope valve, the "ground truth" for its performance is typically established through engineering specifications, material science, and functional testing as defined by recognized standards and internal quality control. This does not involve expert readers establishing ground truth on medical images or clinical outcomes.

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

• Not Applicable. Adjudication methods are relevant for studies where human interpretation of data (e.g., medical images) is involved and consensus is needed. For bench testing of mechanical or functional properties, the "adjudication" is based on objective measurements against engineering specifications.

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. This was not an AI-enabled device. The document describes a traditional medical device accessory (valves for endoscopes). MRMC studies are specific to evaluating diagnostic performance, often of AI algorithms assisting human readers.

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

• No. This device is not an algorithm. It's a physical accessory; therefore, standalone algorithm performance is not applicable.

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

• For these devices, the "ground truth" is defined by engineering specifications and functional requirements (e.g., specific flow rates, suction pressures, resistance to leakage, durability under repeated use, compatibility with specified endoscopes). Successful performance means meeting these pre-defined technical criteria. There is no biological "ground truth" like pathology for these mechanical accessories.

8. The sample size for the training set:

• Not Applicable. This is not an AI/machine learning device; hence, there is no "training set."

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

• Not Applicable. Since there is no training set, no ground truth was established for it.

In summary: The provided FDA 510(k) clearance letter pertains to reusable endoscope valves. The clearance was based on bench testing demonstrating the functional performance and durability of the devices. It is a traditional medical device clearance and does not involve AI, clinical efficacy studies with human participants, or complex ground truth establishment methods typical for AI or diagnostic imaging devices. The specific quantitative acceptance criteria and detailed performance results from these bench tests are not present in this summary document but would be part of the full 510(k) submission.

Ask a specific question about this device

The ULTRASONIC PROBE UM-S20-17S has been designed to be used with an endoscopic ultrasound center, a probe driving unit, and an endoscope for intraluminal ultrasonic imaging of the upper airways and tracheobronchial tree.

The ULTRASONIC PROBE UM-S20-20R has been designed to be used with an endoscopic ultrasound center, a probe driving unit, and an endoscope for intraluminal ultrasonic imaging of the upper airways and tracheobronchial tree.

The Ultrasonic Probes have been designed to be used with an Olympus Endoscopic Ultrasound Center, a Probe Driving Unit, other ancillary equipment and an endoscope for intraluminal ultrasonic imaging of the of the upper airways and tracheobronchial tree and surrounding organs.

Ultrasonic Probes UM-S20-17S and UM-S20-20R are designed to be used in conjunction with bronchoscopes. The Probes are inserted into the patient through a channel of the endoscope.

The Ultrasonic Probes consist of an insertion tube and a connector section. The connector section is connected to the Probe Driving Unit and the Probe Driving Unit is connected to the Ultrasound Center.

The Ultrasonic Probe sends and receives electrical signals to and from the Ultrasound Center through the Probe Driving Unit. The Probes use a 20MHz frequency piezoelectric transducer and produce B-mode scan. They produce 360-degree mechanical/radial sonograms.

The transducer is built into the insertion tube at the tip of the Probe. The transducer is rotated by the Probe Driving Unit within the insertion tube.

The transducer converts the electrical signal to the ultrasound wave, sends it to the object, receives the reflected wave from the object and converts it to the electrical signal. The electrical signal is input to the Endoscopic Ultrasound Center and the ultrasound image is generated by the Endoscopic Ultrasound Center.

UM-S20-17S and UM-S20-20R use direct contact method only.

The Subject devices submitted for clearance each include one (1) major component: the Ultrasonic Probe, which is packaged with a Probe Holder (MH-245) and a Water-resistant Cap (MH-244).

The provided FDA 510(k) clearance letter for the Olympus Ultrasonic Probes (UM-S20-17S and UM-S20-20R) details the acceptance criteria and the study that proves the device meets those criteria. However, it's crucial to note that this document pertains to ultrasonic probes (hardware) and does not involve Artificial Intelligence (AI). Therefore, many of the typical elements related to AI/ML device testing (e.g., training set, ground truth experts, MRMC studies, standalone algorithm performance) are not applicable or described in this type of submission.

The "studies" described here are non-clinical performance tests (bench testing), demonstrating the safety and effectiveness of the physical medical device by comparing its characteristics and performance to a legally marketed predicate device.

Acceptance Criteria and Device Performance for Olympus Ultrasonic Probes

The acceptance criteria for these ultrasonic probes are predominantly based on meeting the performance specifications of the predicate and reference devices, and complying with established medical device standards for safety and functionality. The "reported device performance" indicates that the subject devices met all the acceptance criteria through various non-clinical tests.

1. Table of Acceptance Criteria and Reported Device Performance

• Frequency: 20 MHz (matching Reference Device UM-3R)
• Axial Resolution: 2mm or less (matching Reference Device UM-3R)
• Lateral Resolution: 2mm or less (matching Reference Device UM-3R)
• Frame Rate: 6.67rps (matching Reference Device UM-3R)
• Scanning Field of View: 360° (matching Reference Device UM-3R)
• Display Mode: B-mode (matching Predicate and Reference Devices) | All acoustic output parameters passed/met the specified criteria and standards. |
  | Durability | Demonstrated adequate operational lifespan and resistance to wear and tear. Specific quantitative thresholds are not detailed but are implied by "Durability" testing. | Testing passed/met acceptance criteria. |
  | Measurement Accuracy | Verified the accuracy of measurements obtained using the ultrasonic probes. Specific accuracy metrics are not detailed but are implied by "Measurement Accuracy" testing. | Testing passed/met acceptance criteria. |
  | Human Use Factors | Compliance with FDA Guidance Documents: Applying Human Factors and Usability Engineering to Medical Devices (Feb 3, 2016). Ensures safe and effective use by healthcare professionals. | Testing passed/met acceptance criteria. |
  | Biocompatibility | Compliance with ISO 10993 series (ISO 10993-1, -5, -10, -12, -17, -18, -23). Ensures materials are safe for patient contact. | Testing passed/met acceptance criteria. |
  | Reprocessing Validation | Compliance with ISO 17664-1, AAMI TIR12, ANSI AAMI ST98, ANSI AAMI ST58, ISO 11135, ISO 11138-2. Ensures safe and effective cleaning and sterilization through methods like ETO. | Testing passed/met acceptance criteria. |
  | Electrical Safety/EMC | Compliance with IEC 60601-1, IEC ES60601-1, IEC 60601-1-2, IEC 60601-2-18, IEC 60601-2-37, IEC TR 60601-4-2. Ensures electrical safety and electromagnetic compatibility. | Testing passed/met acceptance criteria. |

Study Details (Applicable to Non-AI Hardware Verification)

As this is a hardware device 510(k) submission, the study methodology focuses on non-clinical performance data and bench testing rather than clinical trials with human subjects or AI-specific assessment methods.

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

• Sample Size: The document does not specify a "sample size" in terms of cases or patients, as the testing was non-clinical (bench testing). The "test set" in this context refers to the manufactured probes themselves and the conditions under which they were tested (e.g., phantom models for acoustic output, simulated use for durability, material samples for biocompatibility).
• Data Provenance: The tests were conducted internally by Olympus Medical Systems Corporation and its manufacturing site, SHIRAKAWA OLYMPUS CO., LTD. in Japan. The data is from bench testing and in-house validation, not from real-world patient data (retrospective or prospective). The manufacturing site is in Japan.

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

• This concept is not applicable to this type of hardware 510(k) submission. "Ground truth" for these tests is established by adhering to widely accepted international standards (ISO, IEC, AAMI) and FDA guidance documents for medical device performance, safety, and manufacturing. These standards define the measurable physical and electrical properties that the device must meet. The experts involved would be engineers, quality control specialists, and regulatory affairs personnel responsible for designing, manufacturing, and testing the probes to these standards.

4. Adjudication Method for the Test Set:

• This concept is not applicable. Adjudication typically refers to resolving discrepancies in expert interpretations of medical images or data, which is relevant for AI image analysis or clinical trials. For hardware testing, "adjudication" is replaced by adherence to established test protocols, measurement equipment calibration, and standard-defined pass/fail criteria.

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

• No, an MRMC comparative effectiveness study was not performed. MRMC studies are used to assess the impact of AI on human reader performance, which is not relevant for this non-AI hardware device. The equivalence claim for this device is based on technical specifications and non-clinical performance data compared to a predicate device.

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

• No, a standalone algorithm performance test was not done. This device is a physical diagnostic ultrasonic transducer, not a software algorithm. Its "performance" refers to its ability to generate images when connected to an ultrasound center, which is then interpreted by a human operator.

7. The Type of Ground Truth Used:

• The "ground truth" for this hardware device is defined by engineering specifications, physical measurements, and compliance with national and international medical device standards. For example, acoustic output parameters are measured against specific limits defined in IEC standards, and biocompatibility is confirmed by testing against ISO 10993. There is no "expert consensus" on imaging findings, pathology, or outcomes data used as ground truth for this device's clearance.

8. The Sample Size for the Training Set:

• This concept is not applicable. There is no "training set" as this is not an AI/ML device that requires machine learning.

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

• This concept is not applicable as there is no training set.

In summary, the FDA 510(k) clearance for the Olympus Ultrasonic Probes is based on a comprehensive set of non-clinical, bench-level performance tests demonstrating technical equivalence to a predicate device and compliance with relevant safety and performance standards for diagnostic ultrasonic transducers. It does not involve AI/ML components or associated clinical study designs like MRMC or standalone algorithm performance assessment.

Ask a specific question about this device

The DISPOSABLE CYTOLOGY BRUSH BC-202D/203D Series is intended to be used to collect tissue specimens in the tracheobronchial tree in combination with a bronchoscope.

The DISPOSABLE CYTOLOGY BRUSH BC-202D-1210/2010/3010/5010, BC-203D-2006 have been designed to collect specimens or cells endoscopically for cytologic examination in conjunction with bronchoscopes. The Subject Device (SD) is inserted into the bronchoscope through a biopsy valve to reach the target area, where, by rubbing the brush on the target area, specimens or cells can be collected. The Subject Device is then withdrawn from the bronchoscope channel with the collected samples.

The Subject Device consists of:

• Handle
• Insertion portion

The handle consists of a ring and a grip. The insertion portion consists of a plastic tube, stainless steel wire, nylon brush (length 6mm or 10mm, diameters 1.2 – 5 mm) and stainless steel distal tip. The grip is attached to the tube and the ring is attached to the wire. The user can move the brush at the tip of the wire back and forth by holding the grip and moving the ring back and forth. The distal tip is designed to prevent damage to the inside of the tube and/or bronchial wall.

This document is a 510(k) clearance letter for a medical device, specifically a disposable cytology brush. It is not an AI/ML device. Therefore, the questions regarding acceptance criteria and studies for AI/ML performance (e.g., sample size for test/training sets, expert consensus, MRMC studies, standalone algorithm performance, effect size of human reader improvement with AI) are not applicable to this document.

The document focuses on demonstrating substantial equivalence to a legally marketed predicate device through:

• Biocompatibility testing: Ensuring the materials are safe for human contact.
• Performance testing (Bench/Non-Clinical): Evaluating the physical and mechanical performance of the brush.
• Sterilization and Shelf Life Testing: Confirming the device remains sterile and functional over its stated shelf life.

Here's how the provided information relates to the performance acceptance criteria and proof for this medical device:

Acceptance Criteria and Device Performance for Disposable Cytology Brush

For this medical device, substantial equivalence is proven by meeting established performance specifications through non-clinical (bench) testing, biocompatibility testing, and sterilization/shelf life validation. There are no AI-specific acceptance criteria or studies mentioned as this is a physical medical device, not an AI/ML software device.

1. Table of Acceptance Criteria and Reported Device Performance

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

• Sample Size: The document does not specify exact sample sizes for each bench test. For these types of physical device tests, sample sizes are typically determined by established standards (e.g., ISO, ASTM) and statistical validity for mechanical testing (e.g., n=30, or as per specific test method requirements).
• Data Provenance: The tests are explicitly described as "Non-Clinical (Bench) Performance Testing" and "Biocompatibility Testing." This means the data comes from laboratory experiments and material testing, not human or observational data. There is no mention of country of origin for the data; it would be generated in the manufacturer's or contracted test labs. All tests are inherently "prospective" in the sense that they are conducted specifically to validate the device's performance.

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

• Not applicable. This is a physical medical device. Ground truth for its performance is established by objective engineering measurements, material science principles, and biological safety standards (e.g., ensuring a certain tensile strength, confirming sterility, testing for cytotoxicity per ISO standards). It does not involve human expert interpretation of images or patient data in the way an AI/ML device would.

4. Adjudication Method for the Test Set

• Not applicable. See point 3. Testing results are pass/fail based on predetermined numerical or qualitative specifications from recognized standards.

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

• Not applicable. This is not an AI/ML device used for diagnostic interpretation. Therefore, there is no human reader component to improve or compare against.

6. Standalone (Algorithm Only) Performance

• Not applicable. There is no algorithm or software for standalone performance evaluation. The device is a physical tool.

7. Type of Ground Truth Used

• Bench Test Specifications: The "ground truth" for this device's performance lies in its ability to meet predefined engineering specifications (e.g., force measurements, cyclical durability, material integrity) and biological safety standards (e.g., absence of toxicity, successful sterilization). This is determined through standardized laboratory testing methods rather than expert consensus on clinical cases or pathology.

8. Sample Size for the Training Set

• Not applicable. There is no "training set" as this is not an AI/ML algorithm that learns from data.

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

• Not applicable. See point 8.

Ask a specific question about this device

The DISPOSABLE BALLOON CATHETER B5-2C has been designed to be used for injection into bronchi, saline injection associated with bronchoalveolar lavage, and bronchial hemostasis (tamponade).

The DISPOSABLE BALLOON CATHETER B5-2C has been designed to be used for injection into bronchi, saline injection associated with bronchoalveolar lavage, and bronchial hemostasis (tamponade). The Balloon Catheter is inserted into the instrument channel of compatible bronchoscopes and advanced to the target area, where medical fluid or saline solution are injected, or the balloon is inflated to allow for hemostasis (tamponade). When the procedure is completed, the balloon is deflated, and the catheter is removed from the patient.

The Balloon Catheter is provided sterile and is constructed of a natural rubber latex inflatable balloon, polyethylene tube, branch, irrigation port, air feed cap, and stopcock. The inflatable balloon has a maximum diameter of O 11 mm after inflation.

The provided document is a 510(k) clearance letter and summary for a medical device (Disposable Balloon Catheter B5-2C). It does not contain any information about acceptance criteria and associated study results for an AI/ML powered device.

The document describes the device, its intended use, comparison to a predicate device, and various performance testing. However, these tests are standard for a physical medical device (biocompatibility, bench testing for physical properties like balloon diameter, strength, etc.) and do not involve AI/ML performance metrics such as sensitivity, specificity, or the involvement of expert readers for ground truth establishment.

Therefore, I cannot provide accurate answers to the questions based on the provided text, as the information regarding acceptance criteria and studies for an AI-powered device is not present.

Ask a specific question about this device

The DISPOSABLE BIOPSY FORCEPS FB-211D/221D/231D/241D are intended to be used to collect tissue within the tracheobronchial tree in combination with a bronchoscope

The Disposable Biopsy Forceps FB-211D, FB-221D, FB-231D, and FB-241D (also known as "EndoJaw") have been designed to collect tissue within the tracheobronchial tree in combination with a flexible bronchoscope. The Disposable Biopsy Forceps are inserted into the channel of a bronchoscope to collect tissue by biting the surface of the lesion with a pair of forceps located at the distal end of the device. The Disposable Biopsy Forceps are withdrawn from the channel and the tissue sample is collected.

The Disposable Biopsy Forceps consists of a handle and an insertion portion. The handle consists of a handle body and slider. The insertion portion consists of a sheath, metallic coil (operating wire), cups, and forceps cap. The slider is connected to the operating wire in the insertion portion and the cups are opened/closed by advancing/retreating the slider with the function of the connection parts. The subject device is available in multiple distal end cup shapes; FB-211D – Alligator jaw-step, FB-221D – Alligator jaw-step with needle, FB-231D – Oval type, and FB-241D – Oval type with needle.

This response is based solely on the provided FDA 510(k) Clearance Letter for the Olympus Disposable Biopsy Forceps.

Analysis of the Provided Document:

The provided document is an FDA 510(k) Clearance Letter for a medical device: Disposable Biopsy Forceps. This type of clearance is for devices that are "substantially equivalent" to legally marketed predicate devices. The key takeaway from this document is that the clearance is based on comparing the new device to a predicate device through non-clinical bench testing, not on clinical performance studies involving human subjects or AI-assisted diagnostic accuracy.

Therefore, the questions regarding ground truth, expert adjudication, MRMC studies, standalone algorithm performance, and sample sizes for training/test sets related to diagnostic accuracy (which would be typical for an AI/software as a medical device) are not applicable to this specific 510(k) clearance documentation.

The "acceptance criteria" and "study that proves the device meets the acceptance criteria" in this context refer to the engineering and performance specifications tested during the non-clinical bench testing to demonstrate substantial equivalence to the predicate device.

Responses based on the provided FDA 510(k) Clearance Letter:

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

The document states: "All test samples passed pre-defined acceptance criteria." However, the specific quantitative acceptance criteria for each test and the detailed reported performance metrics are not explicitly provided in this summary. The document only lists the types of performance tests conducted.

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

• Sample Size: The document mentions "All test samples" but does not specify the numerical sample sizes used for each of the non-clinical bench tests.
• Data Provenance: The tests were conducted by Olympus. The document does not specify country of origin for data as this is a device clearance based on manufacturing and performance testing, not patient data from clinical studies. The testing was prospective in the sense that the tests were designed and executed to evaluate the new device.

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

This question is not applicable as this 510(k) clearance is for a physical medical device (biopsy forceps) and is based on engineering performance demonstration through non-clinical bench testing, not on clinical diagnostic accuracy or interpretation of medical images/data (which would require expert ground truth).

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

This question is not applicable for the reasons stated in point 3. Adjudication methods are relevant for subjective image interpretation or clinical outcomes, not for objective bench test measurements of a physical device.

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

This question is not applicable as the clearance is for a physical biopsy forceps and does not involve AI or human readers for diagnostic interpretation.

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

This question is not applicable as the clearance is for a physical biopsy forceps and does not involve an algorithm.

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

This question is not applicable as the device is cleared based on non-clinical performance characteristics (e.g., insertion, withdrawal, force of jaws, sterility, shelf-life) which are evaluated against predefined engineering specifications and physical measurements, not against clinical ground truth like pathology or outcomes data.

8. The sample size for the training set

This question is not applicable as this is a physical medical device clearance, not an AI/machine learning device that requires a "training set."

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

This question is not applicable for the reasons stated in point 8.

Ask a specific question about this device

The COLONOVIDEOSCOPE OLYMPUS CF-EZ1500DL & CF-EZ1500DI are intended to be used with a video system center, endoscope position detecting unit, documentation equipment, monitor, endoscopic therapy accessories (such as a biopsy forceps), and other ancillary equipment for endoscopy and endoscopic surgery.

The COLONOVIDEOSCOPE OLYMPUS CF-EZ1500DL/I are indicated for use within the lower digestive tract (including the anus, rectum, sigmoid colon, colon, and ileocecal valve).

The GASTROINTESTINAL VIDEOSCOPE OLYMPUS GIF-EZ1500 is intended to be used with a video system center, documentation equipment, monitor, endoscopic therapy accessories (such as a biopsy forceps), and other ancillary equipment for endoscopy and endoscopic surgery.

The GASTROINTESTINAL VIDEOSCOPE OLYMPUS GIF-EZ1500 is indicated for use within the upper digestive tract (including the esophagus, stomach, and duodenum).

The endoscope consists of three parts: the control section, the insertion section, and the connector section.

1. Control section: The UP/DOWN angulation control knob and the RIGHT/LEFT angulation control knob on the control section are connected to the tip of the bending section by a series of wires. By operating the angulation control knobs, the bending section at the distal end bends vertically or parallel to guide the distal end for insertion and observation. The observation mode can be selected by focus switching function, "near focus mode" featuring innovative resolving power for close observation or "normal focus mode" for normal observation. To realize the dual focus mechanism, Voice Coil Motor (VCM) is incorporated as an actuator. The endoscope contains a cylinder to attach a suction valve for suction and air/water valve. Depressing the suction valve will allow the physician to use the endoscope to suction any fluids which are obscuring a good view of the tissue. Therapeutic instruments can be passed through the instrument channel for performing endoscopic biopsy and other therapies. Depressing the air/water valve will allow the doctor to feed water through the endoscope for lens washing. It also can be operated to feed air for removing any fluids or debris adhering to the objective lens.

2. Insertion section: The insertion section has main parts including the image guide, light guides that bring light from the video system center through the endoscope, and instrument channel where therapeutic tools can be pushed in and out (also the suction channel).

3. Connector section: The connector section connects the endoscope with the video system center (CV-1500) through the universal cord.

It appears crucial information regarding the AI or software components' acceptance criteria and the specific study proving the device meets these criteria is missing from the provided FDA 510(k) clearance letter for the Olympus Endoscopes (K250432).

The document details the substantial equivalence of the new endoscopes (CF-EZ1500DL/DI and GIF-EZ1500) to their predicates based on physical characteristics, intended use, and general performance. It mentions extensive non-clinical testing, including:

• Reprocessing validation
• Biocompatibility testing
• Software verification and validation (mentioning FDA guidance for software in medical devices, but not detailing specific performance criteria or tests related to AI/image analysis capabilities)
• Electrical safety and EMC
• Bench testing (performance characteristics like DOF/FOV, Resolution, Noise, etc.)
• Animal study to confirm WLI, NBI, RDI, and TXI performance.
• Risk management

However, the prompt asks for specifics related to an AI/Software component's acceptance criteria and its performance study. The provided text focuses on the endoscope hardware itself and its basic image capture and display functionalities. There is no mention of any AI-driven image analysis, diagnostic augmentation, or automated detection capabilities that would require the kind of clinical performance study details (e.g., sample size, ground truth, expert adjudication, MRMC studies) outlined in the prompt.

The "Indications for Use" for these endoscopes clearly state their purpose is for "endoscopy and endoscopic surgery" for visualization. They are tools for observation and facilitating therapeutic accessories. They are not described as having an AI component for automated analysis or diagnosis.

Therefore, since the provided document does not indicate the presence of an AI or advanced software feature requiring a separate performance study with associated acceptance criteria, I cannot fulfill most of the requested points.

Here's what can be inferred or stated based on the provided document and the information that is explicitly NOT present:

Based on the provided document, there is no evidence of an AI-driven or advanced image analysis software component in the Olympus Endoscopes (CF-EZ1500DL/DI and GIF-EZ1500) that would necessitate the type of acceptance criteria and performance study described in the prompt.

The 510(k) clearance focuses on demonstrating substantial equivalence to predicate endoscopes by comparing physical and operational characteristics, and through general performance testing of the device hardware and its fundamental image capture capabilities.

However, if we were to hypothetically assume there was an AI/Software component (e.g., for polyp detection), and based on common FDA expectations for such devices, here's how the answer would look, highlighting what's missing:

Acceptance Criteria and Study Proving Device Performance (Hypothetical AI Component)

Note: The provided FDA 510(k) clearance letter for K250432 does not detail any specific AI or advanced image analysis software functionality, nor does it contain the performance study metrics for such a component. The following table and points are constructed under the assumption of such a component existing, to illustrate what an ideal response would contain if the information were available in the document.

1. Table of Acceptance Criteria and Reported Device Performance (Hypothetical AI Component)

Since no AI/software component is described in the provided document, this table cannot be filled with actual data. Below is a hypothetical example of what such a table would contain if an AI component for, e.g., polyp detection, were present.

2. Sample Size and Data Provenance for Test Set

• Sample Size for Test Set: Not specified (as no AI performance study is detailed). Typical AI/CADe clearances involve hundreds to thousands of cases.
• Data Provenance: Not specified. For AI devices, this would typically include details like:
  • Country of Origin (e.g., multi-center, US, Europe, Asia)
  • Retrospective or Prospective collection
  • Diversity of patient demographics, disease prevalence, equipment used, etc.

3. Number of Experts and Qualifications for Ground Truth

• Number of Experts: Not specified. For AI/CADe devices, often 3 or more independent experts are used.
• Qualifications of Experts: Not specified. Typically, this would involve board-certified gastroenterologists or colorectal surgeons with significant experience (e.g., 5-10+ years) in endoscopic procedures, potentially sub-specialized in screening/surveillance or interventional endoscopy.

4. Adjudication Method for Test Set

• Adjudication Method: Not specified. Common methods for AI medical device clearances include:
  • Majority Rule: E.g., 2 out of 3, or 3 out of 5 experts agree.
  • Consensus Reading: Experts review and discuss cases to reach a unanimous decision.
  • 2+1 or 3+1: Initial reads by 2 or 3 experts, with a senior adjudicator resolving discrepancies.
  • Pathology Correlation: Where possible, biopsy/histology results serve as the definitive ground truth.

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

• MRMC Study Conducted?: No mention of an MRMC study. The document states "No clinical study was performed to demonstrate substantial equivalence." This implies that any human-in-the-loop performance evaluation of an AI component was not required or not conducted for this 510(k).
• Effect Size of Human Improvement (if applicable): Not applicable as no MRMC study is mentioned. If performed, this would report metrics like the change in reader sensitivity, specificity, or reading time with AI assistance compared to without.

6. Standalone (Algorithm Only) Performance Study

• Standalone Performance Conducted?: No explicit mention of a standalone algorithm performance study with specific metrics (e.g., AUC, sensitivity, specificity on a defined dataset). The "Software verification and validation" generally refers to software engineering and quality assurance, not necessarily clinical performance of an AI algorithm itself.

7. Type of Ground Truth Used

• Type of Ground Truth: Not specified. If an AI component were present for lesion detection, ground truth would typically be established by:
  • Expert Consensus: As described in point 3.
  • Pathology/Histology: Biopsy results confirming the presence or absence of lesions. This is often considered the gold standard for many gastrointestinal applications.
  • Clinical Outcomes Data: Long-term follow-up to confirm diagnoses, though less common for initial device clearance.

8. Sample Size for Training Set

• Sample Size for Training Set: Not specified. This information is typically not included in FDA clearance letters unless directly relevant to a novel AI claim. Training set sizes for medical AI can range from thousands to hundreds of thousands of images/cases.

9. How Ground Truth for Training Set Was Established

• Ground Truth for Training Set: Not specified. It would generally be established through similar methods as the test set (e.g., expert annotations, pathological confirmation), often with additional data augmentation or curation processes.

Ask a specific question about this device

The Luer-Split MAJ-2092 has been designed to be attached to the instrument channel port of Olympus endoscopes to allow both irrigation and the use of EndoTherapy accessories.

The Luer-Split MAJ-2092 ("Luer-Split") has been designed to be attached to the instrument channel port of endoscopes to allow both irrigation and the use of endoscopy accessories. The Luer-Split is a reusable device after appropriate cleaning and disinfection/ reprocessing as stated in the instruction manual accompanied with device. The Luer-Split is provided as a single model (MAJ-2092) and packed as a single unit. The Luer-Split is mounted on the instrument channel port of endoscope and the sealing accessory (i.e., single use biopsy valve) is attached on the proximal end of the Luer-Split. The side arm/T-Pipe of the Luer-Split, also referred to as the irrigation port, is connected to the irrigation tube and irrigation equipment. EndoTherapy devices are inserted via attached sealing accessory to perform procedures.

It appears you've provided the text of an FDA 510(k) clearance letter for the "Luer-Split MAJ-2092." However, the content of this document does not describe acceptance criteria for a device's performance nor does it detail a study proving the device meets those criteria in the context of an AI/software as a medical device (SaMD) product.

This 510(k) is for an endoscope accessory (a physical device), not a software or AI product. The "Non-Clinical and/or Clinical Tests Summary & Conclusions" section mentions "Performance Testing - Bench" (Flow Rate, Composite Durability), "Human Factors Validation," and "Reprocessing Validation," along with "Biocompatibility Evaluation." It explicitly states "Clinical testing is not applicable."

Therefore, I cannot fulfill your request for:

• A table of acceptance criteria and reported device performance related to AI/software performance.
• Sample size and data provenance for a test set in an AI/software study.
• Number of experts and their qualifications for ground truth establishment for AI/software.
• Adjudication method for an AI/software test set.
• MRMC comparative effectiveness study for human readers with/without AI assistance.
• Standalone AI algorithm performance.
• Type of ground truth for AI/software.
• Sample size and ground truth establishment for an AI/software training set.

The provided document simply isn't about an AI/SaMD product. It's a clearance letter for a mechanical accessory for endoscopes.

Ask a specific question about this device

The VISERA S VIDEO SYSTEM CENTER OLYMPUS OTV-S500, when used with endoscopes, video endoscopes, camera heads, monitors and other ancillary equipment for endoscopic surgery, is intended to receive and process electronic signals transmitted from a video endoscope/camera head and output image signal to monitor.

Also, the VISERA S VIDEO SYSTEM CENTER OLYMPUS OTV-S500, when used with rhinolaryngeal stroboscopes, is intended to observe the intralaryngeal phenomenon of phonation (speech) for endoscopic observation to examine the correct functioning of the vocal apparatus (glottis) and to examine voice disorders.

The OLYMPUS OTV-S500 is a "universal platform" which offers compatibility with various endoscopes for different medical specialties and enables efficient operation to meet the clinical needs of the ENT field; the device has been cleared by FDA for use with OLYMPUS urology/gynecology endoscopes (including CYF-VH, CYF-VHR, URF-V2, URF-V2R, URF-V3. URF-V3R. WA2T400A. WA2T412A. WA2T430A. WA2T43WA. WA2T470A. WA2UR11A, WA2UR12A, WA2UR13A, WA2UR14A, WA2UR21A, WA2UR22A, WA2UR23A, WA2UR31A, WA2UR32A, CYF-5, CYF-5R, URF-P6, URF-P6R, URF-P7, URF-P7R and HYF-XP) under K241371.

The OLYMPUS OTV-S500 consists of electrical circuit boards, electrical units (cooling fan, unit power supply, and control panel), harnesses between circuit boards, and optical components (lens and optical filter). A microprocessor is built into the OLYMPUS OTV-S500 which controls processing of observation images, user interface (front panel switch, indicator LEDs, warning buzzer etc.) and menu. These functions are implemented in the embedded software. Scopes including flexible videoscopes or fiberscopes and rigid videoscopes with camera heads and light source are directly connected to the Subject system. When connected, the endoscope (fiberscopes and rigid videoscope without camera head) acquires and displays images directly to the user or output onto a monitor when using a flexible endoscope or scope and camera head.

The Subject devices submitted for clearance include one (1) major component: the Video System Center (OLYMPUS OTV-S500) with Foor Holder (MAJ-2552), and five (5) optional accessories: an HDMI cable (MAJ-2551); the OTV-S500 Upgrade Pack Strobe (MAJ-2547), which activates the stroboscopy function of the OTV-S500; an Air Microphone (MAJ-2548); a Throat Microphone (MAJ-2549); and the Microphone Extension Cable (MAJ-2550).

The OLYMPUS OTV-S500 is compatible with the following OLYMPUS ENT endoscopes: ENF-VH, ENF-VH2, ENF-V3, ENF-V4, ENF-VT3, ENF-XP, ENF-GP2, WA4KS400, WA4KS430, WA4KS431, WA4KS445, WA4KS446, WA4KS471, WA4KS471, WA96100A and WA96105A.

The provided text is a 510(k) Summary for the OLYMPUS OTV-S500 (VISERA S VIDEO SYSTEM CENTER OLYMPUS OTV-S500).

The 510(k) Summary states that no clinical data were collected to support the performance of the subject device. Therefore, a study proving the device meets acceptance criteria in a clinical setting is not available from this document.

However, the document does describe non-clinical performance data and acceptance criteria based on standard compliance and specified tests.

Here's an analysis of the available information:

1. Table of Acceptance Criteria and Reported Device Performance

The document broadly mentions performance testing, but does not provide specific acceptance criteria in numerical or qualitative form alongside reported device performance for each criterion. Instead, it states that "all testing passed/met the acceptance criteria."

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

This information is not provided in the document. The document refers to "Performance Testing Bench" with the subject device and compatible endoscopes, but does not specify the number of devices or endoscopes tested. Data provenance (country of origin, retrospective/prospective) is also not mentioned for the non-clinical tests.

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

This information is not applicable as the described tests are non-clinical, bench-top performance, electrical safety, and software validation tests, which do not typically involve human expert establishment of ground truth in the same way clinical or diagnostic studies do. The "truth" for these tests comes from objective measurements against defined standards and specifications.

4. Adjudication Method for the Test Set

This information is not applicable for the non-clinical tests described. Adjudication methods like 2+1 or 3+1 are typically used in clinical studies where expert consensus is needed for ground truth.

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 done. The document explicitly states: "No clinical data were collected to support performance of the Subject device." The device itself is a video system center for endoscopes, not an AI-assisted diagnostic tool for human readers.

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

This is not applicable in the context of an "algorithm only" performance. The OLYMPUS OTV-S500 is a hardware device (video system center) with embedded software. The software validation tests mentioned would assess the functionality and safety of the embedded software itself, which is part of the device's standalone operation. However, this is not a standalone "algorithm" in the typical sense of a diagnostic AI.

7. The Type of Ground Truth Used (Expert Consensus, Pathology, Outcomes Data, etc.)

For the non-clinical tests described, the "ground truth" or reference for performance is established by technical standards and specifications. For instance:

• Performance Testing Bench: ISO standards (e.g., ISO 8600-3:2019 for Field of View, ISO 15739:2017 for Image Noise and Dynamic Range), and internal product specifications (e.g., for Brightness, Latency, etc.). The FDA Color Performance Review (CPR) Tool for Endoscopy Devices would provide a standard reference for color analysis.
• Electrical Safety/EMC Testing: Compliance with various IEC/ANSI AAMI standards (e.g., ES60601-1, IEC 60601-1-2).
• Software Validation/Cybersecurity: Compliance with IEC 62304, ISO 14971, and FDA guidance documents.

8. The Sample Size for the Training Set

This information is not applicable. As no clinical data was collected and the device is a hardware system with embedded software, there's no mention of a "training set" in the context of machine learning.

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

This information is not applicable due to the absence of a training set as described above.

Ask a specific question about this device

EVIS EUS ENDOSCOPIC ULTRASOUND CENTER OLYMPUS EU-ME3 is intended to be used with Olympus ultrasound endoscopes and Olympus ultrasound probes to observe real-time ultrasound images and is indicated for use within the gastrointestinal (GI) tract, biliary and pancreatic ducts and surrounding organs, intraluminal ultrasound for airways, tracheobronchial tree, trans-rectal, trans-urethral, and trans-esophageal (non-cardiac).

The EVIS EUS ENDOSCOPIC ULTRASOUND CENTER OLYMPUS EU-ME3 system (including KEYBOARD MAJ-2380), when combined with compatible ultrasound video scopes or ultrasound probes, makes an endoscopic ultrasound imaging system that can acquire and display and record real-time ultrasound images of the target tissue or organs. It supplies the driving voltage to the ultrasound transducers to generate ultrasound waves. Using these transducers, the Subject system can display endoscopic and ultrasound images to provide measurements and calculations of distance, area, circumference, time, blood velocity, strain ratio, strain histogram and shear wave speed of the targeted areas. It also allows the storage and retrieval of images for reviewing and printing. The Subject system enables the user to print and record images to an external recording device. Additionally, the Subject system enables the user to record movies to internal memory. The accessories submitted for clearance with the EU-ME3 that may be purchased separately include three (3) Software Options, which unlock software that is pre-installed on the EU-ME3 (for Contrast Harmonic Echo [MAJ-2381], Elastography [MAJ-2382] and Shear Wave Quantification [MAJ-2383]).

The provided FDA 510(k) summary (K243502 for the OLYMPUS EU-ME3) does not contain the detailed information necessary to answer all the questions regarding acceptance criteria and a study proving the device meets those criteria, particularly concerning AI/software performance in a clinical context.

This document describes a premarket notification for a medical imaging device that primarily focuses on hardware and basic software functionalities of an endoscopic ultrasound system, rather than an AI/ML-driven device that requires performance metrics against a defined ground truth.

Specifically, the document states: "No clinical data were collected to support performance of the Subject device." This explicitly indicates that no clinical study, whether involving human readers, standalone AI, or comparative effectiveness, was performed or presented in this submission to establish performance against acceptance criteria for an AI/ML component.

Therefore, the following points can be directly addressed based on the provided text, while others cannot.

Acceptance Criteria and Device Performance Study Analysis (Based on Provided Document K243502)

The provided document describes the Olympus EU-ME3 (EVIS EUS Endoscopic Ultrasound Center), an ultrasound imaging system. The 510(k) submission primarily focuses on demonstrating substantial equivalence to a predicate device (OLYMPUS EU-ME2 Premier Plus, K203128) based on technological characteristics and non-clinical performance data, not a study of an AI/ML algorithm's clinical performance.

Therefore, it is crucial to note that the document explicitly states: "No clinical data were collected to support performance of the Subject device." This means there is no clinical study described within this submission that proves the device meets specific performance acceptance criteria for an AI-driven functionality.

The "acceptance criteria" discussed in this document are primarily related to non-clinical bench testing, safety, electrical compatibility, and compliance with regulatory standards for an ultrasound imaging system.

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

Based on the "SUMMARY OF NON-CLINICAL PERFORMANCE DATA" section (Page 8), the following non-clinical tests were completed and demonstrated compliance, implying these are the "acceptance criteria" for the non-AI/ML aspects of the device demonstrated in this submission:

Ask a specific question about this device