TRUE VISION, TRUE VISION II by MEDICAL DYNAMICS, INC.

These intra-oral Dental Cameras are for use in dentistry to be able to show the patient abnormalities and pathology within the mouth. The cameras are utilized exclusively to inform the patient of conditions in the mouth which require treatment. It is not intended that the dental intra-oral camera be utilized in any dental operative procedure.

The camera is provided NON-STERILE and the camera is not built so that it can tolerate any sterilization process.

The camera system does provide a "clean", optically clear covering for the distal end of the handpiece. This provides a "clean" covering for the distal handpiece, and is intended for one time use only.

Device Description

Not Found

AI/ML Overview

The provided text contains two distinct documents. The first two pages (pages 0 and 1) are an FDA 510(k) clearance letter for a device named "True Vision, True Vision II" (K974542), which is an Intra-Oral Dental Camera. The subsequent pages (starting from page 2) appear to describe a generic deep learning-based object detection and classification system, completely unrelated to the dental camera.

Therefore, the input does not contain information about the acceptance criteria and study proving the dental camera meets those criteria. Instead, it includes a generic description of an AI system.

Based on the only relevant information about a device (the True Vision Intra-Oral Dental Camera) in the FDA 510(k) letter, I cannot provide the requested details because the document does not contain: acceptance criteria, study details, data provenance, expert ground truth, MRMC study, standalone performance, training set details, or ground truth establishment relevant to the dental camera.

The second part of the provided text, which describes a deep learning system, is a generic explanation and not related to the FDA cleared device. If you intended this to be a separate AI device, the information is still very high-level and lacks specific details to answer most of your questions.

However, if I were to hypothetically extract information based on the generic deep learning system described on page 2 and onwards, it would be as follows (but please note this is NOT tied to the FDA clearance document or a specific device):

Hypothetical Analysis based on the Generic Deep Learning System Description (Page 2 onwards - NOT the FDA cleared dental camera):

It is important to reiterate that the following information is extracted from a generic description of a deep learning system provided within the input and does not relate to the FDA-cleared "True Vision, True Vision II" Intra-Oral Dental Camera. The provided description is a high-level overview of an AI system's design and conceptual evaluation, not a detailed regulatory study.

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

Acceptance Criteria	Reported Device Performance
Not explicitly stated in the document.	Accuracy: 90% on a dataset not used for training.

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

Test Set Sample Size: "a dataset of images that were not used to train the system." - Specific size not mentioned.
Data Provenance: Not mentioned (e.g., country of origin, retrospective/prospective).

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

Number of Experts: Not mentioned.
Qualifications of Experts: Not mentioned.

4. Adjudication method for the test set

Not mentioned.

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 mentioned. This generic description focuses on standalone algorithm performance, not human-AI collaboration.

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

Yes. The system "achieved an accuracy of 90% on the dataset," implying standalone performance evaluation.

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

Object Detection: "labeled with the bounding boxes of the objects in the images."
Object Classification: "labeled with the class of the object in the image."
This implies human annotation/labeling for bounding boxes and class labels, likely by trained annotators, but specific "expert consensus" or other types are not detailed.

8. The sample size for the training set

Training Set Sample Size: "a large dataset of images." - Specific size not mentioned.

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

Object Detection: "training on a large dataset of images. The object detection model is trained on a dataset of images that are labeled with the bounding boxes of the objects in the images."
Object Classification: "trained on a large dataset of images. The object classification model is trained on a dataset of images that are labeled with the class of the object in the image."
The ground truth was established by labeling bounding boxes and class labels on the training images. The method or individuals responsible for this labeling are not specified beyond "labeled."

Summary

{0}------------------------------------------------

Image /page/0/Picture/1 description: The image shows the logo for the U.S. Department of Health & Human Services. The logo consists of a stylized eagle with three stripes forming its body and wings. The eagle faces to the right. The text "DEPARTMENT OF HEALTH & HUMAN SERVICES • USA" is arranged in a circular fashion around the eagle.

Food and Drug Administration 9200 Corporate Boulevard Rockville MD 20850

FEB 1 7 1998

Edwin L. Adair, M.D. Director Medical Dynamics, Incorporated 99 Inverness Drive, East Englewood, Colorado 80112

K974542 Re : True Vision, True Vision II Trade Name: Regulatory Class: I Product Code: EIA Dated: November 28, 1997 December 3, 1997 Received:

Dear Dr. Adair:

We have reviewed your Section 510(k) notification of intent to market the device referenced above and we have determined the device is substantially equivalent (for the indications for use stated in the enclosure) to devices marketed in interstate commerce prior to May 28, 1976, the enactment date of the Medical Device Amendments, or to devices that have been reclassified in accordance with the provisions of the Federal Food, Drug, and Cosmetic Act (Act). You may, therefore, market the device, subject to the general controls provisions The general controls provisions of the Act of the Act. include requirements for annual registration, listing of devices, good manufacturing practice, labeling, and prohibitions against misbranding and adulteration.

If your device is classified (see above) into either class II (Special Controls) or class III (Premarket Approval), it may be subject to such additional controls. Existing major requlations affecting your device can be found in the Code of Federal Requlations, Title 21, Parts 800 to 895. A substantially equivalent determination assumes compliance with -----the current Good Manufacturing Practice requirement, as set forth in the Quality System Regulation (QS) for Medical Devices: General regulation (21 CFR Part 820) and that, through periodic (QS) inspections, the Food and Drug Administration (FDA) will verify such assumptions. Failure to comply with the GMP regulation may result in regulatory In addition, FDA may publish further announcements action. concerning your device in the Federal Register. Please note: this response to your premarket notification submission does not affect any obligation you might have under sections 531

{1}------------------------------------------------

Page 2 - Dr. Adair

through 542 of the Act for devices under the Electronic Chrough 542 of the Act 101 xevisions, or other Federal laws or regulations.

This letter will allow you to begin marketing your device as Inis recei will arrow } premarket notification. The FDA described in your sitem, privalence of your device to a legally marketed predicate device results in a classification for your marketed predicate acvice pur device to proceed to the market.

If you desire specific advice for your device on our labeling regulation (21 CFR Part 801 and additionally 809.10 for in regulation (21 ork rares), please contact the Office of Compliance at (301) 594-4618. Additionally, for questions on the promotion and advertising of your device, please contact the Office of Compliance at (301) 594-4639. Also, please note the office or compleased, "Misbranding by reference to the regulation entification" (21 CFR 807.97). Other general information on your responsibilities under the Act may be Information on Jourision of Small Manufacturers Assistance at its toll-free number (800) 638-2041 or (301) 443-6597 or at its internet address "http://www.fda.gov/cdrh/dsmamain.html".

Sincerely yours,

Willionsk

V A. Ulatowski Timo Director Division of Dental, Infection Control and General Hospital Devices Office of Device Evaluation Center for Devices and Radiological Health

Enclosure

{2}------------------------------------------------

12/10/97 WED 12:17 FAX 301 460 3002

FDA/ODE/DDIGD

0 3002 FDA/O

1. Introduction

This document describes the design and implementation of a system for detecting and classifying objects in images. The system is based on a deep learning model that is trained on a large dataset of images. The system is able to detect and classify objects in real-time, and it is robust to variations in lighting, pose, and occlusion.

The system is designed to be used in a variety of applications, such as autonomous driving, robotics, and surveillance. The system is also designed to be easily integrated into existing systems.

The system is implemented in Python using the TensorFlow deep learning framework.

2. System Overview

The system consists of three main components:

Object Detection: This component is responsible for detecting objects in images. The object detection component is based on a deep learning model that is trained on a large dataset of images.
Object Classification: This component is responsible for classifying the objects that are detected by the object detection component. The object classification component is based on a deep learning model that is trained on a large dataset of images.
System Integration: This component is responsible for integrating the object detection and object classification components into a single system. The system integration component is implemented in Python using the TensorFlow deep learning framework.

3. Object Detection

The object detection component is based on a deep learning model that is trained on a large dataset of images. The object detection model is a convolutional neural network (CNN) that is trained to predict the bounding boxes of objects in images. The object detection model is trained using a supervised learning approach. The object detection model is trained on a dataset of images that are labeled with the bounding boxes of the objects in the images.

The object detection model is trained using the following steps:

The images in the training dataset are preprocessed to normalize the pixel values.
The preprocessed images are fed into the CNN.
The CNN outputs a set of bounding boxes for each image.
The bounding boxes are compared to the ground truth bounding boxes.
The difference between the predicted bounding boxes and the ground truth bounding boxes is used to update the weights of the CNN.
The process is repeated until the CNN converges.

4. Object Classification

The object classification component is based on a deep learning model that is trained on a large dataset of images. The object classification model is a CNN that is trained to predict the class of an object in an image. The object classification model is trained using a supervised learning approach. The object classification model is trained on a dataset of images that are labeled with the class of the object in the image.

The object classification model is trained using the following steps:

The images in the training dataset are preprocessed to normalize the pixel values.
The preprocessed images are fed into the CNN.
The CNN outputs a set of class probabilities for each image.
The class probabilities are compared to the ground truth class labels.
The difference between the predicted class probabilities and the ground truth class labels is used to update the weights of the CNN.
The process is repeated until the CNN converges.

5. System Integration

The system integration component is responsible for integrating the object detection and object classification components into a single system. The system integration component is implemented in Python using the TensorFlow deep learning framework.

The system integration component is implemented using the following steps:

The object detection component is used to detect objects in an image.
The bounding boxes of the detected objects are passed to the object classification component.
The object classification component is used to classify the objects in the bounding boxes.
The class labels and bounding boxes are displayed on the image.

6. Results

The system was evaluated on a dataset of images that were not used to train the system. The system achieved an accuracy of 90% on the dataset.

7. Conclusion

The system is able to detect and classify objects in images with high accuracy. The system is robust to variations in lighting, pose, and occlusion. The system is designed to be used in a variety of applications, such as autonomous driving, robotics, and surveillance. The system is also designed to be easily integrated into existing systems.

Page of

SIO(k) Number (if known): (Not known at this time)

Intra-Oral Dental Camera Device Name:

Indications For Use:

These intra-oral Dental Cameras are for use in dentistry to be able to show the patient abnormalities and pathology within the mouth. The cameras are utilized exclusively to inform the patient of conditions in the mouth which require It is not intended that the dental intra-oral camera be treatment. utilized in any dental operative procedure.

The camera is provided NON-STERILE and the camera is not built so that it can tolerate any sterilization process.

(PLEASE DO NOT WRITE BELOW THIS LINE - CONTINUE ON ANOTHER PAGE IF NEEDED)

Concurrence of CDRH, Office of Device Evaluation (ODE)

MARKA ........................................................................................................................................................................ institution

(Division Sign-Off) (Division Signeral, Infection Control, and General Hospital D 74547 510(k) Number .

Prescription Use √
(Per 21 CFR 801.109)

Over-The-Counter Use

(Optional Format 1-2-96)

Regulation Number and Section

§ 872.6640 Dental operative unit and accessories.

(a)
Identification. A dental operative unit and accessories is an AC-powered device that is intended to supply power to and serve as a base for other dental devices, such as a dental handpiece, a dental operating light, an air or water syringe unit, and oral cavity evacuator, a suction operative unit, and other dental devices and accessories. The device may be attached to a dental chair.(b)
Classification. Class I (general controls). Except for dental operative unit, accessories are exempt from premarket notification procedures in subpart E of part 807 of this chapter subject to § 872.9.