Annotating Laparoscopic Images

Services provided: Data Annotation

Published date: 13.09.2021

Read time: 2 min

Client Profile:

Industry: Healthcare
Location: USA
Size: 50-100

Company Bio

The client operates in the medtech industry developing AI tools that can help medical organizations work more efficiently with computer and robotic-assisted interventions.

Overview

The client was developing an AI solution that would assist doctors perform operations by keeping track of all the necessary equipment needed for surgery. They needed to segment (contour with polygons) medical instruments in photos taken during surgical operations.

Business Challenge

The client was developing a system that could track laparoscopic instruments, since this is often necessary for computer and robotic-assisted interventions. They needed to annotate a large number of images,100,000 to be exact. All of the medical instruments needed to be contoured with polygons. It was very important for the client that we achieved the highest level of accuracy possible, which was challenging since the instruments were covered with other objects and liquids. Also, a lot of times the instruments were only partially visible in certain photographs from different angles, which made it difficult to determine the real dimensions of the instruments.

Solutions Provided by Mindy Support

The first thing we did was schedule and provide additional training for the annotators on the types of instruments and how to define the contours of occluded instruments with the best images. This provided the data annotators with the needed skills and knowledge to achieve the necessary level of accuracy while maintaining the necessary pace of annotations.

As we mentioned earlier, we needed to annotate 100,000 images and we were able to get this done in 6 months with an accuracy level of >99%.

Benefits Delivered to the Client

Thanks to the services provided by Mindy Support, the client was able to keep the costs of the project under control and achieve significant savings on overhead costs. We also helped the client keep the project on schedule by hiring and training the needed number of annotators quickly and providing high-quality annotation services as well.

Results

TABLE OF CONTENTS

    Stay connected with our latest updates by subscribing to our newsletter.

      ✔︎ Well done! You're on the list now

      GET A QUOTE FOR YOUR DATA LABELING PROJECT

        I have read and agree to the Privacy Policy

        We have a minimum threshold for starting any new project, which is 735 productive man-hours a month (equivalent to 5 graphic annotators working on the task monthly).

        TALK TO OUR EXPERTS ABOUT YOUR AI/ML PROJECT

        CONTACT US