Vendors
日本語
search
  2. //
  3. Reports & insights
  4. //
  5. Image Recognition in Insurance: Applied Machine Learning

Image Recognition in Insurance: Applied Machine Learning

Create a vendor selection project
Click to express your interest in this report
Indication of coverage against your requirements
A subscription is required to activate this feature. Contact us for more info.
Celent have reviewed this profile and believe it to be accurate.
We are waiting for the vendor to publish their solution profile. Contact us or request the RFX.
Projects allow you to export Registered Vendor details and survey responses for analysis outside of Marsh CND. Please refer to the Marsh CND User Guide for detailed instructions.
Download Registered Vendor Survey responses as PDF
Contact vendor directly with specific questions (ie. pricing, capacity, etc)
23 February 2020
Max Ang

Key research questions

  • What are the success factors leading to image recognition implementation?
  • What are the use cases to consider for insurance?
  • What are the key terminologies in a convolutional neural network model development?

Abstract

In this report, we look at insurance claims and healthcare use cases for image recognition. Due to the synergy between insurance and healthcare, the areas to cover are telerehabilitation, teledoctor, and vehicle damage estimation. This leads to an observation of partnership models and ecosystem viewpoints.

Next, we explore the specific usage of image recognition in insurance and discuss the challenges when developing an AI-based image recognition module. This section intends to highlight application areas and learning points when embarking on an image-based solution. Lastly, we provide an in-depth explanation of how the neural network algorithm is applied, providing a technical understanding of the mechanics behind a convolutional neural network. The brief technical walkthrough aims to deliver a simple understanding of machine learning and give insights into building a model. There will also be an introduction to developing a more robust image-based model. This is explored in adversarial learning, graph convolutional networks and additional data processing techniques, which are specific to image-based problems.

Image recognition is a unique application method and will see strides in developmental innovation to eventual ecosystem integration.

Client Access