Deep Fundus

Deep Fundus contains a series of state-of-the-art Deep Learning models that can predict 12 diseases / indicators through fundus images. These models are all well-trained in the Healgoo AI solution.

Key features

* Fast: upload a fundus photo and have a result within a few seconds.

* Parallel: multiple models independently output their respective predictions through parallel computing.

* Safe: all data is processed in a cloud platform that complies with the HIPAA security specification.

Models

The following pre-training models are currently available on Deep Fundus. Each model can output the prediction result and the corresponding confidence value, indicating the degree of confidence of the algorithm.

* Valid Fundus Image

* Left / Right Eye

* Referable DR

* DR Severity Scale

* Glaucoma

* Cataract

* DME

* AMD

* Myopic Maculopathy

* Stroke / Arteriosclerosis

* Glycosylated Hemoglobin

* Biological Age

Clinical Uses

1. For doctors: your work assistant

  Deep Fundus can be your daily clinic assistant. By analyzing the patient’s fundus photographs, Deep Fundus can provide you with supplementary diagnostic advice for common eye diseases. In addition, it can analyze the risk of systemic indicators such as stroke/arteriosclerosis and glycosylated hemoglobin.

  Deep Fundus is also able to generate and print user friendly reports for your patients which include detailed analysis and referral advice, enabling personalized treatment for your patients.

2. For Screening Organizers: a tool to increase screening efficiency

  Deep Fundus can become a virtual artificial intelligence reading center, providing professional evaluation of fundus photographs collected by multiple screening sites.

  By combining artificial intelligence and manual reading centers in the context of DR screening, a Healgoo Interactive partner has successfully reduced manual reading workloads by 70% and achieved an accuracy of over 95%. Additionally, the average patient waiting time was reduced to 3 minutes. This case fully demonstrates the advantages of using artificial intelligence in a real-world scenario.

3. For Fundus Camera Manufacturers: a built-in value-added module

  Deep Fundus can be designed as a built-in module for your product and activated with a single click. The module automatically obtains patient information and fundus images, uses artificial intelligence models for analysis, and provides diagnostic decisions.

  Artificial intelligence can now be easily integrated into your fundus camera products to enhance user experience and increase added value.

Click to visit official website