Created On
Updated On
Goal: This web application simulates the interaction of a medical doctor client with a Web service API hosted on Google Cloud Run. The API's primary objective is to process the medical images received from the client using a powerful Deep Learning algorithm (Mobile-Net). It then provides a JSON response containing the probabilities of classification for the image across three classes. User Action: As a medical doctor, you can upload an image through the interface, and the API will analyze it to determine the likelihood of belonging to each of the specified classes. Please note that this is a simulated environment, and the results presented here are for demonstrative purposes only. Enjoy exploring the capabilities of our Deep Learning-based medical image classification system! If you want to see the complete project, you can follow this repository ( to the API Ultrasound Classificator.

More details

Use Cases Limitations Evidence Owner's Insight
  • The application serves as an assistive tool for radiologists and obstetricians, providing a preliminary classification of fetal ultrasound images into predefined categories, which may streamline the diagnostic process.
  • This application can be used for educational purposes to demonstrate the potential of AI to process medical images.
  • This application is for demonstrative and educational purposes only and isn’t intended for clinical use.
  • The model's performance is contingent on the quality and diversity of the data it was trained on. While informative, the outputs aren’t always accurate.
  • Currently, only .png and .jpg images are supported.

As of now, data supporting the app’s effectiveness has not been collected. The application is in the early stages of deployment. Future updates will include feedback from initial users and healthcare professionals.

Matteo is the owner of this app. He has a masters degree from Bergamo University in Management & ICT (Information and Communication Technology) Computer Engineering with a background in Biomedical Engineering. He created this app with a vision to enhance the precision of prenatal diagnostics by accurately categorizing fetal ultrasound images into distinct categories (brain, cervix, thorax).

Your insights and feedback on this app are greatly appreciated. If you have any thoughts, experiences, or feedback, please leave a comment below.


Warning: Not intended for clinical use. Assume outputs are unsafe and unvalidated. Use carefully.

  • Favorites: 6
  • Executions: 90

  • Diagnostics & Imaging


M Matteo Ballabio

Member since