Medical Imaging

Use Artificial Intelligence to improve the processing and interpretation of medical imaging.
We help our clients to develop artificial intelligence (AI) and machine learning (ML) tools for the analysis and interpretation of biomedical data.
Internal components to be inspected using defect detection

As the volume and accessibility of health data increase, artificial intelligence and machine learning are going to play an important role in diagnosis. It makes it possible to offer the highest quality possible care not just to the few but to everyone. Using AI in medicine and medical imaging, it will make the healthcare system better, faster, cheaper, and accessible to everyone.

Why use AI in medical imaging

Early Detection
Enabling early detection, prediction and diagnosis of diseases
Since it is done with a machine, the scans can be processed immediately. No need to wait for a specialist to review the scans. 
Lower Cost
Use machines that are faster and do not need to take breaks.
Making it possible to create personalized interventions and therapies.
Safe, Robust, and Interpretable
Our models ensure safe and robust models from the beginning
We can design our algorithm to work with anonymized data such that privacy is ensured.
Propper uncertainty estimation
Since Bayesian Inference our models get propper uncertainty estimation right out of the box.

Why us?
Whenever one is using Artificial Intelligence (AI) and Machine learning (ML) in high-stakes situations like medical diagnosis, one always worries about the false negatives i.e. when one tells a sick patient that they are healthy and the less critical case of a false positive, i.e. when one tell a healthy patient they are sick.

To make sure that does not happen one wants as accurate predictions as possible but one also needs accurately estimate the uncertainty in the predictions, such that there is a way for the model to communicate the need for further scans and/or involve a medical professional.

Since we focus on Bayesian Deep Learning, we have the perfect tool to address this problem, our models get automatically provide accurate uncertainty estimates with every prediction. It also ensures robust models from the get-go. To learn more about why standard deep learning is il suited for medical problems, read our blog posts Safe AI in medical imaging and Dangerous flaws in computer vision

What can we do?
Depending on the type of medical imaging various techniques can be used, here is an example of some of the things we can help you with.
Image Classification
Classify the entire image into a specific set of classes. The main drawback is it gives one class for the whole image.
Learn more
Object Detection
Detects objects within an image and draws a box around them. The benefit here is that multiple objects can be detected in the same image.
Learn more
Semantic Segmentation
Predicts classes on a pixel level, ex. background, lung, liver, and heart. The upside is that it provides a clear boundary between two objects of different classes. The downside is that semantic segmentation does not separate objects of the same category that overlap.
Learn More
Instance Segmentation
Detects instances of objects and their boundaries. It is similar to semantic segmentation in that one gets contours at a pixel level, but solves the issue of what is a distinct object.
Learn More

Read More

How to proceed

Alvíss AI
Use our Alvíss AI platfor to build it yourself.
Learn more
Custom Project
Utilize our consulting services to outsource all or some of the work to us.
Learn more