Creating a recommendation system for the optimal hormonal stimulation protocol based on DNA polymorphisms and relations between specific polymorphisms and other data.
Invicta provides specialist medical services in their infertility clinics located in Gdańsk, Bydgoszcz, Gdynia, Słupsk, Wrocław and Warsaw. Services provided by Invicta focus on the treatment of fertility disorders and relate to women’s health issues.
The company effectively combines science with business, taking a leading position in the design and application of innovative infertility treatment solutions. Invicta creates both proprietary therapeutic programmes and implements modern solutions.
Prior to presenting the solution, we would like to present the importance of the problem that we will be solving in this project.
More and more families are experiencing difficulties in conceiving naturally. This problem might affect up to one million couples in Poland. The research conducted with the use of artificial intelligence algorithms allowed for the identification of key variables (including genetic ones) needed to develop a system that predicts the result of hormonal stimulation and allows for the selection of the optimal dose of drugs used during the treatment. This way we support the work of a doctor who can plan treatment and control the risks associated with it in a better way.
As part of the activities, we created machine learning models that allow us to determine the optimal hormonal stimulation for a given patient. Stimulation aims to produce the maximum number of eggs while minimising the risks associated with stimulation. The model defines a personalised therapy and takes into account, among other things, the patient’s clinical data, stimulation history, and genetic testing.
The recommendation includes the type of stimulation (long/short protocol), stimulation with agonist or antagonist, stimulation with progesterone, etc.
We also recommend optimal doses of the drug. Defining the scope of the recommendation is a key part of the project.
Another decisive element of the system are functionalities which allow building trust in the algorithms’ results which means determining the certainty of the prediction and indicating the key elements on the basis of which the algorithm made a decision.
The genetic data was provided by molecular biologists at INVICTA Laboratories. The INVICTA Software team worked on creating the application itself.
The success of the application prompted INVICTA to extend the scope of this research to other clinical decisions. The plan is to create a comprehensive recommendation system for the entire treatment process. MIM Solutions will be co-creating this system.