Medical AI Cloud | COMARCH SA

Support your doctors and your patients

Information is key in the healthcare sector. Nowadays, the unprecedented amount of medical data is overwhelming and impossible to process without technical help. This is why all Comarch Health products are supported by our Medical AI Cloud - AI cluster leveraging Machine Learning and structured medical knowledge in order to automate medical data analysis, lower the overall costs of healthcare and streamline treatment processes by taking over some of the doctors’ responsibilities.

Features

Comarch Medical Algorithmic Platform (CMAP)

CMAP is a cloud-based platform based on advanced learning algorithms, which supports our telemedicine systems. It has the power to immediately detect and mark any irregularities and evaluate signal quality, even in the vastness of information from 30-day Holter test. These processes significantly shorten analysis time, helping to improve the efficiency of your doctors.

Medical Knowledge Base

A comprehensive database made up of the most current scientific papers and diagnostic procedures; creating a map of connections between symptoms and diseases. An arsenal of relevant phrases and diagnostic steps supporting patients and doctors with individually tailored information and advice available in our systems and apps.

Document Classifier

Our document classifier analyzes and categorizes images, photos, and content, without any manual labor. Applying Optical Character Recognition (OCR) technology, it automatically adds “tags” to scanned images and photos and assigns them to the correct category. It works on the background of our EHR Cloud - secure, interoperable data archive. Document Classifier helps to link and organize information incoming from HIS systems, diagnostic points, telemedicine platforms, patient applications and more. The capabilities don’t stop there, as we can automatically recognize the content of files and extract specific information.

Medical e-Interview

A proper, extensive medical interview can take up to 30-40 minutes, and sometimes even more. Unfortunately, patients and physicians are usually in a hurry and a lot of crucial information is skipped during appointments. Our Medical e-Interview is here to solve the problem.

It is a set of carefully prepared questions that creates a complete health history. A patient fills it up online or through an app before the visit, and can easily share the results with the doctor, who has an insight into the patient's health. AI analyses the responses, automatically sets the appropriate question path and then highlights the most important information to specialists.

Patient Health Metamodel

A holistic overview of your patient’s health condition powered by information from all Comarch Healthcare systems and applications. It offers you a patient’s full medical history (diseases, medications, tests, genetic loads, treatments, etc.). The most crucial data is highlighted and prioritized through intelligent analysis, and then presented comprehensibly. The preview can be adapted for patients, doctors or insurers, displaying elements most relevant to them.

The Science Behind the Solution

The Medical AI Cloud algorithms are adapted depending on the problem that needs to be solved. Their main task right now is the ECG analysis, which involves Digital Signal Processing, Machine Learning, and Statistical Analysis. Combination of these algorithms is set as a pipeline in which several stages are specified, such as heartbeat detector, signal quality assessment, and disorder detection. To ensure maximum detection accuracy different Deep Neural Architectures are used, e.g. Convolutional Nets for local patterns detection, Recurrent Nets for time series dependencies, Autoencoders for learning on non-annotated data.

Another task for Medical AI Cloud is a semantic text analysis, which is used for Document Classifier and Medical Knowledge Base analysis. Based on Decision Trees, after basic text manipulation, the algorithm is able to detect the type of medical document as well as interpret basic information on it.

The algorithms of statistical data analysis are used to support Medical e-Interview and for Patient Health Metamodel analysis. Together with Machine Learning models, they support Medical e-Interview and make medical treatment recommendations.