childish-page-banner

Case Studies

banner

AI model development and certification for cardiovascular medicine

Overview: Our client is a European MedTech start-up focused on the detection of cardiovascular anomalies. The start-up’s objective is to challenge the market for ECG signal-based detection of heart pathologies, such as, but not limited to, atrial fibrillation, premature beats, subventricular tachycardia, etc, obtained from affordable wearable devices such as the Polar H10. The company required an ML-based model development and preparation for the Class 2A medical device certification, in order to ensure that their system is safe and effective for patients to use at home. 

Team: Our team of experts included a Data Engineer, Data Scientists, ML Engineers and a Project Manager

Approach
Discovery phase

Our Data Scientists collaborated with the client in order to understand their existing ECG signal processing system, as well as their computation-based algorithms. We created an automated testing tool that assessed the performance of the current solution against a range of internal and external medical data sets. Based on this analysis, we provided recommendations to improve the scope of detectable medical conditions and identified areas for improvement. 
We advised the client on the certification process and prepared an action plan for the  ML-related work and documentation needed for the Class 2A certification. We also provided improvement recommendations and execution support, which helped the client enhance their ECG signal processing system and algorithms. Our team completed the analysis and recommendations in less than 8 weeks, providing actionable advice, which helped the client reduce the application processing time. 

ML-based signal processing model development phase

The client provided the raw ECG signal data, including labelled data with annotations of R-peaks, pathologies, as well as ‘soft’ and ‘hard’ noise segments. This labelled data was used as inputs for the models and deemed as their training set. The quality and accuracy of pathology predictions (detection) heavily depend on the cleaness of the ECG signal. This is why it is important to have robust, well-performing models in terms of accuracy and precision, which can distinguish between the clean and noisy parts of the ECG signals. Our Data Scientists developed ML-based models for ‘soft’  and ‘hard’ noise detection. Upon successfully passing the strict model performance acceptance criteria of the client, our ML Engineers created the production pipelines for training and inference and integrated it into the client’s platform and business processes. Our collaboration continues with the development of further pathology detection ML-based algorithms. 

Results: Through our active collaboration with the client's team, we were able to complete the model certification preparation work in just 6 months, helping the client achieve their goal of obtaining a Class 2A certification for their AI-powered medical device system. This certification allows the client to offer their product to a wider range of customers, enabling them to provide early detection and prevention of heart-related medical conditions.

banner

Stock market data tool integrated with Bloomberg

Overview: Our client, a financial services company, approached us with a challenge to improve their data analysis tools and provide automation to their team's daily work. They wanted a solution that would help their surveillance organization gain more insights into potential scenarios. Our approach was to create a solution that would enable them to predict MTD/YTD prices.

Approach: We worked in close collaboration with the client to analyze multiple numerical methods and compose different types of feature vectors to improve the prediction model's accuracy. We developed a script that extracts information and an algorithm that calculates the analysis of historical price changes.
We integrated multiple data sources into the solution, including the Bloomberg service terminal, to ensure we had the most accurate and up-to-date data. We developed an API and a user-friendly interface for filtering and searching the information we extracted and analyzed. We used Keras to predict MTD / YTD prices developing a stock prediction model with a neural network to predict the returns on stocks. 

Tech stack: Python, Django,Vue.js, AWS, Keras. 

Results: The performance of our solution was excellent, and we optimized it to achieve more than 80% accuracy in predicting prices. As a result, the tool became an essential asset for our client, empowering their surveillance organization to understand market trends better and predict future prices with greater accuracy.

banner

Image recognition and NLP for fraud detection

Challenge: Creating a fully functioning solution from scratch

Team: Data Engineers, Data Scientists, ML Engineers, DevOps

Solution: The project started with the investigation phase to analyse the existing data and test multiple solutions which meet the needs of both client and cloud providers. AWS was selected as a core service provider for data storage combined with ElasticSearch. Our team built the full ETL process to start operations. Once the data structure and pipelines were set, we introduced two teams of Data Scientists - one specialized in semantic analysis (NLP) and the other with experience in image recognition. The ML part was further taken care of by our engineers. The output was a dashboard of data visualisations made in Kibana to present the results.

Status: Successful completion of the data engineering phase with ongoing data science stages.

Results: The project significantly improved the client's ability to detect security breaches behaviour, which led to a reduction in risk.

media-card-img
AI model development and certification for cardiovascular medicine
media-card-img
Stock market data tool integrated with Bloomberg
media-card-img
Image recognition and NLP for fraud detection
media-card-img
Traffic analysis and prediction based on AI
media-card-img
IoT tool for electricity consumption analysis
media-card-img
Research Overview: Cardiac Comorbidity and Radiation-Induced Lung Toxicity (RILT)
media-card-img
Point of View: Opportunities for AI adoption in Banking
media-card-img
AI model development and certification for cardiovascular medicine
media-card-img
Video platform and chatbot for historical education museum
media-card-img
A CRM model for effective customer offer optimization in retail banking
media-card-img
Empowering athletes: crafting success with the Sika Strength App
media-card-img
NLP development for a pharmaceutical company
media-card-img
AI screening tool for a recruitment software
media-card-img
Software with computer vision analysis in a manufacturing plant
media-card-img
Image recognition and NLP for fraud detection
media-card-img
Traffic analysis and prediction based on AI
media-card-img
Developing a comprehensive PD model for accurate credit risk assessment in banking
media-card-img
Federated Learning for personalized healthcare prediction models in oncology
media-card-img
Research Overview: Cardiac Comorbidity and Radiation-Induced Lung Toxicity (RILT)
media-card-img
Application of AI in Medicine
media-card-img
Video platform and chatbot for historical education museum
media-card-img
Empowering athletes: crafting success with the Sika Strength App
media-card-img
Web platform with CRM for travel reservations exchange
media-card-img
Software with computer vision analysis in a manufacturing plant
media-card-img
Educational gamification platform with mobile applications
media-card-img
Maintenance management ERP and storage for an international manufacturing plant
media-card-img
Healthcare app giving full communication between doctors and patients
media-card-img
Special education learning management system
media-card-img
A platform generating civic education content for teachers
media-card-img
IoT tool for electricity consumption analysis

Contact us

By clicking the Accept button, you are giving your consent to the use of cookies when accessing this website and utilizing our services. To learn more about how cookies are used and managed, please refer to our

Cookie Statement & Privacy Policy