Skip to main content

Science with a Future

In 2021, America for Bulgaria Foundation launched the second edition of its Science with a Future competition for young Bulgarian researchers, scientists, and engineers aged 16-40. 

 

The competition selects and awards funding to promising proposals in robotics, artificial intelligence, and new technologies. The European Software Institute – Center Eastern Europe participates as an administrative coordinator and technological mentor for the approved applications.

 

The selected projects are of tangible and measurable effect on business and address existing problems of public interest. A distinctive feature of all selected proposals is their high potential for practical implementation. The competition's primary goal is to support scientific projects and products in the early stage of development.

 

Other partners in the initiative are Sofia Tech Parkthe Professional Association of Robotics and AutomationAI Cluster Bulgaria, and the British Council.

 

The selected projects:

  • Thermoself - A Pre-Screening Method for Diseases and Abnormalities in Outpatient Conditions By Images from a Thermal Camera. Thermoself is an intelligent solution to detect and monitor various health issues and inflammatory processes in different stages and conditions using AI analysis applied to digital thermal images captured with the Kelvin mobile app. Examples of diseases include cancer, arthritis, vascular, respiratory infections, migraines, muscle pain, and more. Using Kelvin requires no physical contact and is thus absolutely non-invasive as it works through a portable thermal camera and smartphone. The goal of Thermoself is to compare the capabilities of an innovative self-screening technique (thermography) with an established one (mammography) for women with newly diagnosed breast cancer before patients receive any treatment and to develop an artificial intelligence-based machine learning model to distinguish between infrared images of healthy and diagnosed patients.

  • Decision Support Models for the Prediction of Reached Sugar Content and Risk of Infection in Vineyards. Bivine is a software solution aiming to help winegrowers and wine producers with solutions to common problems. It combines a system for informed decisions and management of winery processes. The latter allows tracking and planning of all activities and resources in the winery, starting from the treatment of the vines, through wine production, until the moment when a glass on the table is filled with the divine drink. The system for informed decision-making takes weather data from sensors placed in vineyards and weather forecasts. Such raw data include air temperature and humidity, soil and leaf moisture, and rainfall. Information from this analysis is used to predict the development of disease, infections, and grape quality.

  • АзБуки.ML - A Platform for Processing Unstructured Speech. In various forms, technologies such as voice assistants, chatbots, intelligent devices, robotic interlocutors, and artificial intelligence are increasingly common in our everyday lives. On the other hand, human-technology interaction is no longer limited to the use of a mouse and a keyboard as it increasingly takes place through sound, gestures, voice commands, and natural language. However, most of these interactions are primarily supported in English and a small group of a few other languages. Although Bulgarian language support is available in translation and sound recognition portals, there is a deficit in the practical implementation of classification models, grammatical analysis, content analysis, etc. The main goal of the project is to address the lack of such tools in Bulgarian, as well as to develop new approaches and additions to existing models, algorithms, and programs and to create an easy-to-use API that can be made available to companies, organizations, and institutions.

  • ReBox 2.0 – A Smart Bin for Recyclable Waste. ReBox is an intelligent container for separate collection. The project's authors aim to build sustainable habits related to waste separation, disposal, and environmental protection in children, young people, and their parents. Using a variety of Internet of Things (IoT) sensors, microcontrollers, and software solutions, the smart containers count objects, measure weight, calculate the recyclable value (based on its pre-defined market price), measure volume, and feature an audible alarm. Thanks to these technologies, it is possible to visualize the amount of waste collected on a display and on a dedicated website where the information is updated every 10 minutes. The website aggregates the data from all active containers. Visualizing the results is an essential element of the project concept because, through a combination of information technology and behavioural science methods, the project aims to increase the motivation for separate waste collection in children.

  • Intelligent Hand Rehabilitation Robot. Currently, there is a lack of sufficiently intelligent rehabilitation robots that are teleoperated and allow for rehabilitation at home. The idea of the project is to create a prototype of an intelligent hand rehabilitation robot. The robot will be based on the principle of telemedicine. It will perform the rehabilitation process autonomously after being given control parameters by a medical person. A remote control application will be developed for remote communication. Furthermore, a separate mechanism will be constructed that will allow the robot to perform complex rehabilitation movements according to the needs of each patient. Based on sensor data and intelligent algorithms, the robot can report the strain and load on the patient's arm during rehabilitation. Performance reports for each procedure will be recorded and generated. The aim is thus to prevent overloading or complications in the health condition and speed up the recovery process.

  • Latency-Aware IoT Fog Platform. Current cloud platforms provide the ability to access a large volume of computing resources but fail to operate fast enough (in real-time) in the face of the vast number of hardware devices. Currently, cloud platforms do not consider the limitations of devices such as low computing volume, networking, and energy resources, rapidly changing location and availability. This problem is addressed by decentralizing the architecture into fog platforms. Currently, there is no optimal resource management algorithm in a real fog platform. This project aims to create an optimal resource management algorithm in a fuzzy platform under a dynamic environment composed of resource-constrained devices. The goal is to achieve minimum delay time (latency) in the communication between the user and the fog platform, which is the most critical parameter for user experience, along with optimizing the resources used. The project tests the algorithm's validity in a realistic environment with a popular platform (Kubernetes) and hardware devices measuring the physical parameters of cars."

  • A Platform for Analyzing Water Quality Based on Data from Custom-Developed Sensor Systems and Machine Learning. The high concentrations of HAP (2.5, 10), carbon dioxide (CO2), sulfur dioxide (SO2), nitrogen dioxide (NO2), and ozone (O3) are among the leading causes of reduced life expectancy for over 500 000 thousand inhabitants in the EU in 2018 (according to the European Environment Agency). This project aims to develop a sensor system providing citizens with real-time data on the acidity/alkalinity status (PH) of water, the level of oxygen (O2), the presence of nitrogen (N), phosphorus (P), and the presence of uranium (U). The platform aims to improve people's quality of life by giving them access to real-time data from the sensor system and offering the ability to analyze and make predictions (based on the collected data) on the quality of data through a neural network/artificial intelligence platform.

  • Autonomous Laser Signalling Boxes for Mountain Rescue. This project aims to equip mountain marker stakes at key locations with an innovative box fitted with a dedicated solar panel, lasers, a small first aid kit, a button to send a message to the rescue services, a battery to store energy and more. The alert button will send a message with the box coordinates and information about the temperature at the box's location. The goal is to empower the alerting process of nearby hikers or rescuers via laser signalling, who can quickly locate and assist the person in need after seeing the laser or receiving an emergency message.

  • Autonomous Mailboxes for Individuals and Businesses. The traditional mailbox is increasingly becoming a container for paper spam. This is primarily due to its public access. This is why mailboxes have begun to lose their actual value. People rely on them less and less when receiving valuable mail. Therefore, mailbox owners often ignore it and try to find alternative ways of receiving mail. The intelligent mailbox project aims to solve this problem. The autonomous mailbox for homes and businesses is a platform that consists of smart IoT mailboxes and management software. The platform aims to provide the ability to monitor and control your mailbox at any time and from any location. The boxes send real-time notifications while remaining simple to use and integrate with external systems. The platform is also heavily focused on reducing environmental impact, offering autonomous mailboxes powered entirely by solar energy in various sizes and shapes.

A decorative image of a robot and a human hand approaching each other.

 

Support for the European Software Institute - Center Eastern Europe is provided by America for Bulgaria Foundation. The statements and opinions expressed herein are solely those of the European Software Institute - Center Eastern Europe and do not necessarily reflect the views of America for Bulgaria Foundation or its affiliates

 

The America for Bulgaria Foundation (ABF) is an independent, nonpartisan, and nonpolitical American grant-making foundation working in partnership with Bulgarians to strengthen the country’s private sector and related democratic institutions.

 

ABF builds on the success of the Bulgarian-American Enterprise Fund, an investment fund created in 1991 by the U.S. Congress with support from the U.S. Agency for International Development. For more information, please visit www.us4bg.org.