Moscow

Third_Opinion

AI-screening - services for pathologies identification on medical images (chest x-rays, mammograms, morphology,&nbsp;orthopantomograms, retina pathology). The intelligent system, built as a neural network, is able to determine the type of pathology from the relevant medical image online with an accuracy of 75–95%. To train a neural network, you must first collect, digitize and mark up medical analyzes in the form of images.

AI-monitoring -&nbsp;AI-based clinical observation system for increased patient safety. Using computer vision technology, while securely protecting patient information with HIPAA compliance, the system monitors patient safety and the quality of care provided by medical staff.&nbsp;

medicine

covidtech

artificial_intelligence

neural_networks

computer_vision

Third Opinion

Diagnocat

The service uses neural networks and analyzes tomographic studies, finds teeth with pathologies and creates a diagnostic report for doctors. &nbsp;Diagnocat identifies the main anatomical areas in 3 minutes and diagnoses them according to 122 characteristics (norm, filling, crown, implant, periodontitis).

The service includes the following solutions: Cloud&nbsp; - a platform for storing and managing DICOM patient files,&nbsp; Viewer&nbsp; - for quick familiarization with the general state of the oral cavity,&nbsp; Implant Study&nbsp; - for the planning of dental implantation,&nbsp; Endo Study&nbsp; - for planning endodontic treatment,&nbsp; Radiology Report&nbsp; - for dental diagnosis and documentation.

Over 20 thousand dental computed tomograms were used to train artificial intelligence.

Service for diagnosing dental problems based on neural networks

iTeco

SmartSel is used for automatic processing of information received by the organization. The system includes NLP technologies (Natural language processing), machine and deep learning. SmartSel processes the content of the received message, classifies it and builds the optimal route for information in the organization.

The system interacts with data in text, voice and graphic form. With the help of SmartSel, companies automate the classification and distribution of incoming messages by responsible persons, storage locations or information systems.

Additional opportunities such as tools for creating dialogue systems, allow to automate the maintenance of dialogues, generate responses in individual cases. SmartSel increases the speed and accuracy of processing requests, improving the quality of companies’ interaction with partners and customers, and government organizations - with citizens.

Intelligent system for automatic classification and routing of incoming information

machine_learning

SmartSel

Beorg

Smart Vision provides recognition of documents within 2 minutes with a quality assurance of more than 99%. Types of documents: passports with registration, including handwritten; sets of documents of legal entities and individual entrepreneurs; working documents and personal data of employees; sets for registration of bank cards. Scan formats: image files: pdf, png, tiff, jpg, receiving documents via API.

The service is useful for banks - when registering new clients, insurance companies - when applying for insurance, credit organizations - when issuing loans or plastic cards (bonus, discount, credit), mobile applications services - where documents are required to prove your identity.

Document recognition and image digitization service

Beorg Smart Vision

Neurodata Lab software for emotion recognition doesn’t require wearable sensors or special equipment except for webcam and microphone. The technology detects, recognizes and analyzes emotional and social behavior using neural networks, computer vision and deep learning techniques.

Emotions can be expressed through several channels (modalities): facial expressions and microexpressions of the face, oculomotor activity, voice characteristics, movements and position of the body, as well as expressing through physiology such as increased heartbeat.&nbsp; Emotion recognition technologies are applied in various fields: banking and insurance industries, retail, automobile industry, gaming industry, communications, robotics, neuromarketing, security, digital medicine.

<ul>
	<li>Face Tracker: facial recognition in the frame, as well as its individual elements (positions of eyes, mouth, nose, and other neural network characteristics that are not amenable to common interpretation).</li>
	<li>Diarization to identify human speech in the audio stream, as well as isolating the speaker’s voice, if several other voices are there in the audio stream at the moment of speech.</li>
	<li>Body Tracker: tracking body and arm movements in a video stream. </li>
	<li>Eye Tracker: Eye movement data extraction.</li>
	<li>Heart Rate Tracker: Heart rate using a video camera.</li>
	<li>Respiration Rate Tracker: respiratory rate detection using a video camera.</li>
</ul>

System for recognition and synthesis of emotions, detection and analysis of movements

Neurodata Lab

Gero

The platform is based on the analysis of people's physical activity. The neural network identifies health risks according to data from a fitness bracelet about the physical activity of a person. The neural network links repetitive sequences of movements with medical records and test results.

The algorithm will be useful to medical and insurance organizations. It will enable them to identify people in the risk group remotely. Based on the same information, the biological age of the trial subject has already been successfully determined.

AI-based platform for health status assessment

Brain2 online service allows a person without special skills to train and use artificial intelligence systems through normal texts or databases. BRAIN2 has the advantage of user-friendliness, it is enough to download data or connect the platform to the CRM system, and BRAIN2 will start to process data.

Using AI in business optimizes business processes and reduces costs. Brain2 allows you to read texts of any length and answer questions on it, carry out smart searches on commodity bases, calculate housing costs.

Service for creation of knowledge management models for artificial intelligence systems

chatbots

predictive_analytics

Brain2

MegaFon

LegalApe 2.8 is based on neural networks using data arrays that include judicial practice, business correspondence, scientific works and legal position of lawyers. Chatbot answers questions, provides comments and participates in debate using the information from regulatory legal acts.



Arrays of data about about judicial practice, business correspondence, scientific works and legal position of lawyers were used for neural network training. The bot answers the questions and provides commentary based on the information from regulatory legal acts and various authoritative source.

LegalApe 2.8

Dbrain

Dbrain is a set of services for workflow automation and action recognition system.



Products:


<ul>

	<li>Dbrain — a service based on machine learning for processes automation and document management.</li>

	<li>Handl — a service for data labeling and validation for machine learning. Labelers receive payment for completed tasks in currency convenient for them - dollars, euro or rubles.</li>

</ul>

Pattern recognition, detection, calculation of speed, coordinates and direction of movement of objects (people, animals, cars) by means of radio waves, including behind radio-transparent obstacles (walls, obstacles, rain, snow, fog, dark time of day).

Range: through radio-transparent obstacles up to 9 meters, in open space up to 300 meters, passive detection up to 1000 meters. Over 30% of sensors and sensors, AI-navigation, ADAS, automation and security tools are replaced by software. Scopes of application: Auto industry, Robotics, Automation, IoT / IIoT, Smart home.

Requirements: Software EMIIA, Linux or Android OS, Wi-Fi 2.4 / 5 GHz module.

AI application for recognizing patterns through radio waves

EMIIA.AI

ABBYY

ABBYY FlexiCapture increases opportunities of working with data by applying Artificial Intelligence technologies. They allow to process any type of documents - both structured and unstructured documents (contracts, supplements, charters).
NLP (Natural language processing) identifies entities, roles and facts in the text. NER (Named Entity Recognition) technology allows to extract named entities, for example, full name, organization name or address. It is possible to extract any entities, such as life of a contract. Expanded NLP functionality allows to extract more complex concepts, including facts, roles and commitments, for example, fact of appointment to a post or contract conditions.
Due to machine learning and text processing using natural language, the platform automatically identifies, classifies, extracts, verifies and transfers data to information systems from any text or document used in business processes, such as client requests and transactions.

Tech Solutions