| Developers: | Smart Engines (Smart Endzhins) |
| Last Release Date: | 2025/03/12 |
| Branches: | Mechanical and Instrument Engineering, Pharmaceuticals, Medicine, Healthcare |
| Technology: | Big Data, Data Mining, CAD, EDMS |
Content |
The main articles are:
Smart Tomo Engine is a domestic software complex that conducts three-dimensional tomographic reconstruction of objects of different nature using a set of X-ray projections in real time. The use of reconstruction algorithms in the Smart Tomo Engine can significantly reduce computing time without increasing computing power.
2025: AI announcement to diagnose turbojet engines in assembly
On March 12, 2025, Smart Engines announced that it had made AI to diagnose turbojet engines in assembly. The researchers proposed a technique using artificial intelligence to perform tomography of the entire jet engine - indiscriminately on parts and in one tomographic measurement on a serial industrial installation. $5 million was invested in the development of AI technology.
According to the company, the specifics of tomographic studies of jet engines are associated with fundamentally different requirements for materials used to create parts of the "hot" and "cold" parts of the engine.
For the "hot" part of the engine, it is important that the parts withstand high temperatures and overloads. The "cold" part - the housing and compressor - is created from lightweight materials to reduce the total weight of the engine as much as possible. At the same time, dense heat-resistant materials can absorb all radiation, and light materials are almost transparent. As a result, some of the real defects are not visible on the engine tomography, and false defects occur in places where there is not enough information. These effects are so powerful that instead of an entire shoulder blade, you can see the void.
The researchers proposed a non-destructive testing technique combining computed tomography and AI to obtain an objective tomographic reconstruction of multicomponent all-metal aggregates "assembled." The technology builds a digital engine twin and makes it possible to reliably detect defects - cracks, voids, delamination - and foreign objects such as metal chips in one measurement.
The goal of Smart Engines scientists was to scan the entire jet engine - indiscriminately for parts - and in one measurement on a serial industrial installation. Smart Engines researchers have proposed a comprehensive solution to the problem - the use of intermediate "compromise" energy radiation ~ 450-600 keV, which allow you to reconstruct both denser and less dense parts, as well as high-performance algorithms computer tomography to correct the distortions that occur.
The developed algorithms make it possible to reliably compensate for the effects of scattering of radiation by an object, scattering of radiation by a detector, polychromatic radiation during the study of multimaterial products, as well as noise associated with photonic starvation. AI allows high-quality reconstruction even in the case of complex measurements made on the verge of sensitivity of the device, as a result of which it became possible to carry out flaw detection of aircraft jet engines models indiscriminately into individual components.
 | Engine failure is a disaster. To prevent it, it is not enough to check each part. If there is no control over the finished product, the consequences will be sad, although, as Raikin said, there are no complaints about the buttons. With our technology, it is possible to put quality control of assembled engines on the conveyor, taking their reliability to the next level. told Vladimir Arlazarov, CEO of Smart Engines, Doctor of Technical Sciences |  |
The most promising direction for the further development of the technology is scaling for the study of large and long-sized products, such as aircraft engines and space rockets. This requires the development of next generation installations that will have sufficient source brightness to pass through such large metal products and have dimensions corresponding to the products.
2024
Obtaining the status of software with artificial intelligence
The Ministry of Digital Development of Russia has assigned software solutions of the Russian company Smart Engines the status of software related to the field of artificial intelligence. The corresponding changes were made to the Unified Software Register. Smart Engines announced this on August 29, 2024.
Now users of all flagship products of the company, including, Smart ID Engine, and Smart Document Engine Smart Code Engine Smart Tomo Engine, can receive additional support from the outside, states including by reducing to tax profit payments. More. here
Obtaining a patent for technology for oil field research
Smart Engines scientists have introduced and patented technology in the United States that will allow companies to more efficiently explore oil fields. The company announced this on July 24, 2024. The developed algorithms build a digital twin, which can be used to analyze the digital core, and detect false "densifications" of the rock, which reduces risks when deciding to start production.
A core is a rock sample extracted from a well using a core collector. Creating a digital 3D core model using x-ray the tomography method algorithms artificial intelligence and allows analysis virtual and experiments with the reservoir rock. The digital core is one of the most accurate methods for assessing nanoporosity and, as a result, helps to determine the amount of oil in the rock, to assess the potential of the studied fields.
However, when building a digital core twin, false "compaction" can occur due to bowl artifacts, which hides the pores of the core. It is such artifacts that are able to suppress the algorithm of Smart Engines scientists described in the patent.
Cup-shaped distortions as well as bands between high absorption regions arise from the use of polychromatic radiation for probing, that is, wideband spectrum radiation. The reason for this lies in the fact that classical computed tomography approaches to reconstruction data are designed so that the energy of all photons will be the same (i.e. data recorded in the monochromatic case). X-ray monochromators can be used to combat them, but this reduces the effectiveness of the method. Another approach is to develop more complex and accurate mathematical models underlying the reconstruction. It is to this type that the invention of Russian scientists belongs.
The correction method presented by Smart Engines scientists allows, when processing projection data collected in polychromatic mode, to reduce the number of artifacts associated with increasing beam stiffness, without the need for additional calibration experiments or changing the hardware. Its essence lies in the fact that after determining the optimal correction value, the projection data is linearized (in other words, they are transformed in such a way that they can be processed by reconstruction algorithms), and then the final image is reconstructed from these linearized data. At the same time, the main advantage of this method is the automatic adjustment of correction parameters.
To create 3D models of the core, a microtomograph with a software product for tomographic reconstruction of the Smart Tomo Engine is used. The artificial intelligence system allows you to obtain a reconstructed three-dimensional image of an object in high resolution, compare with an ideal model for the correspondence of composition, size, the presence of cracks and other hidden defects. The program is intended for use in the field of industrial tomography in the study and quality control of the internal structure of objects of different sizes: from microcircuits to industrial internal combustion engines.
In 2024, this is the second US patent received by Smart Engines. The previous one was associated with a model of neural networks - bipolar morphological networks.
Red OS Compatibility 8
Smart Engines has tested the compatibility of the latest version of its technologies with Red OS 8. The company announced this on March 28, 2024.
Previously, the company's software products - Smart ID Engine, Smart Code Engine, Smart Document Engine and Smart Tomo Engine - received the 2.3.0 update. During testing, it was confirmed that they work correctly with the operating system of Red Software.
| | Smart Engines is committed to making all of the company's solutions cross-platform. And, of course, it is important to ensure the operation of document recognition algorithms that are used in key processes in banks, government agencies and other organizations, - said Vladimir Arlazarov, Doctor of Technical Sciences, General Director of Smart Engines. | |
2023
Smart Tomo Engine 2.2 with the ability to play only the area of interest
Smart Engines on October 11, 2023 presented an updated version of the Russian software product for tomographic reconstruction Smart Tomo Engine 2.2. The artificial intelligence system allows you to obtain a reconstructed three-dimensional image of an object in high resolution, compare with an ideal model for the correspondence of composition, size, the presence of cracks and other hidden defects. The program is intended for use in the field of industrial tomography in the study and quality control of the internal structure of objects of different sizes: from microcircuits to industrial internal combustion engines.
In the updated version of the software product, you can now recreate the image of not the entire investigated object placed in the tomograph, but only the area of interest. This is important, since often the detector's working area exceeds the object itself and a lot of uninformative space is formed, which does not need to be restored.
In addition, the task is often to examine only part of the object - for example, a repaired piece of the product, be it a weld or part of a microcircuit. Depending on the size of the object, the reconstruction time can be reduced to 50-70%. The use of tomographic complexes with the new version of the Smart Tomo Engine will help speed up production processes in mechanical engineering, rocket and space, oil and gas, microelectronics and other areas.
В Smart Tomo Engine 2.2. a fundamentally different data loading mechanism is implemented. Now the user does not need to wait until all the measured data is loaded, which can be very large to obtain a digital image of a high-resolution object. In the new version, the processes of loading data and selecting the reconstruction mode are carried out in parallel. While the data is loaded, the user in preview mode can literally "on the fly" configure the reconstruction parameters. This not only made working with the software product more convenient, but also noticeably accelerated it through background loading.
In addition, this version has improved the user interface, as well as increased convenience with the program.
Smart Tomo Engine 2.2 can be installed both on tomographs already in operation to replace standard software, and can be included as a software part in newly created tomographic complexes. The system is included in the register of domestic software and is compatible with both domestic and foreign tomographic devices.
Smart Tomo Engine 2.0 with Industry Expansion
Smart Engines on March 23, 2023 announced the entry into the industrial tomography market. The relevance of the task of non-destructive quality control of the internal structure of parts has increased sharply with the arrival of additive technologies, said Vladimir Arlazarov, CEO of Smart Engines, Candidate of Technical Sciences. X-ray tomography easily copes with this task.
| | Over the past four years, Smart Engines has invested 200 million rubles in computed tomography. We are releasing the second version of our Smart Tomo Engine program. This is a completely Russian product for the reconstruction and analysis of 2D and 3D CT images. Smart Tomo Engine in combination with a domestic tomograph will allow you to get a special service. It will be possible to compare the restored 3D digital image of the object, obtained in high resolution, with the ideal model for compliance with the composition, dimensions, the presence of cracks or other damage, as well as hidden defects - said Vladimir Arlazarov. | |
In general, the second version of the Smart Tomo Engine has expanded the scope of software application to industry, added the CEO of Smart Engines. As he explained to TAdviser, in difficult circumstances it is important for the company to offer the Russian industry a high-quality software product that can be used to control quality, create new materials and production technologies.
The Smart Tomo Engine 2.0 system is designed for accurate tomographic reconstruction of a three-dimensional digital image of an object based on a set of its transmission projections in the X-ray range (X-rays).
Compared to the previous version of the Smart Tomo Engine, this system has improved the quality of the resulting images, significantly improved performance, improved technological compatibility, and an expanded list of compatible tomographs. The Smart Tomo Engine 2.0 also automated the operator's routine tasks to select the best reconstruction parameters and introduced the latest technologies of domestic development.
| | For example, for cases when an object does not fit entirely in the detector type field, the author's patent algorithm FOVEA has been added, which improves the image in the view field and evaluates the background structure, "said computer Marina Chukalina, head of the Smart Engines tomography department. | |
For design work in this system, neural network models were developed that allow high-quality reconstruction at an extremely low dose load (low-angle shooting and small exposure), Marina Chukalina noted.
The line of supported computing platforms in this version of the Smart Tomo Engine has been supplemented with Russian products. For March 2023, it is represented by x86-64, ARM, MIPS, Komdiv and Elbrus architectures for CPU and CUDA for GPU.
2022: Inclusion in the Unified Register of Russian Software
The Smart Tomo Engine tomographic reconstruction software complex, developed by scientists from Smart Engines, is included in the Unified Register of Russian programs for electronic computers and databases. This was announced on December 30, 2022 by representatives of Smart Engines.
According to the order Ministry of Digital Development of the Russian Federation of December 30, 2022, the Smart Engines solution is included in the class that ON includes data mining () Data Mining and processing () tools Big Data Big Data , processing, analysis recognitions and image tools, as well as subroutine libraries (). SDK The Smart Tomo Russian programs Engine entered the register under No. 16192.
Smart Tomo Engine is an artificial intelligence technology that allows you to perform a three-dimensional reconstruction of an X-ray image immediately during the process of tomographic photography and stop when a result is achieved that has sufficient quality. The resulting digital 3D image allows the doctor to easily diagnose, and the technologist to build an expert assessment of the quality of the part.
The completely Russian development allows you to reduce the dose of radiation that a patient or part receives during a CT study. Using tomographic reconstructions performed by the Smart Tomo Engine, it is possible to reduce the registration time or exposure of one image.
| | "The number of CT tests performed is increasing every year. In 2016, 8.9 million procedures were carried out in Russia, and in 2018 - already 10 million studies. Therefore, the issue of reducing the radiation dose is becoming more and more urgent. We are glad that Smart Engines technologies make it possible to at least partially solve this problem, since our solution can be used to examine patients with a large number of diagnoses, including COVID-19, "said Vladimir Arlazarov, CEO of Smart Engines. | |
The software product works autonomously on the local workstation, does not require a network connection and does not transfer data to third-party servers. The inclusion of Smart Tomo Engine in the register of domestic software confirms its compliance with the established rules and requirements of Russian legislation.
2021: Compatibility with Red OS
The developers RED SOFT Smart Engines have confirmed the compatibility operating system RED OS and reconstruction systems X-ray tomography of Smart Tomo Engine images. Tests were carried out on a computing platform processor Baikal-M from the production the Russian of the company "." Baikal Electronics Smart Engines announced this on September 30, 2021. The obtained results confirm the possibility of creating a completely domestic software and hardware complex, which has sufficient performance to solve the problems of computational tomography industrial medical in both industries.
Smart Engines scientists have created the Hough Filtered Back Projection algorithm, which provides a high rate of tomographic reconstruction. This algorithm, together with Monitored Reconstruction technology, also developed by Smart Engines scientists, significantly reduces the beam load by interrupting the collection of radiographs in real time as soon as their set is sufficient for recovery.
| | Providing critical information infrastructure facilities, including medical institutions, with the necessary tools for working on the basis of domestic information technologies is a state priority. Together with partners RED SOFT offers suitable solutions. Another example is the compatibility of products that confirm the possibility of creating a PAC for X-ray tomography Smart Tomo Engine based on the Baikal-M processor with the RED OS operating system on board, "said Rustamov Rustam, Deputy General Director of RED SOFT. | |
| | Tomographic reconstruction using Smart Tomo Engine software and RED OS operating system based on the Baikal-M architecture processor opens up prospects for creating fast and safe tomographs based on Russian hardware and software. We are glad that cooperation with RED SOFT allows us to offer the domestic market a safe solution for the reconstruction of X-ray images obtained during tomographic examinations, - said Vladimir Arlazarov, Ph.D., CEO of Smart Engines. | |
| The site content is translated by machine translation software powered by PROMT. The machine-translated articles are not always perfect and may contain errors in vocabulary, syntax or grammar. Read original article If you find inaccuracies or errors in the results of machine translation, please write to editor@tadviser.ru. We will make every effort to correct them as soon as possible. |