DP Technology Launches Upgraded Uni-AIMS AI-Powered Microscopy Imaging System with All-in-One Appliance

Oct 10, 2024 19:38 CST Updated 19:38

DP Technology

Simulation R&D Platform Developer

DP Technology Launches Solution for Electron MicroscopyIntelligent Characterization Analysis System Uni-AIMSBased on deep learning algorithms and representation pre-training large models, Uni-AIMS achieves precise recognition of microscopic images and performs a series of representation analysis calculations, automatically generating intuitive and comprehensive statistical reports, which can significantly improve efficiency.The Intelligence Level of Microscopic Image Characterization Analysis, reducing the corresponding characterization analysis processMore than 95%Human resources investment, acquisitionHigh-Quality and Comprehensive Consistency Analysis Results。In response to the demand for lightweight deployment and data localization, fully launchedAll-in-One Function Treasure Box, achieving a brand-new experience of seamless integration of hardware and software, ready to use right out of the box. The intelligent characterization analysis system Uni-AIMS provides strong support for the characterization analysis phase in accelerating scientific research on new materials and drugs, and offers crucial evidence for production quality evaluation and guidance on optimizing production directions.

Advances in experimental characterization techniques play a crucial role in promoting the development of science and technology. In the 17th century, Antonie van Leeuwenhoek observed microorganisms using a homemade microscope, marking the birth of microbiology as this was the first time humans had observed single-celled organisms. The application of thermogravimetric analysis (TGA) in the field of polymer materials has facilitated researchers' in-depth studies on thermal stability, compositional analysis, and thermal decomposition mechanisms of materials. In-situ X-ray spectroscopy techniques have revealed microscopic structural changes in battery materials during charge and discharge processes, which is of great significance for improving battery performance and developing new battery materials.

In the scientific research process across multiple key fields such as materials science, new energy, biomedicine, and environmental monitoring, a large amount of experimental data containing fundamental behavioral characteristics of substances has been generated. How to quickly and accurately process these complex data sets, discover patterns within them, and thereby obtain true and objective characterization information is very important for advancing scientific research.

Scanning Electron Microscope (SEM)）By using a focused electron beam to scan the sample surface and capturing the signals generated by the interaction between the electron beam and the sample through a detector, surface images are generated, which haveHigh-Resolution Accuracy(Nanometer level),Easy to Operate、Feature-RichFeatures such as these have enabled this technology to gain widespread application in disciplines like materials science, biology, and chemistry in recent years, continuously driving research progress in these fields. In the field of battery materials, scanning electron microscopy (SEM) can be used to analyze the particle morphology characteristics of cathode and anode materials, such as particle size, shape, and distribution, for standardization and quality control of materials and products. In the field of metallic materials, SEM allows for the characterization of grain and second-phase particle distributions, while SEM equipped with EBSD can accurately characterize grain orientation and texture in metallic materials, enabling the analysis of the structure-property relationship in metal microstructures. In the biomedical field, the identification and characterization of pathogens and large biomolecular particles provide crucial evidence for disease diagnosis and treatment. In environmental science, the identification and characterization of atmospheric particulates contribute to assessing air quality and understanding the impact of particulates on human health.

Figure 1 SEM Image Characterization Types Are Rich and Diverse

However, traditional SEM image analysis mainly relies on manual statistics by experimental technicians and some manually annotated software, which strongly depends on the technicians' experience. The efficiency of image analysis is low, the consistency of data over time is poor, and some key characterization results are difficult to obtain directly. This leads to the objective material behavior information reflected in the image characterization results not being fully utilized, and may even introduce some guiding biases in research directions. There is an urgent need for new intelligent tools for the characterization analysis of SEM micrographs.

Uni—AIMS Intelligent Characterization Analysis Software

Figure 2Uni-AIMS Microscopic Image Intelligent Analysis Tool

Uni-AIMS (AI-Powered Microscopy Imaging System )Developed by DP Technology, Uni-AIMS is an electron microscopy image analysis software that integrates cutting-edge artificial intelligence technology. As a tool for intelligent analysis of microscopic images, Uni-AIMS uses deep learning algorithms to automatically identify objects in electron microscopy images and perform characterization calculations automatically. It supports one-click output of multi-faceted statistical reports, providing efficient and comprehensive data resources for scientific research exploration and production optimization.

Uni-AIMS' AI engine adopts the latest technology to achieve high-precision automatic recognition of targets in electron microscope images, completing high-accuracy image analysis without human intervention. This enhances the automation and intelligence level of image analysis and significantly improves analysis efficiency and accuracy.

The software adopts a one-click operation process, equipped with an intuitive and user-friendly interface, allowing non-professional users to get started quickly. Meanwhile, Uni-AIMS provides rich data charts and in-depth analysis reports, offering solid data support for material research and development.

After the completion of this version update, the Uni-AIMS software has been launched on the Bohrium® Scientific Computing Platform. Compared with the SEM Particle Detector tool released at the end of last year, it has been significantly upgraded in terms of software functionality, application scope, recognition accuracy, and functional dimensions. Currently, the main features of the Uni-AIMS product, developed based on the pre-trained large model foundation for characterization, are:

1.Fast parsing speedThe use of Uni-AIMS software has significantly increased the speed of characterization analysis. For an SEM image containing approximately 1000 particles, it takes half an hour for manual processing to perform particle size distribution statistics, whereas with Uni-AIMS software, it only takes 15 seconds.

2.Results are reliable with high accuracy.Uni-AIMS software significantly reduces human error in measurement and analysis, ensures the consistency and reproducibility of experimental results, promotes the standardization of experimental design, and markedly enhances the efficiency and precision of material research and development.

3.More Comprehensive Analysis. It can analyze some important indicators that are difficult for humans to obtain, such as sphericity, second phase proportion, etc.;

4.Strong generalization abilityBy fine-tuning a pre-trained model with a small amount of electron microscopy images from a completely new field, the ability to analyze representations in that new field can be obtained.

5.Beautiful interface, convenient operation.Full UI interface operation, supporting various visualization forms such as violin plots, histograms, and multi-graph comparisons. Characterization reports can be exported and downloaded with one click.

Application Scenarios

Application Scenario 1: Scene of Particle Analysis for Lithium Battery Cathode Materials

Figure 3: Microscopic image analysis has wide applications in the characterization of battery materials.

NCM Ternary Material, namely lithium nickel cobalt manganese oxide (LiNi_xCo_yMn_zO2), is a high-performance cathode material used in lithium-ion batteries. This material is valued for its high specific capacity, excellent discharge voltage plateau, good cycle stability, and cost-effectiveness.

Ternary Material Cathode Particles and Their Precursor ParticlesParticle Size Distribution (Uniformity), Sphericity, Specific Surface AreaThis will directly affect the electrochemical performance of lithium batteries and is crucial for the energy density and safety of the battery. A particle size distribution that is too narrow or too wide can reduce the compaction density of the material. An appropriate particle size distribution will increase the tap density by filling the voids between larger particles with smaller ones; however, an overly wide particle size distribution may cause slurry stratification, affecting the tap density. Additionally, varying degrees of overcharging and over-discharging between larger and smaller particles can easily lead to a decline in the battery's cycle life.

The figure below shows the SEM images of ternary precursor particles produced through a synthetic process: a) SEM image of ternary precursor particles; b) SEM image recognition results of ternary precursor particles (yellow lines indicate identified grain boundaries).

Figure 4 Automatic recognition effect of ternary precursor particle microscopic images

Particle size analysis results are shown below:

Figure 5 Automatic Analysis Effect of Ternary Precursor Particle Size Distribution

Sphericity analysis results are shown as follows:

Figure 6 Automatic Analysis Effect of Sphericity Distribution of Ternary Precursor Particles

Application Scenario 2: Metal MaterialsGrain Analysis Scenarios in DP Technology

Metallographic AnalysisIt is a characterization method used in materials science to study the internal microstructure of metals and alloys, which is crucial for understanding and controlling the microstructure and macroscopic properties of materials. The characterization analysis of alloy profile grains can extract typical features of the microscopic morphology of alloy materials, such asGrain Size Distribution, Grain Morphology Characteristics(isometric, rod-like, plate-like, etc.), and then study the structure-property relationship between microstructure and macroscopic properties.

Among these factors, the size and morphology of metal grains significantly influence the mechanical properties and anisotropy of alloy materials. According to the classical Hall-Petch relationship, a reduction in grain size generally leads to an increase in the material's strength. This is because grain boundaries add obstacles to dislocation movement, thereby enhancing the material's yield strength.

The figure below shows the results of automatically analyzing the EBSD image of extruded AZ31 magnesium alloy using the Uni-AIMS tool, including key outcomes such as grain distribution and grain morphology characteristics: a) EBSD image of extruded AZ31 magnesium alloy; b) EBSD image recognition results of extruded AZ31 magnesium alloy (yellow lines represent the identified grain boundaries).

Figure 7 Automatic recognition effect of magnesium alloy EBSD images

Particle size analysis results are shown below:

Figure 8 Automatic analysis effect of magnesium alloy grain size distribution

Shape Factor A/R Analysis Results are shown below:

Figure 9 Automatic analysis effect of magnesium alloy grain morphology distribution

Operation Interface Display

Uni-AIMS features a fully visualized operational interface, allowing users to quickly upload SEM images for analysis through interactive controls, and perform tasks such as recognition, filtering, statistics, and report generation. The workflow is illustrated below:

Create a New Session

Upload Image

Select Single or Multiple Images to Upload

Image Recognition

Preview Recognition Effect

Switch recognition results via buttons for convenient and intuitive comparison of recognition effects.

a) Display Recognition Boundary:

b) Hidden Identification Boundary:

c) Support manual modification of recognition results and scale:

Screening and Filtering

Support filtering and targeting particles by numerical ranges of indicators such as particle size, perimeter, sphericity, and aspect ratio.

Summary Statistics

a) Support direct presentation of the distribution of key indicators such as particle size, sphericity, and perimeter, along with the values of P0, P10, P50, P90, and P100;

Particle Size Distribution Presentation Effect:

Sphericity Distribution Presentation Effect:

b) Support for switching between histograms and violin plots for distribution effects:

c) Support scatter plots for any two attributes of all particles, which can be used to preliminarily test correlations:

d) In the case of uploading multiple images, it also supports the comparison of effects among multiple pictures:

Analysis Report Export

The above is only a demonstration of the operation of commonly used functions in Uni-AIMS. More rich features await your exploration!

The product demo video is shown below:

New All-in-One Function Treasure Box

Achieve Rapid Delivery and Data Security Assurance

In order to meet the needs of some enterprisesLocalization, LightweightDeployment Requirements: DP Technology Officially Launches Uni-AIMS Intelligent Characterization and Analysis All-in-One Machine for Out-of-the-Box Use. The all-in-one machine is equipped with mainstream high-performance CPUs, high-performance GPUs, redundant hot-swappable hard drives, dual power supplies, and other configurations to ensure a smooth user experience while meeting the high concurrency, high availability, and security needs of enterprise users. The all-in-one machine primarily adopts a tower form factor but also supports deployment using rack servers with equivalent configurations, allowing for easy deployment in office or server room environments. Important information such as images, data, and reports are stored and used locally, ensuring maximum protection of enterprise data security. The appearance of the all-in-one machine box is shown in the image below:

Figure 10: Integrated Machine Appearance Picture

Facing the Future

Uni-AIMS has demonstrated significant application potential and market value in the characterization analysis process across multiple research fields.

In the future, Uni-AIMS will continue to iterate and upgrade throughPre-trained Representation Large Model, combined with a small amount of downstream image annotation, automatically adaptsMulti-domain, Multi-scenario, Multi-category, Multi-scaleThe microscopic characterization analysis scenario provides more efficient and precise intelligent characterization analysis services for researchers and production line engineers in the fields of materials, energy, pharmaceuticals, and environment.

Apply for Trial

Uni-AIMS Basic Function Trial Version:

https://bohrium.dp.tech/apps/uni-aims

About DP Technology

DP Technology is a leader and practitioner of the "AI for Science" research paradigm. AI for Science involves using AI to learn a series of scientific principles and knowledge, and further address key issues in scientific research and industrial R&D.

DP Technology, relying on its deep cultivation in interdisciplinary fields, has built "Deep Potential · Cosmic Insight^®"The AI for Science large model system brings research methods in many disciplines from the 'experimental trial and error/computer' era into the 'pre-trained model era,' using 'micro-scale industrial design and simulation' as the entry point to create...Bohrium^®Bohr^®Research Space Station, Hermite^®Drug Computational Design Platform, RiDYMO^®Difficult-to-Drug Target R&D Platform and Piloteye^®Research and industrial R&D infrastructure such as battery design automation platforms have formed the "innovation-implementation" chain and open ecosystem for AI for Science, empowering "thousands of industries." This creates a new generation of industrial design and simulation systems for the most fundamental areas of human economic development: biomedicine, energy, materials, and information science and engineering research.

DP Technology is a national high-tech enterprise and a national specialized and new "little giant" enterprise, with research and development centers located in cities such as Beijing, Shanghai, and Shenzhen. The scientific research and technology team is led by an academician of the Chinese Academy of Sciences, bringing together over a hundred outstanding young scientists and engineers from various fields such as mathematics, physics, chemistry, biology, materials, and computer science. Among the company's members, more than 35% are PhDs and postdoctoral researchers. Core members have received the "Gordon Bell Prize," the highest award in the global high-performance computing field in 2020, and their related work was selected as one of China's top ten scientific advances in 2020 and one of the ten major technological breakthroughs in the global AI field.

DP Technology Launches Upgraded Uni-AIMS AI-Powered Microscopy Imaging System with All-in-One Appliance

DP Technology

Application Scenario 1: Scene of Particle Analysis for Lithium Battery Cathode Materials

Application Scenario 2: Metal MaterialsGrain Analysis Scenarios in DP Technology

Create a New Session

Upload Image

Image Recognition

Preview Recognition Effect

Screening and Filtering

Summary Statistics

Analysis Report Export