This medical imaging application data analysis for arkelson gives the answer

altera

If the patient the whole diagnostic process as a paper, medical imaging is like an application problem. Answers to this problem often, is based on the doctor’s eye and clinical experience to judge. However, even experienced doctors in rich, high-precision imaging equipment, are likely to be misdiagnosed.

Based on a misdiagnosis of the China Medical Association data shows that total Chinese clinical medical misdiagnosis rate of 27.8% organ misdiagnosis rate of ectopic 60%, average misdiagnosis rate of 40% malignant tumors, such as nasopharyngeal cancer, leukemia, pancreas, liver and tuberculosis, stomach tuberculosis pulmonary tuberculosis misdiagnosis rate is also above 40%.

Misdiagnosis rates are so high, in fact, not all because the doctor inadvertently, and there are deeper reasons. Founded in 2006, arkelson technology, also for this problem and thinking about that year.

Examines the

Diagnostic Imaging is divided into two categories: structural imaging and functional imaging.

Understand imagery and simple structure, is “what you see is what you get” image, such as x-ray, CT belong to this one, it is visually observed physiological structure, determine whether there is a physical change of the lesions, the equivalent of “we comin” game, but it has no obvious physical changes for some diseases, should do nothing.

Functional imaging, not too accurate metaphorically, is the equivalent of structural images of the “complement”. This type of imaging to study organ cells metabolic capacity of a substance, so as to reflect the function of this organ is normal. Machine check condition of radioactive tracer in human metabolism, and record reactions on energy metabolism of data matrix through a two-dimensional image film to render.

Of course, its weakness is also very intriguing: cannot be located, the quantitative analysis of the existing tools do not achieve statistical significance. Image does not reflect true physiological structure, can only be expressed through the image pixel shading level metabolism intensity, is abnormal, and doctors can’t read data matrix. In this way, results only by doctors and experience to see with the naked eye. Even the experts, misdiagnosis rate in 30%-50%.

Multi-dimensional data matrix function behind the image, if accurately quantified on the basis of by statistics than with the normal group, has practical implications in disease diagnosis. Hermes case

anestcadiz

Problem-solving methods

Arkelson medical imaging technology to solve the “problem” is to help doctors accurately quantified. “We don’t deal with surface images, image behind the data mining analysis matrix, combined with China’s normal database by statistical comparison, use scientific means to come to the conclusion, not subjective judgments by experience, increase the accuracy to 95%. “, Says CEO Chen h. Through an analysis of medical image data, automatic diagnosis–this is arkelson way of artificial intelligence.

In 2004, the arkelson technology was originally based IT services software for hospitals, imaging device reporting system. But the founders found, although finished developing the PACS, but has several old pits manufacturers, this was completely dominant.

Transformation of thinking way, experimental subject with the Xuanwu hospital on quantitative analysis of brain, inspired arkelson of inspiration. Experiments need a patient, suitable for China based on SPM analysis software quantitatively. SPM (Statistical Parametric Mapping) the statistical parametric mapping, quantitative analysis of brain images in the world of cutting-edge theory, also was recognized as the industry standard. Based on implementation of SPM results can help quantify the doctors diagnosis and break through the traditional limitations, that is, rates of up to 99% confidence.

CTO Li discuss with teams that, if the thought of quantitative analysis used in the analysis of other organ diseases, this will be the major subversion of traditional diagnostic mode, something worth doing. So in 2008, arkelson began to study theories of SPM, Li began learning of medical knowledge, mathematical reserves. “The thought that 2–3 can implement software, actually found and not so simple: the application of mathematical methods to solve a medical problem, this has gone beyond mere software development. “CEO Chen h said. Hermes S6 case

The application of medical image problem, naturally also surfaced: how to start from the functional imaging, quantitative, positioning, precise medical diagnosis, exclude to the greatest degree subjective, experience disadvantage, address the high rate of misdiagnosis and missed diagnosis of medical imaging problems.

Rather than spell speed intelligent hardware industry, medical imaging rush, no quick fix. After nearly 3 years after SPM and basic knowledge learning, arkelson teams begin strategic cooperation with a number of key national hospital, brain, heart, liver, spleen, tumours and multiple disease complicated by SPM in the areas of technical cooperation.

Faced with rapid advances in medical technology, arkelson technology also need the write answer. After several years of development, “Bohr” precise quantitative analysis platform was born–this “genius”, finished arkelson in precise image analysis goals.

The answer sheet

In “the pen” before the answer, Jassen precise clear images of several major processes of science and technology.

First of all is to collect imaging data. Read image data from imaging devices, or PACS, followed by a series of image correction.

Second is normalization and standardization of images. Through the establishment of different biophysical models of organs, to multidimensional normalized image matrix preprocessing, minimizes the individual lead acquisition difference. In simple terms, normalization is a data image are adjusted to the same standards of different models, and then compared.

Afterwards, will be statistical comparison test data and which database. One of the most important, is in line with the standards of normal database, the library needs to satisfy two conditions:

Contains enough dimensions, including age, gender, medical history, race, dimension more analysis more precise;

Must go through rigorous screening, such as basic intelligence, related medical history, history of mental illness, such as brain diseases.

Current data sources for two main aspects of the arkelson, case data from the cooperative study of hospital data, while normal for company-paid collection of data.

China Group database in order to focus the precise detection, location, quantity, volume. According to Chen h, this process is now up to 99% confidence.

Final arkelson will give out diagnostic advice, charge advice. Department of functional imaging and quantitative analysis of precision, mainly reflected in the description of the abnormal area, including location, size, quality, severity and recommends medication. We will use artificial intelligence methods in the future, through medical records for further processing, automated image analysis and diagnosis.

Up to now, arkelson technology is unique in the market can apply for CFDA-3 medical software companies. What concept? CFDA-3 refers to the treatment planning software, decision support software, including calculation software for automatic diagnosis software and drugs. CFDA-2 refers to HIS, RIS/PACS hospital information system, usually for medical information and data management, viewing, transfer and storage, and does not include automated diagnosis.

CFDA medical software rating

Arkelson answers to questions pay-business model, there are two floor models and SaaS cloud service models.

Former principal for three and top foreign-owned private hospital, purchased License or by providing accurate analysis platform for free, monthly or per-charge split the database used by the service; the latter mainly for the level-II hospital, cloud platform provides analysis and reports, there is further demand transmission diagnosis guide correspond to industry experts to get.

By 2012, the arkelson algorithm for complete image processing and technology platforms by 2014 to enter clinical trials, started production.

Arkelson science and Technology Research Institute President Li gang said, “we believe that the development of quantitative imaging data, will inject vitality into the medicine into the new era of digital imaging; accurate diagnosis for medical development leap provides a steady flow of power. ”

The application of medical image problem, arkelson took 7 years to be formally surrendered to a satisfactory answer. Believe development of medical diagnostic imaging, will better with optimization of quantitative analysis.

If you are interested, welcome to click here to watch the arkelson technology video.

Thunder wear is Lei feng’s network (search for “Lei feng’s network”, public interest), its public, and attention to sports and health-related products and technologies. If you want to talk about your story, you can add small series of micro-417423625 or send an email to guoyixin@leiphone.com.

Like this:

Like Loading…

Related