DIN
When artificial intelligence recognizes the tumour
Analyzing and recognizing images is crucial in many medical fields, from cancer cell diagnosis to blood count determination. Until now, this has mainly been done by humans. Deep learning image recognition systems can automate the process with AI.
DIN SPEC 13288 'Guidelines for the development of deep learning image recognition systems in medicine' formulates specific requirements for this. Among other things, it contains practical guidelines for the development and structure of the systems and deals in particular with the handling of data for the training of artificial intelligence (AI).
The guideline is based on DIN SPEC 13266 'Guidelines for the development of deep learning image recognition systems' and expands it with regard to the special requirements in medicine. "Higher quality standards apply in this area than in others, and it is more heavily regulated. That's why the new standard contains separate explanations and stricter requirements," explains Felix Faber from MindPeak, initiator of the DIN SPEC. The AI on which the systems are based learns statistical patterns from sample data and can therefore solve complex image recognition tasks. DIN SPEC 13288 defines special requirements for this data and the corresponding training of the AI for medical use. "The specification also focuses on ensuring that the results are explainable. This is the only way to create trust and acceptance of the technology among doctors and patients alike," says Faber.
The new standard is aimed in particular at manufacturers of deep learning image recognition systems and those involved in research and development projects.
You might also be interested in
EU Commission wants to protect victims of artificial intelligence
According to the EU Commission, anyone who suffers damage due to artificial intelligence should be able to claim compensation more easily in future. According to the Commission, this should also increase confidence in new technologies.
read more...AI research should be geared towards industry
The time for broad-based research into artificial intelligence in Germany is over, says Peter Liggesmeyer. Instead, more attention should be paid to the requirements of industry.
read more...Predictive maintenance with artificial intelligence
Data generated by machines and systems is used to detect and prevent faults at an early stage. However, they can only develop their full potential through intelligent linking with the relevant data from different source systems. Insight engines help...
read more...AI marketplace goes live
Work on the AI marketplace project took a year. Now the platform for artificial intelligence is going live.
read more...Start-ups drive AI forward
Artificial intelligence has arrived in mechanical and plant engineering. More and more companies are using the technology and cooperating with start-ups.
read more...Nanya accelerates the use of AI in production
Nanya, the world's fourth largest manufacturer of DRAM, uses artificial intelligence to increase productivity and quality in production. Now, with the help of Hewlett Packard Enterprise, the roll-out of new AI applications in the semiconductor...
read more...Optimized for blue light
Within the visible spectrum, monochromatic image examinations with blue light are the most effective. The lenses in the Blue Vision series from Vision & Control are specially designed for this spectral range.
read more...AI marketplace launched
On the IDS virtual AI marketplace, end customers have access to complete image processing solutions and can choose the right, ready-made application. This should open up new opportunities for image processing projects, especially for SMEs.
read more...Bringing AI to the edge
With the UP Xtreme i11 Edge Compute Enabling Kit from the 'UP Bridge the Gap' brand, Aaeon shows how artificial intelligence can be brought to edge computing.
read more...