Artificial Intelligence is an emotional subject. For some it may send the blood pressure soaring or maybe trigger another medical condition, perhaps by conjuring up images of robots or humanoid creatures taking over the world. Of course, Sci-Fi could have a lot to do with these common types of reaction: Brave New World, The Terminator, The Matrix….
But, for the sake of our health, let’s not get too carried away. After all, we’re increasingly exposed to basic levels of AI in our daily lives…with such things as Alexa and Siri, for example, and ‘chatbots’ that pop up on websites when we’re doing online shopping and banking...you may like them or loathe them, but they’re not scary! Some may even say they offer convenience and a sense of ‘wellbeing’.
Speaking of which, in the medical sciences and healthcare arena, considerable strides are being made in the application of the very latest AI techniques, notably Machine Learning, Deep Learning and Neural Networking. You may recall a well-publicized story from last year about a leading dermatologist in Germany who issued a challenge to fellow medical skin cancer experts from around the world: Could they beat his prototype AI neural network system in diagnosing images of historic potential melanoma cancer cases? On this occasion man was thrashed by machine with only 13 of the 58 dermatologists involved managing to beat the algorithm in correctly identifying more actual cases of the skin cancer rather than harmless birthmarks.
For many in the medical profession these AI-enabled developments are welcome, not least due to lack of time, skills shortages, and budget constraints
AI-enabled algorithms used in medical imaging and diagnostics, for example, can achieve results in seconds rather than the hours or days using the human approach. It has the potential to touch almost everything, from tomography (MRT, CRT) systems to ultrasound diagnostic devices, as well as mobile devices for use in diagnostics and care.
Machine learning and deep learning analyse data sets and learn from them to make specific predictions about patient healthcare concerning an increasing array of medical conditions - from different types of cancers and dementia, to kidney and cardio-vascular disease. Deep learning has a kind of ‘sixth sense’ that can even make predictions from discovering data patterns that humans might otherwise overlook, providing early warning signals which may avert an illness or medical emergency such as heart or kidney failure. It can even suggest possible treatments based on tens of thousands of similar recorded medical cases. In some cases, patients will be able to receive diagnoses and medical treatment advice via mobile applications.
In the end, the more medical data becomes available for analysis and unification, the better AI-enabled machines and systems will become at supporting medical professionals when undertaking complex analytical tasks.
However, Machine Learning and Deep Learning magnify medical systems performance requirements
Medical OEMs and systems developers are in growing need of more powerful platforms for supporting their real-time and graphics-intensive AI medical imaging solutions.
Servers must be equipped with the latest multicore CPUs for enabling the massive parallel processing performance now required in medical imaging and AI-supported diagnostics. High speed GPU performance is naturally of major importance too. Deep learning algorithms accelerated on industry leading GPUs can reduce neural network system learning time from weeks to hours.
Kontron’s latest server brings ‘KISS of Life’ to AI-powered medical platforms
Responding to the compute challenges and massive potential of AI in healthcare, Kontron has recently introduced its most powerful KISS rackmount server. It features Dual Intel® Xeon® SP series processors, allowing real-time compute-intensive processes for analysing large amounts of data. Up to three double-width GPU cards (NVIDIA TESLA V100) ensure extremely high graphics performance, and for extended storage, up to eight 2.5" storage trays can be installed.
Like the company’s other KISS server platforms, the super-powerful KISS 4U V3 SKX is based on industry standard components for enabling ease of configuration and ease of maintenance. The flexible, modular design also allows easy adaptation to medical OEM customer-specific requirements. Crucially, its reliability means it can be used for 24/7 operation at consistently low noise levels (<= 35dBA), qualifying it for operation in noise-sensitive areas in close proximity to people. In addition, it is designed for harsh environments, suitable for use at high temperatures and in mechanically stressed areas. Furthermore long term availability and high security are guaranteed.
As a trusted and experienced OEM partner, Kontron’s board level, module, and system products are used extensively throughout the medical industry for enabling diagnostics, therapy, patient monitoring, home healthcare as well as clinical IT.
So, when it comes to the future of healthcare and medical science, what do you prescribe?
For more information about Kontron’s KISS rackmount server solutions please visit Kontron
{{comment.comment}}