Lab Automation Improves Efficiency and Insight

Computer vision and other types of artificial intelligence accelerate work in the lab.

Laboratory automation is freeing technicians and scientists from time-consuming manual tasks, so they can focus on more important work. Patients can receive their diagnoses fast, getting the timely care they need. New drugs can be tested rapidly, leading to breakthrough treatments. In this lab of the future, artificial intelligence is taking automation to the next level.

Whether running a simple blood test or analyzing the effects of a potential treatment on a cell culture, some of the most important answers in health and life sciences come from the lab. A lab thrives on high accuracy, fast speeds, and high throughput. The more efficiently a lab runs, the faster researchers can make discoveries and clinicians can make diagnoses, accelerating the delivery of world-class care.

Lab automation involves a set of technologies to automate manual, high-volume tasks in clinical or research labs. In a growing number of cases, these technologies involve lab robotics and artificial intelligence (AI), including machine learning, deep learning, and computer vision. Laboratory robotics and automation can be applied to a range of processes and equipment, from benchtop instruments to stand-alone systems to microscopes. Depending on how they’re used, lab automation systems may be single-function or combine many different functions.

Clinical Lab Automation

Automation in a clinical laboratory focuses mainly on ensuring accuracy while accelerating the time and efficiency of diagnostic testing. Clinical labs tend to run around the clock. It’s extremely important for technicians in these labs to manage the large number of tests coming in from one or more hospitals or clinics.

The latest solutions in clinical lab automation use computer vision to read barcodes, identify samples, and help robotic arms make accurate movements. Clinical labs are also exploring the use of machine learning in areas like digital pathology, which requires a high level of compute performance on edge servers.

Research and Pharmaceutical Development

Liquid-handling robots, genomics sequencers, high-content screening (HCS), and high-throughput screening (HTS) are among the lab automation systems helping scientists accelerate research and pharmaceutical development. Researchers can perform an incredibly large number of experiments, which can lead to the discovery of new drugs, cancer therapeutics, and other treatments. Machine learning and deep learning are particularly valuable in research labs, with algorithms that accelerate HCS and other imaging workloads.

For example, to support early drug discovery through HCS acceleration, Intel and Novartis used deep neural networks (DNN) to cut the time to train image analysis models from 11 hours to 31 minutes.1 The team used eight CPU-based servers, a high-speed fabric interconnect, and optimized TensorFlow to process microscopic images significantly faster. This solution helps researchers study the effects of thousands of chemical treatments on different cell cultures and evaluate the potential effectiveness of various drugs.

Benefits of Lab Automation

Automating manual processes in the lab leads to a number of benefits, most notably time savings. But even more important is what’s at stake when tasks are completed faster while maintaining accuracy. For example, when researchers can rapidly run a million compounds against a drug target, they can discover a breakthrough treatment at a speed never before possible.

  • Error reduction. By design, lab automation reduces the possibility of human error by taking manual work out of the process.2 This also supports reproducibility and consistency in testing.
  • Fast turnaround time. Automated systems can perform high-throughput screening and other experiments at a pace not possible when performed by humans, all while maintaining accuracy.2
  • Strategic use of human staff. Lab technicians and scientists can work at the higher end of their skill sets and focus their attention on strategic tasks, rather than being tied up with repetitive work.
  • Cost reduction. Lab automation systems may help lower costs by reducing reagent volumes needed and minimizing waste.
  • Workplace safety. By minimizing the need for human intervention, lab automation can help technicians limit exposure to pathogens and harmful chemicals or injuries caused by repetitive motions.

Intel and Novartis used deep neural networks (DNN) to cut the time to train image analysis models from 11 hours to 31 minutes.¹

Lab Automation Technologies

From robotic arms to image processing, Intel® technologies power the latest lab automation solutions. Our broad portfolio of compute technologies gives instrument manufacturers a range of computing options that meet power and performance requirements, along with software-enabled capabilities for vision and other types of AI.

Additionally, servers and storage built on Intel® technologies provide a strong foundation for data management throughout the lab. This supports the principles of FAIR data—making data findable, accessible, interoperable, and reusable to automated systems without human intervention.

Intel® Technologies for Lab Automation
Intel® Core™ processors and Intel Atom® processors Intel processors deliver the right level of performance and power consumption needed to automate processes in the lab. Ideal for sample handling and retrieval, sorting, centrifugation, and other pre- and postanalytical functions.
Intel® Xeon® Scalable processors Intel® Xeon® Scalable processors deliver high performance for edge servers in the lab, especially useful for high-content screening (HCS) and other types of imaging.
Intel® Movidius™ VPUs Intel® Movidius™ VPUs are designed for computer vision at the edge. These low-power VPUs enable barcode reading, robotic arm movement, sample analysis, and much more.
Intel® Optane™ persistent memory and SSDs Intel® Optane™ persistent memory and solid state drives (SSDs) support large in-memory applications, ideal for imaging and AI workloads in lab automation.
AI Software Tools3 For developers, Intel offers software libraries and optimizations for popular frameworks like TensorFlow and Caffe to boost performance on Intel® architecture. The Intel® Distribution of OpenVINO™ toolkit streamlines the development of vision applications on Intel platforms, including VPUs and CPUs.
Intel® Wi-Fi 6 and Intel 5G With support for the latest Wi-Fi and 5G standards, Intel is streamlining the process of connecting instruments in the lab. High-speed connectivity enables remote control, real-time monitoring, and other edge-to-cloud use cases.

Enabling the Lab of the Future

The Internet of Things has already begun to break down data silos and enable a new level of automation. Microscopic images are processed in real time. Experiment results can be analyzed and shared with labs around the world. Sensor data can be applied to AI algorithms to inform predictive maintenance, which in turn prevents instrument downtime.

Faster processing, storage, and network technologies will continue to enhance the efficiency of the lab of the future. For example, researchers at the Translational Genomics Research Institute (TGen) are sequencing patient genomes, then performing genomics analytics on a high performance computing (HPC) infrastructure powered by Intel® Xeon® Scalable processors. Using modern HPC hardware to perform analytics faster enables genetics counselors and physicians to identify more-timely treatment options. Modern HPC hardware also provides a foundation that empowers researchers to apply machine learning methods to massive amounts of data, revealing insights that can take precision medicine to new heights.

TGen has built a high-performance computing (HPC) infrastructure. Optimized for life sciences, it includes Intel® Xeon® Scalable processors, Intel® Optane™ memory, and Dell rack servers.

As clinical, research, and pharmaceutical laboratories become more connected and automated, Intel will provide a foundation of technology that moves, stores, and processes data efficiently. Whether it’s genomics analytics in the cloud or robotic arms at the edge, Intel® technologies enable intelligence at every step in the automated lab.

Frequently Asked Questions

Lab automation uses robotics, AI, and other technologies to automate manual, high-volume tasks in clinical or research labs.

Automation can accelerate turnaround time and discoveries in both clinical and research labs. This includes labs in hospitals, pharmaceutical and biotech companies, universities, and other research institutions.

Laboratory robotics and automation are powered by a range of hardware and software, sometimes with special capabilities for computer vision or other types of AI.

More Resources in Health and Life Sciences

Find more information on the latest technologies for health providers, healthcare systems, and life sciences professionals.

Health and Life Sciences Technology

Health and life sciences is advancing with Intel® technology-powered medical imaging, data analytics, genomics, telemedicine, and robotics.

Learn more

Accelerating Drug Discovery

Intel collaborated with Novartis to accelerate high content screening, a key element of early drug discovery, using multiscale convolutional neural networks (CNNs).

Read more

Accelerating Genomics Research

Intel and Broad Institute are collaborating on data center solutions to drive genomics analytics and research worldwide.

Explore genomics research

Precision Medicine

Learn about the technologies needed to support precision medicine workloads such as genomic analysis, molecular imaging, and molecular dynamics.

Read about precision medicine

AI in Healthcare and Life Sciences

AI in healthcare and life sciences helps drive efficiency and enhance care in medical imaging, lab automation, and more.

Discover AI in healthcare

Edge Computing

Edge computing places servers, devices, and data processing at many different points, wherever resources are needed most.

Discover edge technologies

Notices and Disclaimers

Intel® technologies may require enabled hardware, software, or service activation.

No product or component can be absolutely secure.

Your costs and results may vary.

Product and Performance Information

120x claim based on 21.7x speedup achieved by scaling from single node system to 8-socket cluster. 8-socket cluster node configuration: CPU: Intel® Xeon® 6148 processor @2.4 GHz, Cores: 40, Sockets: 2, Hyper-threading: Enabled, Memory/node: 192 GB, 2666 MHz,NIC: Intel® Omni-Path Host Fabric Interface (Intel® OP HFI), TensorFlow: v1.7.0, Horovod: 0.12.1, Open MPI: 3.0.0, Cluster: ToR Switch: Intel® Omni-Path Switch. Single node configuration: CPU: Intel® Xeon Phi™ processor 7290F, 192 GB DDR4 RAM, 1x 1.6 TB Intel® SSD DC S3610 Series SC2BX016T4, 1x 480 GB Intel® SSD DC S3520 Series SC2BB480G7, Intel® Math Kernel Library (Intel® MKL) 2017/DAAL/Intel Caffe. *References: BBBC-021: Ljosa V, Sokolnicki KL, Carpenter AE, Annotated high-throughput microscopy image sets for validation, Nature Methods, 2012. ImageNet: Russakovsky O et al, ImageNet Large Scale Visual Recognition Challenge, IJCV, 2015. TensorFlow: Abadi M et al, Large-Scale Machine Learning on Heterogeneous Systems, Software available from tensorflow.org. Tests document performance of components on a particular test, in specific systems. Differences in hardware, software, or configuration will affect actual performance. Consult other sources of information to evaluate performance as you consider your purchase. For more complete information about performance and benchmark results, visit www.intel.com/benchmarks. Intel® technologies’ features and benefits depend on system configuration and may require enabled hardware, software, or service activation. Performance varies depending on system configuration. No computer system can be absolutely secure. Check with your system manufacturer or retailer or learn more at intel.com.
2“Advantages and limitations of total laboratory automation: a personal overview,” Clinical Chemistry and Laboratory Medicine (CCLM), February 2019, degruyter.com/view/journals/cclm/57/6/article-p802.xml.
3

Performance varies by use, configuration and other factors. Learn more at www.Intel.com/PerformanceIndex.