Vision Systems Applications for In-Vitro Diagnostics & Lab Automation

Machine vision is helping to improve efficiency and throughput in areas from simple diagnostic tests to sophisticated DNA analysis. With the help of vision systems and laboratory automation, the healthcare industry is reducing potential errors and more accurately detecting infection, diagnosing medical conditions, and monitoring drug therapies. Vision systems have improved in-vitro diagnostics (IVD) laboratory processes.

Barcode Reading

Labs must track every sample and make sure that ordered tests link back to the right patient. The process is critical as the labs increase their throughput. To meet demands, labs need higher barcode reading rates. And as labs transition to 2D codes, they need fast, reliable, easy-to-use solutions.

Machine vision systems have improved barcode reading methods. The systems also track the sample tubes and vials. After capturing an image of vials, machine vision algorithms look for specific areas on the image to find the barcode. And thanks to machine learning, the system can even read a poor-quality barcode that could not be scanned otherwise.

Vial Detection and Alignment

IVD systems are using robotic arms for more and more tasks. As samples get smaller, more precise arm control is needed. Machine vision helps to accurately navigate the robots to vials to speed processes and ensure no breakage occurs. Machine vision helps to identify what vial is in each position for the robotic arm to pick up.

Sample Quality

Blood testing analyzers must have accurately prepared samples and test setups. Blood samples are ranked by different indices, such as hemoglobin and bilirubin. These indices can vary based on how the samples are loaded and oriented in the rack. Blood separation labels are crucial to the workflows in highly automated labs.

ROI Calculator

Discover the potential cost savings of robotic automation over a 20-year system life

This calculator compares your current manual labor costs against the total cost of owning and operating a robotic system over its 20-year lifespan.

EXPLORE TODAY

Deep learning-based machine vision can analyze centrifuged blood and determine if it’s been effectively separated into distinct phases. It can then classify the samples for processing. Based on the classifications, it can separate passing and failing samples.

Liquid Levels and Sample Volumes

In IVD lab systems, the liquid volume is important to know to determine if the system is operating correctly and make sure the result is of acceptable quality. Machine vision systems can find the meniscus in a vial when measuring very small amounts. They can also measure the levels of separated phases of blood run through a centrifuge. This confirms an adequate sample for testing and reduces the likelihood of contamination.

Additional Applications

Some additional machine vision system applications include:

• Cap detection. Machine vision identifies the sample and the correct tests to run on each sample. Checking the caps helps to avoid pipette breakage. Machine vision can find out if the cap and vial match and if the caps are installed properly.
• Color detection. Machine vision measures color reliability for chemistry analysis.
• Shape detection and cell counting. Machine vision accurately identifies cell counts and shapes. It can be trained to count blobs, pills, caps, and even a lack of items.