Vision & Imaging Blog
Advanced Microscope Leverages Machine Learning to Improve Malaria Diagnosis
More than 200 million people are infected with malaria every year. Fortunately, researchers have made a breakthrough in medical research imaging thanks to a new microscope outfitted with machine learning. The new microscope could potentially help speed up the diagnosis of one of the world’s most deadly infectious diseases.
If you’re a world traveler, you’ve probably heard of the risk of malaria in areas of the world like sub-Saharan Africa, South Asia, and Central America. This new technology could help save lives the world over and make global travel safer.
How the Microscope Uses Lighting to Find Malaria
Researchers designed this new microscope technology to change lighting angles, colors, and patterns on the fly. But that’s not all. It can learn the best settings to perform complex diagnostics. Thanks to its machine learning capabilities, the microscope can find malaria parasites faster and more accurately than trained physicians.
Most microscopes use light optimized for humans to see. The same amount of light bounces off a sample in all directions. But the new microscope includes lighting options designed for the microscope’s machine learning to examine instead of the human eye. A bowl shape of LEDs allows samples to be imaged with light up to 90-degree angles.
Market Intelligence News & Insights:
How Will Apple’s $500 Billion Investment for US Manufacturing, Education, and AI Play Out?
On February 24th, Apple announced a “commitment… to spend and invest more than $500 billion in the U.S. over the next four years”, with plans for new manufacturing plants, enhanced support for manufacturing partners via their Advanced Manufacturing Fund, and a myriad of other initiatives. As part of this plan, Apple projects 20,000 new hires, with roles focused on R&D, silicon engineering, software development, and AI/ML engineering.
Machine Learning Helps to Find Disease Fast
To train the microscope’s machine learning algorithms, researchers scanned hundreds of samples of malaria-infected red blood cells. A convolutional neural network learned what characteristics were most important when diagnosing malaria. It could then adjust what LED settings were the most effective in finding the disease.
Doctors typically use a type of microscope that has a very limited field of view to find malaria. Scanning a sample slide for signs of malaria can take an expert 10 to 15 minutes. But, the new machine learning microscope learned how to detect malaria in less than ten seconds. This is inspiring news, as many areas don’t have the number of physicians needed to scan every sample.
Not only is this microscope faster, but it’s far more accurate. Conventional microscopes only allow a physician to see a few dozen cells at a time when searching for malaria. Even with its wider field of view that can see a thousand cells at once, the microscope is classifying with 95 - 99% accuracy. That’s better accuracy than most trained physicians and software out there.
Share This On X:
Recent Posts