Prototype microscope teaches itself the best illumination settings for diagnosing malaria
Engineers at Duke University have developed a microscope that adapts its lighting angles, colors and patterns while teaching itself the optimal settings needed to complete a given diagnostic task.
In the initial proof-of-concept study, the microscope simultaneously developed a lighting pattern and classification system that allowed it to quickly identify red blood cells infected by the malaria parasite more accurately than trained physicians and other machine learning approaches.
The results appear online on November 19 in the journal Biomedical Optics Express.
“A standard microscope illuminates a sample with the same amount of light coming from all directions, and that lighting has been optimized for human eyes over hundreds of years,” said Roarke Horstmeyer, assistant professor of biomedical engineering at Duke.
“But computers can see things humans can’t,” Hortmeyer said. “So not only have we redesigned the hardware to provide a diverse range of lighting options, we’ve allowed the microscope to optimize the illumination for itself.”
Rather than diffusing white light from below to evenly illuminate the slide, the engineers developed a bowl-shaped light source with LEDs embedded throughout its surface. This allows samples to be illuminated from different angles up to nearly 90 degrees with different colors, which essentially casts shadows and highlights different features of the sample depending on the pattern of LEDs used.
The researchers then fed the microscope hundreds of samples of malaria-infected red blood cells prepared as thin smears, in which the cell bodies remain whole and are ideally spread out in a single layer on a microscope slide. Using a type of machine learning algorithm called a convolutional neural network, the microscope learned which features of the sample were most important for diagnosing malaria and how best to highlight those features.
The algorithm eventually landed on a ring-shaped LED pattern of different colors coming from relatively high angles. While the resulting images are noisier than a regular microscope image, they highlight the malaria parasite in a bright spot and are correctly classified about 90 percent of the time. Trained physicians and other machine learning algorithms typically perform with about 75 percent accuracy.
“The patterns it’s picking out are ring-like with different colors that are non-uniform and are not necessarily obvious,” said Horstmeyer. “Even though the images are dimmer and noisier than what a clinician would create, the algorithm is saying it’ll live with the noise, it just really wants to get the parasite highlighted to help it make a diagnosis.”
Horstmeyer then sent the LED pattern and sorting algorithm to another collaborator’s lab across the world to see if the results were translatable to different microscope setups. The other laboratory showed similar successes.
“Physicians have to look through a thousand cells to find a single malaria parasite,” said Horstmeyer. “And because they have to zoom in so closely, they can only look at maybe a dozen at a time, and so reading a slide takes about 10 minutes. If they only had to look at a handful of cells that our microscope has already picked out in a matter of seconds, it would greatly speed up the process.”
The researchers also showed that the microscope works well with thick blood smear preparations, in which the red blood cells form a highly non-uniform background and may be broken apart. For this preparation, the machine learning algorithm was successful 99 percent of the time.
According to Horstmeyer, the improved accuracy is expected because the tested thick smears were more heavily stained than the thin smears and exhibited higher contrast. But they also take longer to prepare, and part of the motivation behind the project is to cut down on diagnosis times in low-resource settings where trained physicians are sparse and bottlenecks are the norm.
With this initial success in hand, Horstmeyer is continuing to develop both the microscope and machine learning algorithm.
A group of Duke engineering graduate students has formed a startup company SafineAI to miniaturize the reconfigurable LED microscope concept, which has already earned a $120,000 prize at a local pitch competition.
Meanwhile, Horstmeyer is working with a different machine learning algorithm to create a version of the microscope that can adjust its LED pattern to any specific slide it’s trying to read.
“We’re basically trying to impart some brains into the image acquisition process,” said Horstmeyer. “We want the microscope to use all of its degrees of freedom. So instead of just dumbly taking images, it can play around with the focus and illumination to try to get a better idea of what’s on the slide, just like a human would.”
The Latest on: Machine learning microscope
- Machine Learning Reduces Microscope Data Processing Timeon June 9, 2021 at 1:05 pm
In this new study, recently published in the journal Small Methods, the researchers opted for a new technique to speed up the microscope data processing. This time, they used machine learning ...
- Researchers create quantum microscope that can see the impossibleon June 9, 2021 at 9:37 am
In a major scientific leap, researchers have created a quantum microscope that can reveal biological structures that would otherwise be impossible to see.
- New digital tool makes the analysis of microscopy images much easieron June 9, 2021 at 6:26 am
Information technology can make life easier in many areas - including research. In medicine, for example, it is still standard practice to evaluate microscopy images of tissue sections by hand. This ...
- Faster, Cheaper Diagnosis of Rare Cancer in Fat Cellson June 9, 2021 at 2:30 am
Scientists have created a gene panel test to quickly and more precisely diagnose common forms of liposarcoma. This breakthrough development leverages advan | Clinical And Molecular Dx ...
- Diagnoses with Deepflashon June 8, 2021 at 10:46 pm
Microscopic images of tissue sections can now be analyzed much more easily - with an innovative digital tool. Two researchers from Würzburg have ...
- Machine learning reduces microscope data processing time from months to just secondson June 8, 2021 at 12:20 pm
Ever since the world's first ever microscope was invented in 1590 by Hans and Zacharias Janssen—a Dutch father and son—our curiosity for what goes on at the tiniest scales has led to development of ...
- Winner: 2021 Harrison-Meldola Memorial Prizeon June 8, 2021 at 5:15 am
As computational chemists, Professor Duarte’s team use computers as 'molecular microscopes' to understand how reactions ... combining classical, quantum and machine-learning methods. Aside from ...
- Google and Harvard Unveil the Largest High-Resolution Map of the Brain Yeton June 6, 2021 at 7:00 am
Last Tuesday, teams from Google and Harvard published an intricate map of every cell and connection in a cubic millimeter of the human brain.
- Future of chip making to lean heavily on AI for spotting defects, says Applied Materialson May 31, 2021 at 1:01 pm
For the first time, a machine learning program will tell a scanning electron microscope where to look closer for defects in a silicon wafer ...
via Bing News
The Latest on: Smart microscope
- How we review internet service providerson June 10, 2021 at 12:56 pm
I've been reviewing tech for CNET for nearly eight years now, and I've never had an assignment quite like writing reviews of internet service providers. From refrigerators , waffle makers and toasters ...
- Microscope reveals the secrets of a material's structureon June 9, 2021 at 5:48 am
EPFL scientists have made an important discovery about the structure of barium titanate, a material used in everyday objects. Their findings refute existing theories on the displacement of the ...
- Microscope reveals secrets of a material’s structureon June 8, 2021 at 11:16 pm
Themis" microscope © Alain Herzog / 2021 EPFL EPFL scientists have made an important discovery about the structure ...
- Using Your Phone As A Microscope On The Electronics Workbenchon June 5, 2021 at 5:00 pm
I’m guessing that almost any smartphone made over the last few ... so why does a real 2D LCD microscope not have this problem? I am theorising when I say this, but I suspect that the solution ...
- Best iPhone 7 cases: how to protect your older iPhoneon June 4, 2021 at 6:10 am
Options for those who like to bundle their cards and cash with their smartphone in a case tend ... whose job it is to look at stuff under microscopes, so the images you see on these cases are ...
- 3D printing tiny parts for big impacton June 3, 2021 at 1:52 am
(Nanowerk News) Whether it’s computer chips, smartphone components ... “The process shares a lot of similarities with regular microscopes, except we’re delivering a digital picture instead of ...
- Engineers Have Created a Simple Slide Coating That Can Impressively Boost the Resolution of a Microscopeon June 2, 2021 at 10:25 pm
A light-powered microscope has a resolution limit of around 200 nanometres — which makes observing specimens smaller or closer together than that all but impossible. Engineers at the University ...
- Researchers find new technology that lets ordinary microscope see in super resolutionon June 1, 2021 at 4:56 am
A technology that improves the resolution of an ordinary light microscope so that it can be used to directly observe finer structures and details in living cells has been developed by a team of ...
- Why Advanced Human and other ASX imaging stocks have more than doubled in 12 monthson May 31, 2021 at 1:47 pm
Adanvanced Human Imaging (AHI) and other ASX imaging stocks such as Telix (TLX), Micro-X (MX1), and Optiscan (OIL) have doubled in 12 months.
- Smart Scale Market Present And Upcoming Trends, Leading Players - Fitbit, Withings, Blipcareon May 28, 2021 at 5:31 am
The report elaborates the essential market components that are important to strategic planning for the new players involved in the Smart Scale market. The report offers a critical account of the Smart ...
via Bing News