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
- Behold the xenobots – part frog, part robot. But are they alive?on January 31, 2020 at 10:43 am
At first, they are just blurry blobs. But when Douglas Blackiston turns a knob on the microscope, they suddenly come to life in sharp relief. They look like miniature, lumpy matzo balls suspended in ...
- New Machine-Learning Method Independently Determines Crystal Structureon January 31, 2020 at 8:49 am
This method involves using a machine-learning algorithm, just as the one used in self-driving cars and facial ... In this method, a scanning electron microscope (SEM) is used to obtain electron ...
- Machine learning technique speeds up crystal structure determinationon January 30, 2020 at 11:09 am
The method uses a machine learning algorithm, similar to the type used in facial ... Their method involves using a scanning electron microscope (SEM) to collect electron backscatter diffraction (EBSD) ...
- Opening the AI box: can deep learning predict cancer recurrence?on January 29, 2020 at 2:13 am
The extracted samples of tissue are examined under a microscope and, if cancerous cells are found ... Even though automated machine learning tools capable of accurately classifying histopathology ...
- Largest Brain Wiring Diagram to Date Is Publishedon January 22, 2020 at 10:35 am
Electron microscope image shows neurons in the fruit fly brain ... That last step was carried out by combining machine learning algorithms from Google with a team of proofreaders at Janelia. According ...
- Scanning Raman picoscopy: A new methodology for determining molecular chemical structureon January 21, 2020 at 6:40 am
Scanning tunneling microscope and atomic force microscope have outstanding ... and can become a more powerful tool with the aid of imaging recognition and machine learning techniques. The ability of ...
- Future-proofing under the microscope as global security industry converges at Intersec 2020on January 19, 2020 at 2:47 am
The emergence of automation, biometrics, artificial intelligence and machine learning is transforming the face of the ... to address critical threats are all key topics under the microscope. Held ...
- New approach using deep learning could speed up brain image microscopy by 16 timeson January 8, 2020 at 3:19 pm
They believed that deep learning -- a form of machine-learning that uses multiple ... be very useful for increasing the resolution of microscope images, a process called super-resolution.
- Using Machine Learning to Analyze Gene Regulationon December 31, 2019 at 11:44 am
Massive amounts of data has been generated with powerful microscopes, and by applying techniques like high-throughput robotics or advanced genetic sequencing technologies. Machine learning has been ...
- Technology 2020. Algorithms in the cloud, food from printers and microscopes in our bodieson December 18, 2019 at 5:04 pm
Machine learning, natural language processing ... Swallowed capsules containing miniature microscopes will offer insights into our bodies. The data sourced in this way will go to digital files ...
via Bing News
The Latest on: Smart microscope
- Jayson Tatum's elite defense a major factor in his growthon January 31, 2020 at 4:43 pm
I hope I haven't offended you': The potential of Jayson Tatum https://t.co/iauFHjHj9d pic.twitter.com/qA3pIrNVqA The Duke product tried to hide within himself in his seat while the laser-focus of ...
- Artificial lifeforms designed by supercomputers are fully programmableon January 31, 2020 at 2:25 pm
Robots are made to mimic living creatures, and as smart as they’re becoming, we can still look at them and understand ... They gathered frog stem cells, isolated and incubated them, then cut and ...
- Offensive and defensive players to watch in Chiefs vs. 49erson January 31, 2020 at 3:43 am
As the Kansas City Chiefs get set to square off against the San Francisco 49ers in Super Bowl LIV, each player will be under the microscope. Kansas City will ... Mahomes will have to walk a fine line ...
- 4K Microscope To Expand From Hamilton County Schools To Other Gigabit Communitieson January 30, 2020 at 8:50 am
The expansion of the 4K Microscope program will be possible because of a $30,000 grant that US Ignite recently awarded to Chattanooga through its National Science Foundation-funded Smart Gigabit ...
- Novel microscope views ultrafast processes at quantum scaleon January 30, 2020 at 8:37 am
To be able to video the component-level functioning of next-gen computers or smart phones, researchers need a high-speed ... laser pulses in conjunction with a scanning tunnelling microscope (STM).
- Household dust is being put under the microscope as scientists seek to better understand iton January 29, 2020 at 10:12 am
"If you then dump that on your veggie patch and start growing your veggies in there, you're not being smart with producing clean green vegetables. It smothers them. "Put it in the waste bin, outside.
- Advantest to Showcase Latest Semiconductor Test Solutions for 5G, Automotive and IoT Applications at SEMICON Korea 2020on January 27, 2020 at 11:05 pm
Leading semiconductor test equipment supplier Advantest Corporation (6857) will feature its newest test solutions for advanced ICs at SEMICON Korea on February 5-7 at COEX in Seoul, South Korea. In ...
- High-Performance, Aqueous Lithium-Ion Battery Is Safer Than Current Designson January 27, 2020 at 2:12 am
A microscope image of niobium tungsten oxide particles that make up an anode in a ... The team envisions their battery design well-suited for applications in which energy needs to be stored in compact ...
- Global Automated Microscopy Market 2020-2024| Evolving Opportunities with Bruker Corp. and Carl Zeiss AG | Technavioon January 24, 2020 at 12:20 pm
Smart Life & Ecofriendly Systems, and Others. The company offers Transmission Electron Microscope HT7700, Scanning Electron Microscopes SU3800/SU3900, and Scanning Electron Microscope FlexSEM 1000 / ...
- Lafayette wins award to promote replication of successful smart, connected community applicationson January 24, 2020 at 10:58 am
Chattanooga, Tennessee In collaboration with Hamilton County Schools, the city of Chattanooga will expand a program using 4K microscopes and remote instrumentation capabilities ... and the local ...
via Bing News