Novel AI model explains retinal sex difference

reposted from https://www.vchri.ca/stories/2024/03/20/novel-ai-model-explains-retinal-sex-difference Novel AI model explains retinal sex difference Stories Mar 20, 2024 3 minutes The approach could lead to further biomarker discoveries to assist in disease diagnostics and early treatment interventions. Artificial intelligence (AI) applications are revolutionizing health diagnostics and precision medicine, yet how they identify patterns of disease in layers of data has largely remained a mystery. Research led by Vancouver Coastal Health Research Institute researcher Dr. Ipek Oruç revealed for the first time AI behaviours that differentiated between female and male retinal images, paving the way for other novel biomarkers of disease to aid in precision care. Dr. Ipek Oruc is an associate professor in the Department of Ophthalmology and Visual Sciences; associate member of the School of Biomedical Engineering; investigator with the Data Science Institute; and director of the NOVA Lab at the University of British Columbia. She is also a principal investigator with ICORD. An expert in visual neuroscience and artificial intelligence applications of retinal image analysis, Oruç’s PNAS Nexus study charts a path for other research teams to peek under the hood of AI algorithms to understand how the algorithm was developed in order to improve patient diagnostics and outcomes using medical imaging. “Our research opens the black box of AI, setting the stage for future research to apply this methodology to leverage its potential and identify previously unseen characteristics of various conditions.” Oruç and her team’s proof-of-concept study examined a convolutional neural network (CNN) model trained to classify patient sex in retinal images. CNNs are a type of deep neural network AI that can mimic human cognition in image processing and classification at superhuman speeds, processing thousands of images in a matter of seconds. The technology is already applied to a number of different AI algorithms using medical imaging for the detection and classification of conditions such as cancers and heart disease. The CNN used in Oruç’s study was trained to detect male and female retinal scans taken using a specialized fundus camera, a mainstay tool of optometry and ophthalmology used to capture images of the retina and other eye features. Ophthalmologists and nonexperts better able to detect retinal sex difference The research team developed and applied a novel methodology to discover for the first time retinal features that differentiate between male and female eyes. They tested 14 exploratory research questions that were derived from the behaviours and decisions of the CNN in what investigators termed the “Inspiration” phase of the model. Nine of the hypotheses revealed significant findings, five of which were verified. These biomarkers of sex difference in retinal images included greater retinal vascularization and a darker ring around the optic disk region in male retinas as compared to female retinas. These two images show subtle differences between male and female retinal images. The male image has a darker area around the optic disk, which appears as a bright spot in the images here, as well as more vasculature and nodes. Researchers then shared these distinguishing features with 26 expert ophthalmologists and 31 nonexperts. Prior to receiving this information, both groups’ ability to detect sex difference in retinal images averaged 50 per cent. After receiving training on the distinguishing characteristics between female and male retinal images discovered through Oruç’s novel methodology, both groups were able to identify the sex of the retinal images with approximately 66 per cent accuracy in the post-training block. This significant improvement is still a ways away from a 100 per cent detection rate, which indicates that additional differences have yet to be discovered to enable greater accuracy, Oruç says. "Our findings showcase an opportunity for biomarker discovery through CNN applications, with the added benefit of equipping medical practitioners with new diagnostic options that can be added to their clinical toolkit.” “We are now investigating other biomarkers present in fundus images that can be identified from CNNs, such as whether changes in the eye might signal a risk of stroke or dementia.” Oruç is also currently working on a study to apply this novel methodology to identify biomarkers in women that could lead to improvements in disease detection and treatment.

Eyes, Nose as Windows to Alzheimer’s

reposted from

Eyes, Nose as Windows to Alzheimer’s

Failing a sniff test or screening positive on an eye exam may predict people’s chances of developing the neurodegenerative disorder.

By Kerry Grens | July 14, 2014

• Comment
• Print

• Link this
• Stumble
• Tweet this

WIKIMEDIA, ADAMANTIOSA suite of studies presented at the ongoing Alzheimer's Association International Conference in Copenhagen offers a variety of ways to diagnose the neurodegenerative disease at early stages. In some cases, researchers worked out the chances of developing the disease among people who can’t smell certain odors, and in other studies scientists tested the utility of Alzheimer’s biomarkers in the eye.

“We envision a future where we can predict risk and then do things to lower risk,” Matthew Growdon, a medical student at Harvard Medical School and the Harvard School of Public Health who is working on a such a smell test, told theWashington Post.

Alzheimer’s sniff tests have been proposed before. The idea is that a poorer performance for odor detection could predict cognitive decline. “A loss of sense of smell does not mean you have Alzheimer's disease,” Kenneth Heilman, a professor in the department of neurology at the University of Florida College of Medicine in Gainesville, told HealthDay. “But if someone has episodic memory loss and also has a loss of smell, a degenerative disease like Parkinson’s or Alzheimer’s is a possibility.”

Just north of the nose, the eyes may also be windows into one’s cognitive state. Researchers reported at the conference two methods to detect beta amyloid protein using fluorescent tags either in the lens or the retina of the eye. According to The Guardian, “both tests showed that levels of beta-amyloid in the eye mirrored those seen in the brain by PET [positron emission tomography] imaging.”

Shaun Frost from Australia’s Commonwealth Scientific and Industrial Research Organisation who led one of the eye biomarker studies, told the Guardian, “if further research shows that our initial findings are correct, it could potentially be delivered as part of an individual's regular eye check-up. The high resolution level of our images could also allow accurate monitoring of individual retinal plaques as a possible method to follow progression and response to therapy.”

Such tests are not likely to be perfectly accurate, USA Today pointed out, but they may help determine who needs more definitive—and costly—testing. “Everyone with Alzheimer’s has beta amyloid in their brain, but not everyone with beta amyloid buildup will develop Alzheimer’s, limiting the predictive ability of some of these tests.”

Tags

smell, olfaction, diagnostics, biomarkers, amyloid and Alzheimer's Disease

Extra Eyeballs on the Eye

reposted from

Extra Eyeballs on the Eye

A legion of citizen-scientist gamers helps a team of researchers explain a long-standing riddle of how the retina processes motion.

By Rina Shaikh-Lesko | May 6, 2014

• Comment
• Print

• Link this
• Stumble
• Tweet this

WIKIMEDIA, AMY LEE ROBINSONResearchers have solved a 50-year mystery of how the retina processes motion with the aid of citizen scientists playing an online game, according to a study published this week (May 4) in Nature. A team led by Sebastian Seung of MIT found that two bipolar neurons in the retina firing together activate a third neuron, a starburst amacrine cell, which is wired to the brain.

Initially, scientists thought the eye passed all visual information for processing up to the brain. Since the mid-1960s, though, researchers have realized that the retina is sensitive to direction and speed.

Seung is the creator of a neuron-mapping game called EyeWire. More than 120,000 gamers from 140 countries helped Seung’s team to map and color-code individual neurons and their connections in the retina. The researchers used retinal neuron wiring diagram created by the players and pieced together by Seung’s team to understand what was going on inside the retina.

“This is a very nice paper that poses a very clear and testable prediction about direction-selective computation in the retina,” neuroscientist Botond Roska from the Friedrich Miescher Institute for Biomedical Research in Basel, Switzerland, who was not involved in the study, told Nature News. “It’s an exciting idea, and I bet it’ll be followed by research from many labs trying test this hypothesis.”

Seung noted that the wiring diagram depicts a small fraction of the total neural connections in the retina. “There are probably other neurons that are a part of this motion-detection circuit,” he told Nature. “We need to map those out and eventually reconstruct the entire retinal connectome.”

reposted from SfN2013

PUBLIC RELEASE DATE:

13-Nov-2013

[ Print | E-mail ]  Share

 [ Close Window ] 

Contact: Karen Mallet
km463@georgetown.edu
Georgetown University Medical Center 

Can the eyes help diagnose Alzheimer's disease?

		VIDEO: Scott Turner, M.D., Ph.D., discusses research to be presented at the Neuroscience 2013 meeting. He and an international team of researchers are studying the link between vision loss and Alzheimer's... Click here for more information.

SAN DIEGO — An international team of researchers studying the link between vision loss and Alzheimer's disease report that the loss of a particular layer of retinal cells not previously investigated may reveal the disease's presence and provide a new way to track disease progression.

The researchers, from Georgetown University Medical Center (GUMC) and the University of Hong Kong, examined retinas from the eyes of mice genetically engineered to develop Alzheimer's disease (AD). They presented their findings today at Neuroscience 2013, the annual meeting of the Society for Neuroscience.

"The retina is an extension of the brain so it makes sense to see if the same pathologic processes found in an Alzheimer's brain are also found in the eye," explains R. Scott Turner, MD, PhD, director of the Memory Disorders Program at GUMC and the only U.S. author on the study. "We know there's an association between glaucoma and Alzheimer's in that both are characterized by loss of neurons, but the mechanisms are not clear."

Turner says many researchers increasingly view glaucoma as a neurodegenerative disorder similar to AD.

Most of the research to date examining the relationship between glaucoma and Alzheimer's focused on the retinal ganglion cell layer, which transmits visual information via the optic nerve into the brain. Before that transmission happens, though, the retinal ganglion cells receive information from another layer in the retina called the inner nuclear layer.

In their study, the researchers looked at the thickness of the retina, including the inner nuclear layer (not previously study in this setting) and the retinal ganglion cell layer. They found a significant loss of thickness in both. The inner nuclear layer had a 37 percent loss of neurons and the retinal ganglion cell layer a 49 percent loss, compared with healthy, age-matched control mice.

In humans, the structure and thickness of the retina can be readily measured using optical coherence tomography. Turner says this new tool is increasing finding applications in research and clinical care.

"This study suggests another path forward in understanding the disease process and could lead to new ways to diagnose or predict Alzheimer's that could be as simple as looking into the eyes," Turner says. "Parallel disease mechanisms suggest that new treatments developed for Alzheimer's may also be useful for glaucoma."

###

Support for this research comes from the Hong Kong University Alzheimer's Disease Research Network and from Ms. Kit Wan Chow.

The lead author of the research is R. C. Chang from the University of Hong Kong. Other authors include K. Chiu, C. K. M. Lok, Y. Matsuoka and K. F. So, all of the University of Hong Kong. The authors report having no personal financial interests related to the study.

About Georgetown University Medical Center

Georgetown University Medical Center (GUMC) is an internationally recognized academic medical center with a three-part mission of research, teaching and patient care (through MedStar Health). GUMC's mission is carried out with a strong emphasis on public service and a dedication to the Catholic, Jesuit principle of cura personalis -- or "care of the whole person." The Medical Center includes the School of Medicine and the School of Nursing & Health Studies, both nationally ranked; Georgetown Lombardi Comprehensive Cancer Center, designated as a comprehensive cancer center by the National Cancer Institute; and the Biomedical Graduate Research Organization, which accounts for the majority of externally funded research at GUMC including a Clinical and Translational Science Award from the National Institutes of Health.

---

[ Print | E-mail  Share ] [ Close Window ] 

AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert! system.

HOME   DISCLAIMER   PRIVACY POLICY   TERMS & CONDITIONS  CONTACT US   TOP Copyright ©2013 by AAAS, the science society.

reposted from:

http://www.eurekalert.org/pub_releases/2013-09/meae-rus092413.php

Public release date: 24-Sep-2013
[ Print | E-mail |  Share

] [ Close Window ] 

Contact: Mary Leach
Mary_Leach@meei.harvard.edu
Massachusetts Eye and Ear Infirmary 

Researchers use smart phone photography to diagnose eye disease

Cheap, common device can aid in telemedicine

BOSTON (Sept. 24, 2013) – Retinal (or fundus) photography is an essential part of any ophthalmology practice. Commercial fundus cameras can cost tens to hundreds of thousands of dollars, making the technology out of reach for smaller ophthalmic practices and to physicians in third-world countries. In a recent study now on line, Massachusetts Eye and Ear researchers describe the relatively simple technique of fundus photography in human and rabbit eyes using a smartphone, an inexpensive app for the smartphone, and instruments that are readily available in an ophthalmic practice.

Smartphones are now being used more routinely in ophthalmology to document patients' ocular conditions, the authors write. Previously described techniques of fundus imaging often proved difficult to repeat, partly because video capture using Apple's built-in camera app in the iPhones cannot independently control the focus and the exposure during filming, which results in glare and poor image quality.

"Our technique provides a simpler and higher quality method to more consistently produce excellent images of a patient's fundus," said senior author Shizuo Mukai, M.D., Mass. Eye and Ear retina specialist and Harvard Medical School associate professor of Ophthalmology. "This technique has been extremely helpful for us in the emergency department setting, in-patient consultations, and during examinations under anesthesia as it provides a cheaper and portable option for high-quality fundus-image acquisition for documentation and consultation. This technique is well tolerated in awake patients most likely since the light intensity used is often well below that which is used in standard indirect ophthalmoscopy."

Using the described technique of smartphone fundus photography with the use of iPhone 4 or iPhone 5, the app Filmic pro, and a 20D lens with or without a Koeppe lens, researchers were was able to capture excellent, high-quality fundus images in both children under anesthesia and in awake adults.

The best results were achieved in the operating room when a Koeppe lens was used in addition to the 20D lens; however, excellent images were acquired with the 20D lens alone in the clinic and emergency room setting as well as in the operating room. Researchers report that even first-year ophthalmology residents were able to master this technique in a relatively short period.

"This technique is relatively inexpensive and simple to master, and takes advantage of the expanding mobile-telephone networks for telemedicine," Dr. Mukai said. "We expect that the quality of the images achieved using this technique will continue to improve as higher-resolution cameras with larger sensors and better image stabilization is incorporated into newer smartphones."

###

In addition to Dr. Mukai, authors of the paper are Luis J. Haddock and David Y. Kim. This study was funded in part by gifts to the Mukai Fund at the Massachusetts Eye and Ear Infirmary.

About Massachusetts Eye and Ear

Mass. Eye and Ear clinicians and scientists are driven by a mission to find cures for blindness, deafness and diseases of the head and neck. After uniting with Schepens Eye Research Institute in 2011, Mass. Eye and Ear in Boston became the world's largest vision and hearing research center, offering hope and healing to patients everywhere through discovery and innovation. Mass. Eye and Ear is a Harvard Medical School teaching hospital and trains future medical leaders in ophthalmology and otolaryngology, through residency as well as clinical and research fellowships. Internationally acclaimed since its founding in 1824, Mass. Eye and Ear employs full-time, board-certified physicians who offer high-quality and affordable specialty care that ranges from the routine to the very complex. U.S. News & World Report's "Best Hospitals Survey" has consistently ranked the Mass. Eye and Ear Departments of Otolaryngology and Ophthalmology as among the top hospitals in the nation.

URL:http://www.hindawi.com/journals/jop/2013/518479/

Findings published in Journal of Ophthalmologyonline.

---

[ Print | E-mail |  Share] [ Close Window ] 

 

AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert! system.

HOME   DISCLAIMER   PRIVACY POLICY   TERMS & CONDITIONS   CONTACT US   TOP Copyright ©2013 by AAAS, the science society.

For paper and images see

http://www.hindawi.com/journals/jop/2013/518479/

Scientists Make Retina Wiring Diagram

• • reposted from 
  • http://www.the-scientist.com//?articles.view/articleNo/36925/title/Scientists-Make-Retina-Wiring-Diagram/
  • 
    
  • 
    
  • 
    
  • The Scientist »
  • The Nutshell

Scientists Make Retina Wiring Diagram

Researchers have created a neural wiring diagram of a chunk of mouse retina by analyzing thousands of electron microscopy images.

By Chris Palmer | August 7, 2013

• 0 Comments
• Print

• Link this
• Stumble
• Tweet this

Photomicrograph of axons sprouting out of mouse embryo retinal neurons that were cultured in a Petri dish.Anna Guzik-Kornacka, University of Zurich

The human brain has about 100 billion neurons, with more than 100 trillion connections, or synapses, among them. A long-term goal of neuroscience research is to create a wiring diagram of each of the brain’s neurons and its connections. Knowing all of the connections could enable a greater understanding of the function of the various types of neurons, as well as of individual brain areas. Scientists at the Massachusetts Institute of Technology (MIT) and the Max Planck Institute for Medical Research have taken an important step toward this goal by creating a complete neural wiring diagram of a small piece of the mouse retina. Their findings were published today (August 7) in Nature.

“It’s the complete reconstruction of all the neurons inside this [area],” said study coauthor Sebastian Seung, a computational neuroscientist at MIT, in a statement. “No one’s ever done that before in the mammalian nervous system.”

The researchers used an innovative imaging technique called serial block-face scanning electron microscopy developed by the study’s senior author, Winfried Denk, a neuroscientist at Max Planck. The technique involves slicing 25-nanometer-thick sections of tissue off a chunk of the retina and imaging them one at a time. In total, the researchers imaged 3,200 such sections from a 200-micrometer by 300-micrometer portion of the mouse retina called the inner plexiform layer—a processing structure sandwiched between photoreceptors, which receive visual input, and ganglion cells, which send out visual information to the brain.

The retina sections were then reconstructed into a complete three-dimensional wiring diagram of the 950 neurons in the tissue sample. First, 225 German undergraduate students spent 2 years, and a total of 20,000 hours, tracing skeleton structures of every neuron. Then, using a computer algorithm developed by Seung, the team analyzed the images and filled out the fine details of the skeletons.

With the fleshed-out wiring diagram in hand, the researchers were then able to classify each of the neurons into five basic types: the aforementioned photoreceptors and ganglion cells, as well as horizontal cells, bipolar cells, and amacrine cells. The exhaustive survey uncovered a new type of bipolar cell, though its function remains unknown. The analysis of neural connections also suggests that a particular type of ganglion cell is motion-sensitive.

“We haven’t completed the project of classifying types, but this shows that it should be possible,” said Seung in the statement. “This method should be able to do it, in principle, if it’s scaled up to a larger piece of tissue.”

Seung is now working on mapping a larger section of retina—0.3 millimeter by 0.3 millimeter. He plans to first analyze the electron microscopy images of the section with his computer algorithm, then have human volunteers check over the computer’s work through a crowdsourced project called EyeWire.

Tags

wiring diagram, synapse, retina, mouse study, electron microscopy and brain