PRESS-NEWS.org - Press Release Distribution
PRESS RELEASES DISTRIBUTION

New AI model draws treasure maps to diagnose disease

The model can accurately identify tumors and diseases in medical images. The tool draws a map to explain each diagnosis, helping doctors follow its line of reasoning, check for accuracy, and explain the results to patients

New AI model draws treasure maps to diagnose disease
2024-03-04
(Press-News.org) Medical diagnostics expert, doctor’s assistant, and cartographer are all fair titles for an artificial intelligence model developed by researchers at the Beckman Institute for Advanced Science and Technology.

Their new model accurately identifies tumors and diseases in medical images and is programmed to explain each diagnosis with a visual map. The tool’s unique transparency allows doctors to easily follow its line of reasoning, double-check for accuracy, and explain the results to patients.

"The idea is to help catch cancer and disease in its earliest stages — like an X on a map — and understand how the decision was made. Our model will help streamline that process and make it easier on doctors and patients alike,” said Sourya Sengupta, the study’s lead author and a graduate research assistant at the Beckman Institute.

This research appeared in IEEE Transactions on Medical Imaging.

Cats and dogs and onions and ogres First conceptualized in the 1950s, artificial intelligence — the concept that computers can learn to adapt, analyze, and problem-solve like humans do — has reached household recognition, due in part to ChatGPT and its extended family of easy-to-use tools.

Machine learning, or ML, is one of many methods researchers use to create artificially intelligent systems. ML is to AI what driver’s education is to a 15-year-old: a controlled, supervised environment to practice decision-making, calibrating to new environments, and rerouting after a mistake or wrong turn.

Deep learning — machine learning’s wiser and worldlier relative — can digest larger quantities of information to make more nuanced decisions. Deep learning models derive their decisive power from the closest computer simulations we have to the human brain: deep neural networks.

These networks — just like humans, onions, and ogres — have layers, which makes them tricky to navigate. The more thickly layered, or nonlinear, a network’s intellectual thicket, the better it performs complex, human-like tasks.

Consider a neural network trained to differentiate between pictures of cats and pictures of dogs. The model learns by reviewing images in each category and filing away their distinguishing features (like size, color, and anatomy) for future reference. Eventually, the model learns to watch out for whiskers and cry Doberman at the first sign of a floppy tongue.

But deep neural networks are not infallible — much like overzealous toddlers, said Sengupta, who studies biomedical imaging in the University of Illinois Urbana-Champaign Department of Electrical and Computer Engineering.

“They get it right sometimes, maybe even most of the time, but it might not always be for the right reasons,” he said. “I’m sure everyone knows a child who saw a brown, four-legged dog once and then thought that every brown, four-legged animal was a dog.”

Sengupta’s gripe? If you ask a toddler how they decided, they will probably tell you.

“But you can’t ask a deep neural network how it arrived at an answer,” he said.

The black box problem Sleek, skilled, and speedy as they may be, deep neural networks struggle to master the seminal skill drilled into high school calculus students: showing their work. This is referred to as the black box problem of artificial intelligence, and it has baffled scientists for years.

On the surface, coaxing a confession from the reluctant network that mistook a Pomeranian for a cat does not seem unbelievably crucial. But the gravity of the black box sharpens as the images in question become more life-altering. For example: X-ray images from a mammogram that may indicate early signs of breast cancer.

The process of decoding medical images looks different in different regions of the world.

“In many developing countries, there is a scarcity of doctors and a long line of patients. AI can be helpful in these scenarios,” Sengupta said.

When time and talents are in high demand, automated medical image screening can be deployed as an assistive tool — in no way replacing the skill and expertise of doctors, Sengupta said. Instead, an AI model can pre-scan medical images and flag those containing something unusual — like a tumor or early sign of disease, called a biomarker — for a doctor’s review. This method saves time and can even improve the performance of the person tasked with reading the scan.

These models work well, but their bedside manner leaves much to be desired when, for example, a patient asks why an AI system flagged an image as containing (or not containing) a tumor.

Historically, researchers have answered questions like this with a slew of tools designed to decipher the black box from the outside in. Unfortunately, the researchers using them are often faced with a similar plight as the unfortunate eavesdropper, leaning against a locked door with an empty glass to their ear.

“It would be so much easier to simply open the door, walk inside the room, and listen to the conversation firsthand,” Sengupta said.

To further complicate the matter, many variations of these interpretation tools exist. This means that any given black box may be interpreted in “plausible but different” ways, Sengupta said.

“And now the question is: which interpretation do you believe?” he said. “There is a chance that your choice will be influenced by your subjective bias, and therein lies the main problem with traditional methods.”

Sengupta’s solution? An entirely new type of AI model that interprets itself every time — that explains each decision instead of blandly reporting the binary of “tumor versus non-tumor,” Sengupta said.

No water glass needed, in other words, because the door has disappeared.

Mapping the model A yogi learning a new posture must practice it repeatedly. An AI model trained to tell cats from dogs studying countless images of both quadrupeds.

An AI model functioning as doctor’s assistant is raised on a diet of thousands of medical images, some with abnormalities and some without. When faced with something never-before-seen, it runs a quick analysis and spits out a number between 0 and 1. If the number is less than .5, the image is not assumed to contain a tumor; a numeral greater than .5 warrants a closer look.

Sengupta’s new AI model mimics this setup with a twist: the model produces a value plus a visual map explaining its decision.

The map — referred to by the researchers as an equivalency map, or E-map for short — is essentially a transformed version of the original X-ray, mammogram, or other medical image medium. Like a paint-by-numbers canvas, each region of the E-map is assigned a number. The greater the value, the more medically interesting the region is for predicting the presence of an anomaly. The model sums up the values to arrive at its final figure, which then informs the diagnosis.

“For example, if the total sum is 1, and you have three values represented on the map — .5, .3, and .2 — a doctor can see exactly which areas on the map contributed more to that conclusion and investigate those more fully,” Sengupta said.

This way, doctors can double-check how well the deep neural network is working — like a teacher checking the work on a student’s math problem — and respond to patients’ questions about the process.

“The result is a more transparent, trustable system between doctor and patient,” Sengupta said.

X marks the spot The researchers trained their model on three different disease diagnosis tasks including more than 20,000 total images.

First, the model reviewed simulated mammograms and learned to flag early signs of tumors. Second, it analyzed optical coherence tomography images of the retina, where it practiced identifying a buildup called Drusen that may be an early sign of macular degeneration. Third, the model studied chest X-rays and learned to detect cardiomegaly, a heart enlargement condition that can lead to disease.

Once the mapmaking model had been trained, the researchers compared its performance to existing black-box AI systems — the ones without a self-interpretation setting. The new model performed comparably to its counterparts in all three categories, with accuracy rates of 77.8% for mammograms, 99.1% for retinal OCT images, and 83% for chest x-rays compared to the existing 77.8%, 99.1%, and 83.33.%

These high accuracy rates are a product of the deep neural network, the non-linear layers of which mimic the nuance of human neurons.

To create such a complicated system, the researchers peeled the proverbial onion and drew inspiration from linear neural networks, which are simpler and easier to interpret.

“The question was: How can we leverage the concepts behind linear models to make non-linear deep neural networks also interpretable like this?” said principal investigator Mark Anastasio, a Beckman Institute researcher and the Donald Biggar Willet Professor and Head of the Illinois Department of Bioengineering. “This work is a classic example of how fundamental ideas can lead to some novel solutions for state-of-the-art AI models.”

The researchers hope that future models will be able to detect and diagnose anomalies all over the body and even differentiate between them.

“I am excited about our tool’s direct benefit to society, not only in terms of improving disease diagnoses, but also improving trust and transparency between doctors and patients,” Anastasio said.

Editor's notes:

The paper associated with this work is titled “A Test Statistic Estimation-based Approach for Establishing Self-interpretable CNN-based Binary Classifiers” and can be accessed online here: https://doi.org/10.1109/TMI.2023.3348699

Mark Anastasio is also a professor in the Illinois Departments of Electrical and Computer Engineering, Computer Science, and Biomedical and Translational Science, and an affiliate of the Coordinated Science Lab. He can be reached at maa@illinois.edu.

Media contact: Jenna Kurtzweil, kurtzwe2@illinois.edu.

END

[Attachments] See images for this press release:
New AI model draws treasure maps to diagnose disease New AI model draws treasure maps to diagnose disease 2 New AI model draws treasure maps to diagnose disease 3

ELSE PRESS RELEASES FROM THIS DATE:

Breastfeeding after COVID-19 booster can give babies antibodies

2024-03-04
Lactating mothers who get the COVID-19 booster pass along the antibodies to their children via their breast milk – and potentially protect babies too young to receive the vaccine, a study from the University of Florida Institute of Food and Agricultural Sciences (UF/IFAS) and the UF College of Medicine found. The study is the third in a series that looks at antibody protection being transferred via breast milk from mothers who received their first two COVID-19 vaccinations and, now, the booster shot. The second publication reported the same antibody transfer via breast milk. “We think that breast milk may play an important ...

Researchers closing in on genetic treatments for hereditary lung disease, vision loss

Researchers closing in on genetic treatments for hereditary lung disease, vision loss
2024-03-04
PORTLAND, Ore. – Researchers who work with tiny drug carriers known as lipid nanoparticles have developed a new type of material capable of reaching the lungs and the eyes, an important step toward genetic therapy for hereditary conditions like cystic fibrosis and inherited vision loss. Findings of the study led by Gaurav Sahay and Yulia Eygeris of the Oregon State University College of Pharmacy and Renee Ryals of Oregon Health & Science University were published today in the Proceedings of the National Academy of Sciences. Unlike other types of lipid nanoparticles that tend to accumulate in the liver, the ones in this study, ...

COVID-19 associated with increased risk for autoimmune inflammatory rheumatic diseases up to a year after infection

2024-03-04
Below please find summaries of new articles that will be published in the next issue of Annals of Internal Medicine. The summaries are not intended to substitute for the full articles as a source of information. This information is under strict embargo and by taking it into possession, media representatives are committing to the terms of the embargo not only on their own behalf, but also on behalf of the organization they represent.    ----------------------------    1. COVID-19 associated with increased risk for autoimmune inflammatory rheumatic diseases up to a year after infection Vaccination showed some protection against AIRDs, depending on severity ...

UC Irvine receives $15 million NSF grant for integrative movement research

2024-03-04
Irvine, Calif. March 4, 2024 — The National Science Foundation has granted $15 million to the Integrative Movement Sciences Institute at the University of California, Irvine. This six-year funding, part of the NSF’s Biology Integration Institutes program, will support groundbreaking research led by Monica Daley, professor of ecology & evolutionary biology at the UCI School of Biological Sciences. The research funded by this grant aims to understand the intricate mechanics of muscle control during rapid, unsteady movements in complex environments. Muscle ...

University of Houston engineer Metin Akay featured in study highlighting 50 scientists' contributions to biomedical engineering advancements

University of Houston engineer Metin Akay featured in study highlighting 50 scientists contributions to biomedical engineering advancements
2024-03-04
Metin Akay, founding chair of the Biomedical Engineering Department at the University of Houston and John S. Dunn Professor, is one of 50 top scientists from 34 elite universities to publish a roadmap for groundbreaking research to transform the landscape of medicine in the coming decade.  Published on behalf of the Institute of Electrical and Electronics Engineers (IEEE), the world’s largest technical professional organization dedicated to advancing technology for humanity and the IEEE Engineering in Medicine and Biology Society (IEEE EMBS), the paper focus ...

JWST captures the end of planet formation

JWST captures the end of planet formation
2024-03-04
March 4, 2024, Mountain View, CA – The James Webb Space Telescope (JWST) is helping scientists uncover how planets form by advancing understanding of their birthplaces and the circumstellar disks surrounding young stars, in a paper published in the Astronomical Journal, a team of scientists led by Naman Bajaj of the University of Arizona and including Dr. Uma Gorti at the SETI Institute, image for the first time, winds from an old planet-forming disk (still very young relative to the Sun) which is actively dispersing its gas ...

Good news—MS drugs taken while breastfeeding may not affect child development

2024-03-04
EMBARGOED FOR RELEASE UNTIL 4 P.M. ET, MONDAY, MARCH 4, 2024 MINNEAPOLIS – Certain medications for multiple sclerosis (MS) called monoclonal antibodies, taken while breastfeeding, may not affect the development of a child during the first three years of life, according to a preliminary study released today, March 4, 2024. The study will be presented at the American Academy of Neurology’s 76th Annual Meeting taking place April 13–18, 2024, in person in Denver and online. The study examined four monoclonal antibodies for MS: natalizumab, ocrelizumab, rituximab and ofatumumab. MS is a disease in which the body’s immune system attacks ...

Programs intended to reduce health insurance premiums may make coverage less affordable for the middle class

2024-03-04
PITTSBURGH, March 4, 2024 — Reinsurance programs, which were created to help lower premiums and increase enrollment in the Affordable Care Act’s health insurance marketplaces, may have had the opposite effects for many potential marketplace enrollees, according to a study by health policy researchers at the University of Pittsburgh School of Public Health, Duke University and University of Minnesota. The study, published today in Health Affairs, is the first to examine the effects of a post-American Rescue Plan Act ...

PrEP discontinuation in a US national cohort of sexual and gender minority populations, 2017–22

2024-03-04
In the U.S., sexual and gender minority populations are disproportionately affected by HIV. Pre-exposure prophylaxis (PrEP) is a key prevention method, but its effectiveness relies on consistent usage. While a significant body of research has addressed PreP initiation and adherence, far less attention has been paid to the reasons for and consequences of PrEP discontinuation. A team of investigators conducted a four-year U.S. national cohort study exploring PrEP discontinuation among sexual and gender minority people who initiated PrEP.  “Our cohort was entirely comprised of individuals ...

USC Study: Medicare Part D plans increased restrictions on drug coverage

2024-03-04
Medicare Part D plans significantly increased restrictions on prescription drugs, excluding more compounds from coverage or subjecting more of them to review before patients could access the treatments, according to a new study from USC researchers. Among drugs not in Medicare “protected classes,” the share of drug compounds restricted or excluded by Part D plans surged from an average of 31.9% in 2011 to 44.4% in 2020, according to the study published in the March 2024 issue of Health Affairs. Brand-name-only compounds (those without a generic alternative) were especially limited, with more ...

LAST 30 PRESS RELEASES:

UK/Portuguese study strongly suggests antibiotic-resistant “superbugs” are being passed from cats and dogs to their owners

Researchers study effects of solvation and ion valency on metallopolymers

Physicists solve puzzle about ancient galaxy found by Webb telescope

Clear guidelines needed for synthetic data to ensure transparency, accountability and fairness study says

Report finds significant gender and racial inequities in the educational measurement profession

University of Houston and Scotland’s Heriot-Watt University forge strategic energy alliance

Rice team demonstrates miniature brain stimulator in humans

Jennifer Stinson receives prestigious Barer-Flood Prize in health services research

First insights into the genetic bottleneck characterizing early sheep husbandry in the Neolithic period

Theories that explain the crisis in democracy are inadequate for Latin America, experts say

Starving cells hijack protein transport stations

Where have all the right whales gone?

Researchers find no link between COVID-19 virus and development of asthma in children

Cell’s ‘garbage disposal’ may have another role: helping neurons near skin sense the environment

Study reveals potential to reverse lung fibrosis using the body’s own healing technique

International team co-led by a BSC researcher discovers more than 50 new deep-sea species in one of the most unexplored areas of the planet

Cleveland Innovation District partners exceeding many targets set by state and JobsOhio

A third of women experience migraines associated with menstruation, most commonly when premenopausal

MD Anderson Research Highlights for April 12, 2024

Soft Robotics appoints new Deputy Editor-in-Chief, Barbara Mazzolai, PhD

Wiley releases Mass Spectra of Designer Drugs 2024 to accelerate forensics analysis of fentanyls, cannabinoids, and more

Freestanding emergency departments are popular, but do they function as intended?

University of Cincinnati experts present at national neurology conference

Bonobos are more aggressive than previously thought

How seaweed became multicellular

Melanomas resist drugs by ‘breaking’ genes

Africa’s iconic flamingos threatened by rising lake levels, study shows

Vaccination timeliness among US children ages 0-19 months

Changes in permanent contraception procedures among young adults following the Dobbs decision

Semaglutide vs endoscopic sleeve gastroplasty for weight loss

[Press-News.org] New AI model draws treasure maps to diagnose disease
The model can accurately identify tumors and diseases in medical images. The tool draws a map to explain each diagnosis, helping doctors follow its line of reasoning, check for accuracy, and explain the results to patients