(Press-News.org) Mass General Brigham investigators have developed a robust new artificial intelligence (AI) foundation model that is capable of analyzing brain MRI datasets to perform numerous medical tasks, including identifying brain age, predicting dementia risk, detecting brain tumor mutations and predicting brain cancer survival. The tool. known as BrainIAC, outperformed other, more task-specific AI models and was especially efficient when limited training data were available. Results are published in Nature Neuroscience.
“BrainIAC has the potential to accelerate biomarker discovery, enhance diagnostic tools and speed the adoption of AI in clinical practice,” said corresponding author Benjamin Kann, MD, of the Artificial Intelligence in Medicine (AIM) Program at Mass General Brigham. “Integrating BrainIAC into imaging protocols could help clinicians better personalize and improve patient care.”
Despite recent advances in medical AI approaches, there is a lack of publicly available models that focus on broad, brain MRI analysis. Most conventional frameworks perform specific tasks and require extensive training with large, annotated datasets that can be hard to obtain. Furthermore, brain MRI images from different institutions can vary in appearance and based on their intended applications (such as in neurology versus oncology care), making it challenging for AI frameworks to learn similar information from them.
To address these limitations, the research team designed a brain imaging adaptive core, or BrainIAC. The tool uses a method called self-supervised learning to identify inherent features from unlabeled datasets, which can then be adapted to a range of applications. After pretraining the framework on multiple brain MRI imaging datasets, the researchers validated its performance on 48,965 diverse brain MRI scans across seven distinct tasks of varying clinical complexity.
They found that BrainIAC could successfully generalize its learnings across healthy and abnormal images and subsequently apply them to both relatively straightforward tasks, such as classifying MRI scan types, and very challenging tasks, such as detecting brain tumor mutation types. The model also outperformed three more conventional, task-specific AI frameworks at these applications and others.
The authors note that BrainIAC was especially good at predicting outcomes when training data was scarce or task complexity was high, suggesting that the model could adapt well to real-world settings where annotated medical datasets are not always readily available. Further research is needed to test this framework on additional brain imaging methods and larger datasets.
Authorship: In addition to Kann, Mass General Brigham authors include Divyanshu Tak, Biniam A. Garomsa, Anna Zapaishchykova, Tafadzwa L. Chaunzwa, Juan Carlos Climent Pardo, Zezhong Ye, John Zielke, Yashwanth Ravipati, Suraj Pai, Omar Arnaout, Hugo JWL Aerts, and Raymond Y. Huang. Additional authors include Sri Vajapeyam, Maryam Mahootiha, Mitchell Parker, Luke R. G. Pike, Ceilidh Smith, Ariana M. Familiar, Kevin X. Liu, Sanjay Prabhu, Pratiti Bandopadhayay, Ali Nabavizadeh, Sabine Mueller, and Tina Y. Poussaint.
Disclosures: None.
Funding: This study was supported in part by the National Institute of Health/ the National Cancer Institute (NIH/NCI) (U54 CA274516 and P50 CA165962), and Botha-Chan Low Grade Glioma Consortium. Authors acknowledge the Children’s Brain Tumor Network (CBTN) for the imaging and clinical data access, and ASCO Conquer Cancer Foundation: 2022A013157 Radiation Oncology Institute: ROI2022-9151.
Paper cited: Tak D et al. “A foundation model for generalized brain MRI analysis” Nature Neuroscience DOI: 10.1038/s41593-026-02202-6
###
About Mass General Brigham
Mass General Brigham is an integrated academic health care system, uniting great minds to solve the hardest problems in medicine for our communities and the world. Mass General Brigham connects a full continuum of care across a system of academic medical centers, community and specialty hospitals, a health insurance plan, physician networks, community health centers, home care, and long-term care services. Mass General Brigham is a nonprofit organization committed to patient care, research, teaching, and service to the community. In addition, Mass General Brigham is one of the nation’s leading biomedical research organizations with several Harvard Medical School teaching hospitals. For more information, please visit massgeneralbrigham.org.
END
A new U of T Scarborough study finds that being mentally sharp can translate into a productivity boost equivalent to about 40 extra minutes of work each day.
The study, published in the journal Science Advances, followed participants over a 12-week period and found that day-to-day fluctuations in mental sharpness helped explain why people sometimes fail to follow through on their goals. On days when participants were mentally sharp, they were more likely to set goals and complete them, whether it was finishing assignments or even just cooking ...
COVID-19 vaccination is not the cause behind a decrease in childbirth, according to a study from Linköping University, Sweden. The results speak against rumours about vaccination and reduced fertility. The findings have been published in the journal Communications Medicine.
“Our conclusion is that it’s highly unlikely that the mRNA vaccine against COVID-19 was behind the decrease in childbirth during the pandemic,” says Toomas Timpka, professor of social medicine at Linköping ...
Key takeaways
New research published in the European Heart Journal - Digital Health shows that an AI-enabled digital stethoscope more than doubles sensitivity for detecting moderate to severe valvular heart disease in a real-world clinical setting compared to a traditional stethoscope.
A total of 357 patients of 50 years old or over were examined with both the traditional and the AI-enabled digital stethoscope and the results were compared.
The AI-stethoscope demonstrated significantly higher sensitivity in detecting the heart sound patterns that indicate valvular heart disease, with 92.3% sensitivity compared with 46.2% with the traditional stethoscope.
Valvular ...
Insulated joints are little known, but many railway lines could not be used without them. They divide the rail network into electrically separated sections and register when a train enters and leaves a section. Only when the section is free again the next train is allowed to enter. Around 33,000 insulated joints are currently installed in Austria, but they wear out quickly on heavily used lines. Together with ÖBB and Martin Schienentechnik, Graz University of Technology (TU Graz) has now developed a prototype for significantly more robust insulated joints using improved materials and new geometries. According to current research findings, ...
When startups scale quickly, founders often make hurried hiring decisions that unintentionally disadvantage women, according to new study from the Stockholm School of Economics in Sweden. The study shows how the pressures of rapid growth increase the likelihood that founders rely on mental shortcuts and make biased decisions.
Drawing on large‑scale Swedish data, the study shows that scaling—when companies hire far more people than their usual growth trend would predict—puts pressure on founders to decide swiftly, which increases the use of mental shortcuts. These ...
A supermassive black hole with a case of cosmic indigestion has been burping out the remains of a shredded star for four years — and it’s still going strong, new research led by a University of Oregon astrophysicist shows.
Already, the jet shooting out of the black hole is a contender for one of the brightest, most energetic things ever detected in the universe. Scientists have now collected enough data on the unusual occurrence to predict that the stream of radio waves belching from the black hole will keep increasing exponentially before peaking in 2027.
“This is really unusual,” said Yvette Cendes, an ...
An international study investigating the genomic diversity of the Sudanese population reveals that the Copts originating in Egypt –who settled in the country between the seventh and eleventh centuries– have acquired a genetic variant that protects them from contracting malaria. “The acquisition of this variant has taken place very quickly, in just 1,500 years, after a group of Copts mixed with Sudanese populations with sub-Saharan characteristics”, explains David Comas, principal investigator at the Institute of Evolutionary Biology (IBE: CSIC-Pompeu Fabra University) and a full ...
Hydrogen plays an important role in society’s energy transition. For the technology to be used on a broad scale, effective hydrogen sensors are required to prevent the formation of flammable oxyhydrogen gas when hydrogen is mixed with air. Now, researchers at Chalmers University of Technology, Sweden, can present a compact sensor that can be manufactured on a large scale and is well suited to the humid environments where hydrogen is to be found. Unlike today’s sensors, the new sensor performs better the more humid it gets.
Image: Chalmers University of Technology | Mia Halleröd Palmgren
Wherever ...
Recent advances in biomolecular archaeology have revealed that ancient objects can retain the molecular fingerprints of past aromatic practices. These molecules provide unprecedented insight into ancient perfumery, medicine, ritual, and daily life.
In a new publication, an interdisciplinary research team led by archaeo-chemist Barbara Huber (Max Planck Institute of Geoanthropology and the University of Tübingen), shows how museums can use this molecular evidence to engage audiences with the sensory worlds of the past. The team combined their expertise to create a new workflow ...
A new Canadian study has found that people living with dementia (PLWD) are often excluded from research due to assumptions of incapacity and variations in institutional processes. The authors argue that with rights-based, supported approaches, PLWD can participate meaningfully in decisions that affect their lives.
“To ignore the PLWD from research is to exclude a critical piece of information that may affect research outcomes,” says Jim Mann, an advocate living with dementia and co-researcher in the study.
The study, based on interviews with dementia researchers across Canada, reveals how a range ...
