Thursday, August 21, 2025

Ambient documentation technologies reduce physician burnout and restore ‘joy’ in medicine



AI-driven scribes that record patient visits and draft clinical notes for physician review led to significant reductions in physician burnout and improvements in well-being, according to Mass General Brigham-led study of two large healthcare systems




Mass Eye and Ear






A new study led by Mass General Brigham researchers reveals that ambient documentation technologies – generative artificial intelligence scribes that record patient visits and draft clinical notes for physician review before incorporating into electronic health records – led to significant reductions in physician burnout. The findings, published in JAMA Network Open, draw on surveys of more than 1,400 physicians and advanced practice providers at both Mass General Brigham based and Atlanta’s Emory Healthcare.

At Mass General Brigham, use of ambient documentation technologies was associated with a 21.2% absolute reduction in burnout prevalence at 84 days, while Emory Healthcare saw a 30.7% absolute increase in documentation-related wellbeing at 60 days.

“Ambient documentation technology has been truly transformative in freeing up physicians from their keyboards to have more face-to-face interaction with their patients,” said study co-senior author Rebecca Mishuris, MD, MPH, MS, chief medical information officer at Mass General Brigham and a primary care physician in the healthcare system. “Our physicians tell us that they have their nights and weekends back and have rediscovered their joy of practicing medicine. There is literally no other intervention in our field that impacts burnout to this extent.”

Physician burnout affects more than 50% of U.S. doctors and has been linked to time spent in electronic health records, particularly after hours. There is additional evidence that the burden and anticipation of needing to complete their appointment notes also contributes significantly to physician burnout.

“Burnout adversely impacts both providers and their patients who face greater risks to their safety and access to care,” said Lisa Rotenstein, MD, MBA, a co-senior study author and director of The Center for Physician Experience and Practice Excellence at Brigham and Women’s Hospital. She is also an assistant clinical professor of medicine at the UCSF School of Medicine. “This is an issue that hospitals nationwide are looking to tackle, and ambient documentation provides a scalable technology worth further study.”

The researchers analyzed survey data from pilot users of ambient documentation technologies at two large health systems. At Mass General Brigham, 873 physicians and advanced practice providers were given surveys before enrolling, then after 42 and 84 days. About 30% of users responded to the surveys at 42 days, and 22% at 84 days. All 557 Emory pilot users were surveyed before the pilots and then at 60 days of use, with an 11% response rate. Researchers analyzed the survey results quantifying different measures of burnout at Mass General Brigham and physician well-being at Emory Healthcare.

Qualitative feedback from users touted that ambient documentation enabled more “contact with patients and families,” improvements in their “joy in practice,” while recognizing its potential to “fundamentally [change] the experience of being a physician.” However, some users felt it added time to their note-writing or had less utility for certain visit types or medical specialties. Since the pilot studies initiated, the AI technologies have continued to evolve as the vendors make changes based on user feedback and as the large language models that power the technologies improve themselves through additional training, warranting continued study.

The study authors added that given that these were pilot users and there were limited survey response rates, these findings likely represent the experience of more enthusiastic users, and more research is needed to track clinical use of ambient documentation across a broader group of providers.

Mass General Brigham’s ambient documentation program launched in July 2023 as a proof-of-concept pilot study involving 18 physicians. By July 2024, the pilot, which tested two different ambient documentation technologies, expanded to more than 800 providers. As of April 2025, the technologies have been made available to all Mass General Brigham physicians, with more than 3,000 providers routinely using the tools. Later this year, the program will look to expand to other healthcare professionals such as nurses, physical and occupational therapists, and speech-language pathologists.

Ambient documentation’s use will continue to be studied with surveys and other measures tracking burnout rates and time spent on clinical notes inside and outside of working hours. Researchers will evaluate whether burnout rates improve over time as the AI evolves, or if these burnout gains plateau or are reversed.

“Ambient documentation technology offers a step forward in health care and new tools that may positively impact our clinical teams,” said Jacqueline You, MD, MBI, lead study author and a digital clinical lead and primary care physician at Mass General Brigham. “While stories of providers being able to call more patients or go home and play with their kids without having to worry about notes are powerful, we feel the burnout data speak similar volumes of the promise of these technologies, and importance of continuing to study them.”

Authorship: Mass General Brigham co-authors include Adam Landman, David Y. Ting, Sayon Dutta, Amanda J. Centi, Molly Macfarlane, Eran Bechor, Jonathan Letourneau, Gabrielle Choo-Kang, Esther H. Kim, Cordula Magee, Brian J. Lang, Laura Angelo, Michelle Frits, Christine Iannaccone, Angela Rui, and David W. Bates. Additional co-authors included Reema H. Dbouk (Emory), Julie C. Wang (Harvard), Jackson Olin (Northeastern), Ivana Salmikova (Emory), Bryan Blanchette (Emory), Rachel Silverman (Emory), Christopher Holland (Emory)

Disclosures: You reported personal fees from MedDocLive outside of the submitted work. Landman reported receiving consultant fees from Abbott outside the submitted work and that Mass General Brigham has an institutional investment in Abridge. Mishuris is a member of the advisory council at Elsevier, Inc. Bates reported receiving equity from  ValeraHealth, CLEW, MDClone, AESOP< FeelBetter, and Guided Clinical Solutions and personal fees from FeelBetter, Guided Clinical Solutions, and Releyens outside the submitted work, as well as a patent issued to Brigham and Women’s Hospital. Rotenstein reported receiving grants from the American Medical Association, Agency for Healthcare Research and Quality, FeelBetter, Endowment for Advancing a Healthier Wisconsin, and The Doctors Company; serving on an artificial intelligence advisory board for Augmedix; and personal fees from Phreesia outside the submitted work. No other disclosures were reported.

Funding: This project received financial support from the Physician’s Foundation and the National Library of Medicine of the National Institutes of Health (NIH) grant T15LM007092.

Paper cited: You, et al. “Impact of ambient documentation technology on physician and advanced practice provider experience.” JAMA Network Open, DOI: 10.1001/jamanetworkopen.2025.28056

 

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.

 

Study links rising temperatures and declining moods


An analysis of social media in 157 countries finds hotter weather is associated with more negative sentiments.




Massachusetts Institute of Technology





CAMBRIDGE, MA -- Rising global temperatures affect human activity in many ways. Now, a new study illuminates an important dimension of the problem: Very hot days are associated with more negative moods, as shown by a large-scale look at social media postings.

Overall, the study examines 1.2 billion social media posts from 157 countries over the span of a year. The research finds that when the temperature rises above 95 degrees Fahrenheit, or 35 degrees Celsius, expressed sentiments become about 25 percent more negative in lower-income countries and about 8 percent more negative in better-off countries. Extreme heat affects people emotionally, not just physically.

“Our study reveals that rising temperatures don’t just threaten physical health or economic productivity — they also affect how people feel, every day, all over the world,” says Siqi Zheng, a professor in MIT’s Department of Urban Studies and Planning (DUSP) and Center for Real Estate (CRE), and co-author of a new paper detailing the results. “This work opens up a new frontier in understanding how climate stress is shaping human well-being at a planetary scale.”

The paper, “Rising Temperatures Are Altering Human Sentiment Globally,” is published today in the journal One Earth. The authors are Jianghao Wang, of the Chinese Academy of Sciences; Nicolas Guetta-Jeanrenaud SM ’22, a graduate of MIT’s Technology and Policy Program (TPP) and Institute for Data, Systems, and Society; Juan Palacios, a visiting assistant professor at MIT’s Sustainable Urbanization Lab (SUL) and an assistant professor Maastricht University; Yichun Fan, of SUL and Duke University; Devika Kakkar, of Harvard University; Nick Obradovich, of SUL and the Laureate Institute for Brain Research in Tulsa; and Zheng, who is the STL Champion Professor of Urban and Real Estate Sustainability at CRE and DUSP. Zheng is also the faculty director of CRE and founded the Sustainable Urbanization Lab in 2019.

Social media as a window

To conduct the study, the researchers evaluated 1.2 billion posts from the social media platforms Twitter and Weibo, all of which appeared in 2019. They used a natural language processing technique called Bidirectional Encoder Representations from Transformers (BERT), to analyze 65 languages across the 157 countries in the study.

Each social media post was given a sentiment rating from 0.0 (for very negative posts) to 1.0 (for very positive posts). The posts were then aggregated geographically to 2,988 locations and evaluated in correlation with area weather. From this method, the researchers could then deduce the connection between extreme temperatures and expressed sentiment.

“Social media data provides us with an unprecedented window into human emotions across cultures and continents,” Wang says. “This approach allows us to measure emotional impacts of climate change at a scale that traditional surveys simply cannot achieve, giving us real-time insights into how temperature affects human sentiment worldwide.”

To assess the effects of temperatures on sentiment in higher-income and middle-to-lower-income settings, the scholars also used a World Bank cutoff level of gross national income per-capita annual income of $13,845, finding that in places with incomes below that, the effects of heat on mood were triple those found in economically more robust settings.

“Thanks to the global coverage of our data, we find that people in low- and middle-income countries experience sentiment declines from extreme heat that are three times greater than those in high-income countries,” Fan says. “This underscores the importance of incorporating adaptation into future climate impact projections.”

In the long run

Using long-term global climate models, and expecting some adaptation to heat, the researchers also produced a long-range estimate of the effects of extreme temperatures on sentiment by the year 2100. Extending the current findings to that time frame, they project a 2.3 percent worsening of people’s emotional well-being based on high temperatures alone by then — although that is a far-range projection.

“It's clear now, with our present study adding to findings from prior studies, that weather alters sentiment on a global scale,” Obradovich says. “And as weather and climates change, helping individuals become more resilient to shocks to their emotional states will be an important component of overall societal adaptation.”

The researchers note that there are many nuances to the subject, and room for continued research in this area. For one thing, social media users are not likely to be a perfectly representative portion of the population, with young children and the elderly almost certainly using social media less than other people. However, as the researchers observe in the paper, the very young and elderly are probably particularly vulnerable to heat shocks, making the response to hot weather possible even larger than their study can capture.

The research is part of the Global Sentiment project led by the MIT Sustainable Urbanization Lab, and the study’s dataset is publicly available. Zheng and other co-authors have previously investigated these dynamics using social media, although never before at this scale.

“We hope this resource helps researchers, policymakers, and communities better prepare for a warming world,” Zheng says.

###

The research was supported, in part, by Zheng’s chaired professorship research fund, and grants Wang received from the National Natural Science Foundation of China and the Chinese Academy of Sciences.

https://www.cell.com/one-earth/abstract/S2590-3322(25)00248-9 

 

 

New AI tool tracks early signs of hurricane formation



An artificial intelligence system developed by the University of Miami is giving forecasters the first automated way to distinguish between key tropical weather patterns in the Atlantic and Pacific—an advance now in use at the National Hurricane Center



University of Miami Rosenstiel School of Marine, Atmospheric, and Earth Science

New AI tool tracks early signs of hurricane formation 

image: 

Satellite image shows three tropical waves in the Atlantic on June 28, 2024. The wave to the far right developed into Hurricane Beryl, which became the earliest Category 5 Atlantic hurricane on record. 

view more 

Credit: NOAA





An artificial intelligence system developed by the University of Miami is giving forecasters the first automated way to distinguish between key tropical weather patterns in the Atlantic and Pacific—an advance now in use at the National Hurricane Center for the 2025 Atlantic hurricane season.

MIAMI — A research team led by a Ph.D. student at the University of Miami Rosenstiel School of Marine, Atmospheric, and Earth Science has developed a new artificial intelligence (AI) tool that can automatically identify and track tropical easterly waves (TEWs)—clusters of clouds and wind that often develop into hurricanes—and separate them from two major tropical wind patterns: the Intertropical Convergence Zone (ITCZ) and the monsoon trough (MT).

“With this wave tracking tool, we have a new way to detect different patterns, and the types of systems that can grow into hurricanes,” said Will Downs, a Ph.D. student in the Department of Atmospheric Sciences at the Rosenstiel School who led the development of the system. “It’s one important step toward improving forecasts and giving communities more time to prepare.”

The artificial intelligence (AI) model, created by Downs and his collaborators, addresses long-standing gaps in forecasting. Existing tools have struggled to track TEWs in regions like the Caribbean, and no automated method has existed to tell the ITCZ and MT apart—until now.

Using four decades of weather data from 1981 to 2023, Downs trained convolutional neural networks (CNNs)—a form of AI—to detect and differentiate these systems in real time. His model combines historical observations from the National Hurricane Center’s (NHC) Tropical Analysis and Forecast Branch with reanalysis data of past weather and climate conditions.

Forecasters at the NHC now have real-time access to the wave tracker. 

“It has captured the waves where they seem to be going, with remarkable accuracy so far,” said Sharan Majumdar, a professor of atmospheric sciences at the Rosenstiel School and Downs’ advisor. “The robust dataset it produces will help researchers more effectively study the behavior of these waves—from weak clusters of clouds to developing tropical cyclones.”

The analysis also found that tropical waves in the Caribbean tend to be weaker than in the open Atlantic yet remain detectable with AI. It identified a westward expansion of the monsoon trough in the Atlantic over recent decades and shifts in the Pacific during strong El Niño events.

Downs’ fascination with storms began in New Orleans during Hurricane Katrina, when his family evacuated to rural Louisiana. After Hurricane Isaac struck in 2012, he began tracking storms online, a passion that eventually led him to pursue a doctoral degree at the Rosenstiel School studying cyclogenesis—the process by which hurricanes form and intensify.

The project also benefited from collaboration with fellow atmospheric sciences Ph.D. student Aidan Mahoney, who interns at the NHC. “What started as a quick question about tropical wave analysis turned into many long discussions about the complexities of tropical wave dynamics,” Mahoney said. “Will developed an expert understanding of the training data, which allowed him to create the best possible version of the tracker.”

The study titled, Using Deep Learning to Identify Tropical Easterly Waves, the Intertropical Convergence Zone, and the Monsoon Trough, was published August 8, 2025 in the Monthly Weather Review of the American Meteorological Society. 

The research was funded by a National Science Foundation grants AGS-1747781 and AGS-2438140, a University of Miami Provost’s Fellowship in Interdisciplinary Computing, and a University of Miami Fellowship.

About the University of Miami and Rosenstiel School of Marine, Atmospheric, and Earth Science

The University of Miami is a private research university and academic health system with a distinct geographic capacity to connect institutions, individuals, and ideas across the hemisphere and around the world. The University’s vibrant academic community comprises 12 schools and colleges serving more than 19,000 undergraduate and graduate students in more than 180 majors and programs. Located within one of the most dynamic and multicultural cities in the world, the University is building new bridges across geographic, cultural, and intellectual borders, bringing a passion for scholarly excellence, a spirit of innovation, and a commitment to tackling the challenges facing our world. The University of Miami is a member of the prestigious Association of American Universities (AAU).

 Founded in 1943, the Rosenstiel School of Marine, Atmospheric, and Earth Science is one of the world’s premier research institutions in the continental United States. The School’s basic and applied research programs seek to improve understanding and prediction of Earth’s geological, oceanic, and atmospheric systems by focusing on four key pillars:

*Saving lives through better forecasting of extreme weather and seismic events. 

*Feeding the world by developing sustainable wild fisheries and aquaculture programs. 

*Unlocking ocean secrets through research on climate, weather, energy and medicine. 

*Preserving marine species, including endangered sharks and other fish, as well as protecting and restoring threatened coral reefs. www.earth.miami.edu.

 

 

Advancing disaster response with the EBD dataset


Extensible Building Damage (EBD)


Journal of Remote Sensing
Illustration of the proposed SS-FT framework. 

image: 

Illustration of the proposed SS-FT framework. (A-B) show the overall SS-FT and the model’s dataflow. (C-D) elaborates on the supervised fine-tuning, and the self-supervised contrastive learning processes. For each mini-batch, Lcontra is calculated on positive ”queries” and negative representations stored in the category-wise memory bank on a pixel level.

view more 

Credit: Journal of Remote Sensing






A new dataset, the Extensible Building Damage (EBD) dataset, offers significant improvements in disaster response mapping by combining satellite imagery and deep learning techniques. This dataset, covering 12 natural disasters, uses semi-supervised fine-tuning (SS-FT) to reduce the time and effort traditionally required for manual damage labeling, speeding up disaster recovery efforts globally.

Building damage assessments (BDA) are crucial for post-disaster recovery, as they help in identifying areas most in need of urgent assistance. However, current BDA methods suffer from slow dataset development, largely due to manual labeling requirements. The new Extensible Building Damage (EBD) dataset addresses this by leveraging deep learning for semi-automated labeling, improving the speed and accuracy of damage assessment in disaster zones. Based on these challenges, or due to these problems, there is a need for further research into semi-automated disaster response technologies.

Researchers from Zhejiang University and the RIKEN Center for Advanced Intelligence in Japan, with collaboration from various international institutions, have introduced the EBD dataset, published (DOI: 10.34133/remotesensing.0733) in Journal of Remote Sensing. This dataset represents a leap in disaster mapping by using machine-driven annotation to assist human experts in quickly categorizing building damage post-disaster. The SS-FT method it uses provides an innovative solution to the traditionally slow and labor-intensive task of damage classification.

The EBD dataset includes over 18,000 image pairs from 12 major natural disasters, with labels for over 175,000 buildings. Unlike earlier efforts, the dataset uses a semi-automatic annotation process, drastically reducing the manual workload by 80%. The SS-FT method not only utilizes a small amount of manually labeled data but also incorporates large sets of unlabeled samples for improved accuracy. This breakthrough provides faster, more reliable damage assessment results, particularly in areas with limited human resources.

The process begins with a pre-trained model using a historical dataset, which is then fine-tuned on disaster-specific data through the SS-FT method. By comparing pre- and post-disaster images, the model automatically classifies damage into four categories: No Damage, Minor Damage, Major Damage, and Destroyed. The SS-FT method has proven to improve model accuracy especially in situations with limited labeled samples. This capability is demonstrated through disaster events such as Hurricane Ian and the Turkey Earthquake, where the model showed significant improvements over pre-trained only setting and supervised fine-tuning setting.

"By reducing the reliance on manual labeling, the EBD dataset represents a major step forward in how we can use artificial intelligence in disaster response," said Dr. Zeyu Wang, a leading researcher on the project. "This system not only accelerates post-disaster recovery but also makes it more scalable, meaning it can be used globally to address future disaster events."

The research used high-resolution satellite imagery from the Maxar Open-Data Program, processing bi-temporal images to assess building damage. The SS-FT method was implemented using the PyTorch framework, with the model optimized on NVIDIA GPUs. The process involved multiple rounds of fine-tuning, using both labeled and unlabeled data to improve damage classification accuracy.

The EBD dataset has the potential to transform emergency response by providing rapid, accurate damage assessments. As this dataset continues to grow, it could be integrated into broader global disaster monitoring systems, offering valuable insights for climate change-related disasters. Additionally, its semi-automated labeling system can be applied to new disaster scenarios, making it an indispensable tool for disaster management worldwide. The future of disaster response relies on datasets like EBD, offering more timely and precise interventions to save lives.

###

References

DOI

10.34133/remotesensing.0733

Original Source URL

https://spj.science.org/doi/10.34133/remotesensing.0733

Funding Information

This work was supported by the National Key Research and Development Program of China under495Grants 2019YFE0127400.

About Journal of Remote Sensing

The Journal of Remote Sensingan online-only Open Access journal published in association with AIR-CAS, promotes the theory, science, and technology of remote sensing, as well as interdisciplinary research within earth and information science.

 

Risk management: Making gene therapy safer and more effective







Yale University




The ability to correct disease-causing genetic mistakes using genome editors holds great promise in medicine, but it is not without risk. When this type of “genetic surgery” is performed on DNA, for instance, there is always the danger of leaving permanent genetic scars that may even be heritable.

To alleviate this risk, researchers have experimented with gene editing processes on messenger RNA (mRNA), a central link between DNA and proteins that doesn’t carry the same risks because it doesn’t involve permanent changes to the DNA. But existing RNA editing tools have proven either too cumbersome to use or too toxic to human cells. 

Yale researchers have developed a new — and safe — family of RNA-editing tools that utilize an RNA-targeting activity that they found “hidden” inside a popular gene editing tool known as CRISPR-Cas9. 

“The solution was surprisingly simple,” said study lead author Ailong Ke, professor of molecular biophysics and biochemistry at Yale School of Medicine and a member of Yale’s Faculty of Arts and Sciences. “We discovered robust RNA-targeting activity hidden inside [the CRISPR tool] and its related enzyme, IscB, and simply unleashed its hidden power to target RNA.” 

Their findings were published in the journal Cell.

CRISPR (clustered regularly interspaced short palindromic repeats) are DNA sequences found in the genomes of organisms — such as bacteria and archaea — whose cells lack a nucleus and other membrane-bound organelles. Cas9 (CRISPR-associated protein 9) is an enzyme that uses CRISPR sequences. Cas9 enzymes and CRISPR sequences form the basis of the CRISPR-Cas9 technology used to edit genes in living organisms.

The approach was guided by “a deep understanding of the molecular structures of IscB,” including findings reported by the lab in the journal Science, said Chengtao Xu, a postdoctoral associate at Yale and first author of the study.

“It would be much harder to come up with the same idea from Cas9, because its structure is way more sophisticated than IscB.” said Xu. “Nature leaves a lot of treasures for us, and it’s challenging but intriguing to reveal them. This is something we’re particularly good at in molecular biophysics and biochemistry.”

Researchers named their new tools R-IscB and R-Cas9 and defined their usage in genome research and medicine.

“They are the Swiss army knives for RNA editing,” Ke said. “We show that they can be used to perturb mRNA functions, to slice and destroy the targeted mRNA, or to correct the coding mistakes in the mRNA target.

“In essence, we have a way to perform any type of genetic surgery at the RNA level, which is a big deal.”

Xu added that the tools worked just as well on the enzyme Cas9 targets, which use CRISPR sequences. “We’re really excited to see how far we can take this approach with other similar tools,” he said. 

Researchers now plan to test the tools in the lab to cure rare genetic diseases or to promote wound healing.

“We’re particularly excited about the trans-splicing reactions performed by the R-IscBs, because it can potentially correct any type of genetic mutations at the RNA level. This is a huge opportunity for genome medicine,” Ke said. 

“There are a lot of potential applications. The new tool is robust, very precise, and quite versatile.”

Other study authors include Xiaolin Niu and Haifeng Sun, who are postdoctoral associates at Yale. The study also involved collaborator Professor Weixin Tang from the University of Chicago.