Wearable smart patch could deliver automatic aid during a fentanyl overdose

Virginia Tech

LinkedInWeChatBlueskyMessageWhatsAppEmail

 

image: 

Researcher Penghui Zhao holds up the iNal patch, a wearable medical device that can prevent accidental opioid overdoses.

view more 

Credit: Photo courtesy of Penghui Zhao.

A patch smaller than a penny could one day save the lives of people experiencing a fentanyl overdose – even when no one else is there to help. 

Researchers in Virginia Tech’s Department of Biological Systems Engineering have developed a wearable microneedle patch that can detect fentanyl in the body and automatically release naloxone, a medication that can reverse an opioid overdose. The technology could change how opioid overdoses are treated by making lifesaving intervention possible without relying on a bystander. 

The patch, called the iNal patch, was developed by a research team led by Wujin Sun, assistant professor of biological systems engineering, which is in both the College of Agriculture and Life Sciences and the College of Engineering. The team's findings were recently published in Advanced Science. 

“In general situations when there’s an overdose, we need to have someone there to save you,” Sun said. “In this case, we don’t need a bystander because we’re protected all the time.”  

A patch that senses and responds 

The patch uses an array of 121 microscopic needles to reach the fluid just beneath the skin. When pressed into place, the needles penetrate only far enough to access that fluid while causing minimal tissue damage. 

Inside the patch are porous silica nanoparticles loaded with naloxone, a medication that blocks the effects of opioids. The openings of those nanoparticles are covered with fentanyl-sensitive molecular gates.

When fentanyl reaches the patch, it causes the gates to open, allowing naloxone to be released into the body. Higher fentanyl concentrations trigger the release of more medication. Because only some of the pores open during each response, the patch retains naloxone that can be released during future fentanyl exposure. 

Penghui Zhao, the paper’s first author and a visiting instructor in Virginia Tech’s Academy of Integrated Science, led the development and testing of the patch.

“One of the greatest technical challenges was finding the right combination of biomaterials and microneedle technology to achieve reliable, on-demand drug release while maintaining mechanical strength, biocompatibility, and responsiveness,” Zhao said. 

The repeated-release feature could be particularly important because the effects of fentanyl may last longer than those of naloxone. After an initial dose of naloxone wears off, overdose symptoms can return. The patch is designed to respond again if fentanyl levels remain high.

Sun describes the device as a harm-reduction tool rather than a replacement for pain medication or other medical care. Opioids are commonly prescribed for legitimate pain relief, and accidental overdoses can occur when someone becomes confused about whether a dose has already been taken. 

“We expect an opioid concentration in the bloodstream because you need pain relief,” Sun said. “But we don’t want that to be too high. We designed the sensor to monitor the opioid concentration, and once it reaches a threshold, it triggers the release of an antagonist that can prevent overdose.”

Although the researchers used fentanyl and naloxone to demonstrate the system, Sun said the underlying chemistry could be adapted to recognize other opioids or release other drugs that block their effects. 

The iNal patch is smaller than a penny. Photo courtesy of Penghui Zhao.

Promising results in early testing

The researchers first tested the patch in laboratory settings to determine whether fentanyl would reliably trigger naloxone release. The system responded within minutes, released larger quantities of naloxone as fentanyl concentrations increased, and continued releasing medication for up to 24 hours. 

They then evaluated the patch in mice. The patch released more naloxone as the fentanyl dose increased and continued to respond through at least three separate fentanyl exposures. 

“The most exciting moment came from the animal studies,” Zhao said. “Seeing the patch respond to fentanyl exposure and effectively reverse opioid-induced effects showed us that the technology could potentially work beyond the laboratory.” 

Mice treated with the patch showed substantially fewer opioid-induced symptoms than mice exposed to fentanyl without it.

The researchers found no significant signs that the patch caused irritation or other harmful effects. More research is needed, however, to understand how well it holds up over time, how consistently it works across different skin types and real-world conditions, and how accurately it can respond to different opioids.  

Moving from the lab to clinical use 

The project began in 2022 and grew from Sun’s broader interest in engineering drug-delivery systems that respond to clinical challenges.  

“I was looking for a challenge where engineering could make a real difference,” Sun said. “When I learned more about opioid overdose and how many lives it claims, I knew this was a problem I wanted to help solve.” 

The study brought together researchers from biological systems engineering, the Virginia Tech College of Science, the School of Neuroscience, the University of Wisconsin–Madison, the University of California Riverside, and the University of Sydney. Zerui Zhou, a doctoral student in Sun’s lab, and Yuanzhi Bian, a BSE postdoctoral fellow, also contributed to the research. 

Sun and Zhao have filed a patent application related to the technology. Sun is now exploring next steps to develop the patch into a product that could eventually be used beyond the laboratory. 

“This is a very general platform,” Sun said. “It has a lot of opportunities.” 

Original study: DOI 10.1002/advs.202524301

---

Journal

Advanced Science

DOI

10.1002/advs.202524301 

Article Title

A Fentanyl-Responsive Microneedle Patch for Harm Reduction

 

Taking screenshots makes you more likely to forget information

Binghamton University psychology research explores the mechanism behind the digital capture effect

Binghamton University

Facebook

XLinkedInWeChatBlueskyMessageWhatsAppEmail

Snapping a photo or a screenshot to remember something? According to recent cognitive research from Binghamton University, State University of New York, the practice may make you more likely to forget.

If you take photographs or screenshots during an experience, consistent evidence has shown that your memory of the information or event is degraded, explained Sophia Fabrizio ’22, MA ’24, a Binghamton University doctoral candidate in psychology and the lead author of “Digital amnesia: The aftermath of a screenshot.” Co-authored with Rebecca Lurie, MA’19, PhD ’23, and Psychology Professor Deanne Westerman, the article was recently published in the journal Memory & Cognition.

Known as the photo-taking impairment effect, the phenomenon occurs for material that the picture-taker doesn’t review afterward. According to other research, using photographs to retrieve and review memories may benefit long-term retention, Fabrizio said. However, many of us take more photos and screenshots than we can use — around 20 photos a day, with around 2,000 photos stored on the average smartphone, according to estimates.

“Unless you are actively reviewing those images as cues for elaborative memory retrieval, it is unlikely to benefit you,” she said.

Another study suggests that our memory isn’t impaired when photographs are captured automatically, using a wearable clip camera; this indicates that there’s something about the act of taking a photo or capturing a screenshot that impairs memory, rather than the knowledge that something is being saved, Fabrizio said.

Binghamton researchers conducted a series of seven experiments to explore the digital capture effect. Participants captured images of artwork and were tested on their memory; some experiments also used math problems to test whether participants’ cognitive resources were reallocated. None showed clear benefits of taking screenshots when compared to the degree of memory impairment for the captured images.

Potential mechanisms

One possible mechanism behind digital amnesia is divided attention: The act of capturing an experience takes away cognitive resources that would otherwise be dedicated to encoding the information in memory. While divided attention plays a role, it’s unlikely to be the main source of memory impairment; people show a comparable deficit when extra time is provided to view an art piece before or after taking a photograph, and when the capture task is made less difficult, which should minimize its effects. 

Another possibility is cognitive offloading, in which we do not allocate cognitive resources to remember information if it’s stored externally. Offloading allows us to redirect those conserved cognitive resources toward aspects of an experience that weren’t captured or toward unrelated tasks.

Individuals should only employ the strategy if the information is reliably saved and accessible. However, research has shown that memory remained impaired even when the picture-takers knew their images would be immediately deleted. With screenshots in particular, people were less likely to remember whether they captured an image or viewed a piece of art, and had worse memory for the art itself when it was captured. 

A third hypothesis is attentional disengagement, in which the act of taking a photograph or screenshot causes us to unconsciously distance ourselves from the experience. 

“This extends beyond captured information to the broader context of an event,” Fabrizio explained. “The poorer ‘source’ memory for whether the art pieces were screenshotted or viewed — and in the original museum study on this effect, worse memory for the location of photographed art pieces in the exhibit — provides some preliminary support for this account.”

This unconscious detachment may be sparked by a longstanding association between capturing images and the ability to offload information. Binghamton researchers and others continue to test predictions related to the attentional disengagement account. 

A better option for remembering information: break out a pen and paper. Writing something down — for example, taking notes during a lecture — forces us to process and organize the information into manageable bullet points, and draw connections. Known as “desirable difficulty,” the mental effort in this kind of processing may make us more likely to remember the information.

However, it’s not foolproof; sometimes, writing something down can lead to cognitive offloading — such as forgetting a friend’s birthday after we add it to our calendar and set reminders. Like the phone numbers in your contact list, you may no longer be able to remember the specific information.

Screenshotting can potentially supplement your memory if you take only a few intentional shots and review them later — similar to how a calendar reminder can support your memory of an upcoming event. But taking frequent screenshots and letting them accumulate unreviewed has the opposite effect, research shows.

“Taken together, the results of the current study suggest that we are likely harming our memory for information and experiences with the press of a button, and that this impairment may even extend beyond what is captured,” Fabrizio said.

---

Journal

Memory & Cognition

DOI

10.3758/s13421-026-01921-2 

Method of Research

Experimental study

Subject of Research

People

Article Title

Digital amnesia: The aftermath of a screenshot

Article Publication Date

14-Aug-2026

 

Multi-source data-driven machine learning reshapes the diagnosis and treatment of lung cancer

A comprehensive review of multi-source data-driven machine learning for lung cancer diagnosis, treatment, and prognosis

Editorial Office of Opto-Electronic Journals Group

Facebook

XLinkedInWeChatEmail

 

image: 

The integrated framework of data, model, and application for machine learning in lung cancer management

view more 

Credit: Editorial Office of Opto-Electronic Journals Group Image source link: https://www.oejournal.org/ioe/article/doi/10.67704/ioe.2026.260006

Section 1: Background
Lung cancer remains one of the malignancies with the highest incidence and mortality worldwide. Clinical practice has long been plagued by core dilemmas, including insufficient sensitivity in early screening, lack of personalized treatment regimens, and limited accuracy in prognostic evaluation, which severely restrict the improvement of patient survival rates. Traditional lung cancer diagnosis and treatment rely heavily on empirical judgment, which can hardly address the high heterogeneity of tumors and the complex evolution of the disease course. With the continuous accumulation of medical resources such as medical imaging, omics detection, liquid biopsy, digital pathology, and electronic health records, lung cancer management has entered a new era driven by multi-source data. Machine learning, with its powerful capabilities in data mining and pattern recognition, can extract latent patterns from complex, heterogeneous, and multi-dimensional medical data. It transforms morphological features, molecular characteristics, pathological structures, blood biomarkers, and clinical information into quantitative evidence for diagnosis, treatment decision-making, and prognosis assessment, thus becoming a key technology to break through the bottlenecks of lung cancer care.

Against this background, a joint team from Shanghai Jiao Tong University, Qilu Hospital of Shandong University, and other institutions systematically analyzed the characteristics of multi-source lung cancer data and the evolution of machine learning technologies. Based on the data-model-application framework, the team comprehensively elaborated the innovative applications of multi-source data-driven machine learning in early screening, diagnosis, treatment optimization, and prognosis evaluation of lung cancer. The review reveals the inherent logic that data characteristics determine model selection and model performance supports clinical value, providing systematic theoretical support and technical references for the deep integration of artificial intelligence and lung cancer clinical decision-making, as well as the implementation of precision medicine.

Section 2: Summary of Review Content
The joint research teams from the Institute of Medical Multimodal Sensing at the School of Automation and Intelligent Sensing, Shanghai Jiao Tong University, the Research Center for Internet of Medical Things at Qilu Hospital of Shandong University, and the Research Center of Shaoxing Keqiao Laboratory Medicine, Chongqing Medical University have published a review article in Volume 2 of Intelligent Opto-Electronics journal on June 29, 2026, entitled Multi-Source Data-Driven Machine Learning for Lung Cancer: Diagnosis, Treatment, and Prognosis. Centered on the main line of data type–model adaptation–clinical application, this review systematically summarizes the dimensional characteristics, clinical value, and complementary mechanisms of five core data sources (imaging, multi-omics, liquid biopsy, pathology, and clinical data). It compares the technical features, applicable scenarios, and advantages/disadvantages of traditional machine learning, deep learning, multimodal fusion, foundation models, and interpretable models, and systematically analyzes the latest progress of multi-source data-driven machine learning in lung cancer diagnosis, treatment, and prognosis.

At the data level, the review details the feature dimensions and clinical value of multi-modal data such as imaging, multi-omics, pathology, liquid biopsy, and clinical data, revealing their complementarity and synergistic effects in lung cancer care.

At the model level, it compares the technical characteristics and applicable scenarios of mainstream machine learning paradigms and clarifies the optimal selection strategies for different data types and clinical tasks.

At the application level, it deeply discusses the advances and challenges of multi-source data-driven machine learning in improving early diagnosis accuracy, optimizing personalized treatment, and realizing dynamic prognostic risk stratification, offering strong AI support for precise clinical decision-making.

This review innovatively proposes a complete logical chain of data characteristics–model adaptation–clinical value, defines optimal model selection strategies in different scenarios, and puts forward systematic solutions to key challenges including insufficient data standardization, bottlenecks in multimodal fusion, lack of algorithm interpretability, and delayed clinical translation, pointing out a clear path for the field to move from laboratory research to clinical application.

Section 3: Outlooks
Multi-source data-driven machine learning is driving a profound transformation of lung cancer care from experience-driven to data-driven, and from extensive intervention to precise stratification. Future efforts should focus on data standardization, adaptive multimodal fusion, improved interpretability, and prospective clinical validation to accelerate clinical translation. The ultimate goal is to build a full-cycle closed-loop diagnosis and treatment system with precise early screening, personalized therapy, and dynamic prognosis, thereby significantly improving the survival rate and quality of life of lung cancer patients.

Section 4: Introduction to the Research Group
This research was jointly completed by the team led by Prof. Jinhong Guo from the School of Automation and Intelligent Sensing, Shanghai Jiao Tong University, Qilu Hospital of Shandong University, and Chongqing Medical University. The team focuses on interdisciplinary research of multimodal medical sensing, medical artificial intelligence, and medical-industrial translation, especially the clinical translation of machine learning in biomedicine. The team has published more than 200 papers in journals including Nature Electronics, Advanced Materials, Angewandte Chemie, and IEEE Transactions series. It has authored 3 monographs, filed over 100 national invention patents (70+ granted), and obtained more than 30 medical device registration certificates. The team has won two First Prizes of Science and Technology Progress Award of Chinese Optical Engineering Society and led more than 10 national, provincial, and industrial projects. Committed to clinical demand, the team promotes the application of multi-source data fusion, deep learning, and interpretable AI in tumor screening, precision treatment, and prognosis evaluation, building an interdisciplinary innovation platform to boost the high-quality development of precision and smart healthcare.

 

Reference
Title of original paper: Multi-source data-driven machine learning for lung cancer: diagnosis, treatment, and prognosis
Journal: Intelligent Opto-Electronics
DOI: https://doi.org/10.67704/ioe.2026.260006

Funding Information 
This work was supported by the National Key R&D Program of China (No. 2023YFF0724300).

---

DOI

10.67704/ioe.2026.260006 

Method of Research

Literature review

Subject of Research

Not applicable

Article Title

Multi-source data-driven machine learning for lung cancer: diagnosis, treatment, and prognosis

 

Students who view more social media use more products that often harbor hormone-disrupting ingredients

Rutgers University

WeChatEmail

Personal care products are among the largest everyday sources of exposure to endocrine-disrupting chemicals, and high schoolers who follow social media influencers or watch beauty tutorials seem to use considerably more of them, according to Rutgers researchers.

In a Rutgers-led study published in the Journal of Exposure Science & Environmental Epidemiology, researchers conducted a survey of product usage among 143 students at Princeton High School in New Jersey. They found that more time on social media and greater engagement with beauty-related content predicted more usage. After researchers accounted for gender, grade, race, ethnicity and parents' education, they found students who followed influencers with product lines reported using 30% more products in the previous 24 hours. Those who watched hair or beauty tutorials used 40% more.

It was a novel finding that researchers on the study team credited a pair of then-high school students, Gabrielle Kaputa and Vita Moss-Wang, who helped design and conduct the survey and suggested it include questions about social media usage.

"Social media turned out to be the most interesting part of our results," said Emily Barrett, the George G. Rhoads Endowed Legacy Professor and vice chair of the Department of Biostatistics and Epidemiology at the Rutgers School of Public Health and the lead author of the study.

“We often hear that social media may hurt kids’ mental health and socioemotional development,” she added. “This study suggests another cause for concern: social media’s potential to increase the use of personal care products that often include hormone-disrupting and potentially carcinogenic chemicals.”

Barrett and her colleagues, who have conducted several studies of adult product use, see much need for more information about student usage patterns.

"Teenagers are a particularly interesting population to study because they are undergoing rapid changes, including in their hormone systems,” Barrett said. “It may be a sensitive period in which better product choices could reduce chemical exposure."

Such products include soap, toothpaste, shampoo, deodorant, sunscreen, skin care products, fragrance and makeup. Some contain phthalates, parabens, phenols or other chemicals that may affect hormone signaling and cancer risk. Previous research has found that people who use more products tend to have higher levels of some of these chemicals in their bodies.

In this study, girls reported using a median of 14 products in the previous day, twice as many as boys. In addition to near universal reports of using soaps and toothpaste in the past day, 57% reported the use of fragrances, which often use endocrine-disrupting phthalates without reporting them on labels. Overall product use among the students was comparable to levels reported in adult studies.

The students voiced some concern about potential chemical exposure, but few took the necessary steps to minimize it. Nineteen percent regularly read ingredient lists, and 5% regularly used an app or website intended to identify healthier products.

"Teenagers don't read labels, it turns out," said Moss-Wang, who was a sophomore at Princeton High School when the project began and is now a sophomore at Wellesley College.

The unusual collaboration began when investigators from the Rutgers Center for Environmental Exposures and Disease approached the Princeton High School research program in 2023. Moss-Wang and Kaputa joined the team, helped design the survey and urged the researchers to explore social media. They also prepared research ethics materials, created a recruitment video and flyers, visited classes and tracked responses. A second recruitment wave reached high school students in Rutgers summer programs.

“Dr. Barrett allowed us to manage the high school population outreach and data collection," said Kaputa, now a junior at the University of Connecticut. "We identified interesting aspects of social media exposure within our student questionnaires." Both students used the data for a three-year research project and became coauthors of the scientific paper.

The professional researchers, meanwhile, are now doing follow-up research with focus groups to learn which accounts, messages and influencers shape teenagers' choices.

Larger studies across more diverse communities could test whether the pattern holds and could pair product inventories with urine samples that measure chemical exposure. Researchers also want to learn whether trusted online voices can steer adolescents toward safer choices.

The study was supported by the National Institute of Environmental Health Sciences through the Rutgers Center for Environmental Exposures and Disease.

---

Journal

Journal of Exposure Science & Environmental Epidemiology

DOI

10.1038/s41370-026-00948-6 

Method of Research

Survey

Subject of Research

People

Article Title

Adolescent personal care product use and social media engagement: a pilot study in NJ high school students

Article Publication Date

17-Aug-2026

 

One step closer to efficient autonomous dental surgery

Advanced robotics could help improve the precision of tooth transplantation procedures, while reducing unnecessary bone removal

Editorial Office of West China School of Stomatology, Sichuan University

Facebook

XLinkedInWeChatBlueskyMessageWhatsAppEmail

 

image: 

In vitro study examines autonomous robotic socket preparation for anatomically complex tooth transplantation procedures 

view more 

Credit: pmuellr via Creative Commons Search Repository Image source link: https://openverse.org/image/3ca803cf-04b0-478e-8cfd-50dd7cf65ab1?q=tooth+implant&p=2

Tooth autotransplantation is a recognized treatment option for replacing missing teeth, particularly in young patients, as it preserves natural tooth function, supports continued jaw development, and eliminates the need for dental implants. A key determinant of success is the precise preparation of the recipient socket, which must closely match the donor tooth while minimizing trauma to the surrounding bone and preserving the delicate periodontal ligament. However, conventional socket preparation relies heavily on surgical experience and manual refinement, making it difficult to consistently achieve the required level of precision. These limitations have driven interest in robotic-assisted surgical systems that can automate socket preparation and potentially improve accuracy, efficiency, and procedural consistency. 

Against this backdrop, researchers from China developed an autonomous multi-axis robotic system capable of creating recipient sockets through nonlinear, surface-conforming osteotomy. The study was led by Dr. Shizhu Bai and Dr. Yimin Zhao from the School of Stomatology, The Fourth Military Medical University, China. The study was published in International Journal of Oral Science on June 30, 2026.  

“Our goal was to evaluate whether robotic-assisted socket preparation could achieve greater geometric accuracy and preserve more surrounding bone than a conventional static guide-assisted approach,” shares Dr. Bai.  

The study was conducted using 40 three-dimensional printed mandibular models representing tooth autotransplantation scenarios. The models were divided equally between robot-assisted and static guide-assisted groups, with each group containing 10 single-rooted and 10 double-rooted tooth anatomies. Digital planning software was used to position donor teeth and design the ideal recipient sockets. To preserve periodontal ligament space, donor root surfaces were expanded by 0.5 mm, and undercuts were removed to allow easier insertion. In the robotic workflow, an autonomous multi-axis robotic system executed preplanned surface-conforming milling paths, while the conventional approach combined guided pilot drilling with freehand socket refinement. Following preparation, digital scans were used to evaluate positional accuracy, socket morphology, bone removal, and procedure time. 

The robotic system demonstrated superior accuracy and geometric fidelity compared with the static guide-assisted approach. While both methods achieved similar accuracy at the socket entry point, robotic preparation significantly reduced deviations at the deepest part of the socket and produced more precise drilling angles. The robot-generated sockets also more closely matched the planned root morphology, showing higher volumetric agreement, lower surface deviation, and substantially less unnecessary bone removal. These advantages were particularly pronounced in double-rooted teeth, where complex anatomy poses greater surgical challenges. Despite these improvements, overall preparation times were comparable between the two techniques. 

The findings highlight the potential of autonomous robotics to make tooth autotransplantation more predictable and less dependent on operator experience. “The technology enables the creation of recipient sockets that closely replicate donor root anatomy while minimizing unnecessary bone removal, potentially preserving surrounding bone, reducing repeated trial insertions, and creating more favorable conditions for periodontal ligament healing and primary stability,” notes Dr. Zhao. The greatest improvements were observed in double-rooted teeth, suggesting that robotic assistance may be particularly valuable for anatomically complex cases where conventional techniques are most challenging. 

Although these findings were obtained using three-dimensional printed laboratory models, they provide an important proof of concept for integrating robotics into tooth autotransplantation. Future clinical studies will be needed to determine whether the improved geometric accuracy translates into shorter extra-alveolar time, enhanced healing, greater transplant stability, and better long-term outcomes in patients. Further research should also evaluate workflow efficiency, learning curves, and the performance of robotic systems under real surgical conditions. 

In conclusion, this study demonstrates that autonomous multi-axis robotic osteotomy can prepare recipient sockets with greater precision than conventional static guide-assisted techniques while maintaining comparable procedure times. By improving socket fidelity and reducing unnecessary bone removal, particularly in complex root anatomies, the technology represents a promising step toward more precise, reproducible, and patient-specific tooth autotransplantation procedures. 

 

*** 

 

Reference 
DOI: 10.1038/s41368-026-00446-3   

 

About The Fourth Military Medical University 
The Fourth Military Medical University (FMMU), also known as the Air Force Medical University, is one of China's leading medical institutions, located in Xi'an, Shaanxi Province. Founded in 1941, the university is renowned for excellence in medical education, clinical care, and biomedical research. FMMU is home to several nationally recognized research platforms, including the State Key Laboratory of Oral & Maxillofacial Reconstruction and Regeneration and the National Clinical Research Center for Oral Diseases. The university has made significant contributions to regenerative medicine, stomatology, trauma care, neuroscience, and military medicine, while fostering innovation through interdisciplinary research and international scientific collaborations. 

Website: https://www.fmmu.edu.cn/ 

 

About Professor Shizhu Bai from The Fourth Military Medical University 
Dr. Shizhu Bai is working as a Professor at the School of Stomatology, The Fourth Military Medical University, China. Prof. Bai’s research focuses on integration of facial scanners, intraoral scanners, and computer-aided design in the field of dentistry. He has published more than 50 articles, with citations over 1,500.  

 

About Dr. Yimin Zhao from The Fourth Military Medical University 
Dr. Yimin Zhao is a Professor and Doctoral Supervisor of the School of Stomatology, The Fourth Military Medical University, China. His research interests include repair functional reconstruction of oral and maxillofacial defects. He has published 280 papers and authored three monographs. Dr. Zhao has established the intelligent simulation and restoration technology system for facial defects, created a systematic technology for masticatory function reconstruction following jaw defects, which significantly improved patients' quality of life and has become an internationally adopted technique. 

 

Funding information 
This study was supported by the National Natural Science Foundation of China (Grant No. 82501241). 

---

Journal

International Journal of Oral Science

DOI

10.1038/s41368-026-00446-3 

Method of Research

Experimental study

Subject of Research

Not applicable

Article Title

Autonomous robotic execution of nonlinear toolpaths for geometry-matched osteotomy in tooth autotransplantation: an in vitro study