DeepSeek-R1 offers promising potential to accelerate healthcare transformation
Peer-Reviewed Publication
Updates every hour. Last Updated: 5-Nov-2025 07:11 ET (5-Nov-2025 12:11 GMT/UTC)
In a Perspective article published in MedComm – Future Medicine, a joint team from The Hong Kong University of Science and Technology and The Hong Kong University of Science and Technology (Guangzhou) explores how the emerging large language model DeepSeek-R1 may accelerate the transformation of healthcare. Highlighting its open-source, low-cost and interpretable capabilities, the study discusses how DeepSeek-R1 can enhance diagnostic efficiency, support clinical decision-making, and improve patient engagement across diverse medical settings.
Researchers from the State Key Laboratory of Green Pesticide at Guizhou University have discovered how the Cucumber Green Mottle Mosaic Virus (CGMMV) exploits the host protein cytosolic Fructose-1,6-bisphosphatase (FBPase) to form biomolecular condensates (BMCs) through liquid-liquid phase separation (LLPS), thereby facilitating viral replication. The study further indicates that a novel compound, C1, can effectively disrupt this interaction, leading to significant inhibition of CGMMV infection. The relevant research findings have been published in Science Bulletin.
Led by corresponding authors Prof. Runjiang Song and Academician of Chinese Academy of Engineering Baoan Song, the team identified NbFBPase as a key interacting protein of CGMMV’s capsid protein (CP). Mutations in CP residues Tyr18 impaired BMCs formation and reduced viral pathogenicity. Compound C1, a benzo[d]oxazole derivative, specifically targets Tyr18, outperforming existing antiviral agents. Moreover, the researchers found that Tyr18 of CGMMV-CP plays a critical role in regulating photosynthesis-related processes during infection by modulating the expression of genes involved in the Calvin cycle.
This work not only elucidates a key virus-host interaction but also provides a blueprint for designing targeted antiviral drugs.
The National Heart Centre Singapore (NHCS) announces a major advancement in cardiac care research with the implementation of SENSE (Singapore hEart lesioN analySEr), a nationwide AI initiative that reduces the time taken to analyse cardiac scans, from hours to minutes. This breakthrough system will transform the detection and prediction of coronary artery disease (CAD) through advanced machine learning technology.
The NHCS CardioVascular Systems Imaging and Artificial Intelligence (CVS.AI) Research Laboratory1 is spearheading SENSE, a project co-led with A*STAR Institute for Infocomm Research (A*STAR I2R), which implements sophisticated computational capabilities and algorithms that can automatically interpret cardiac imaging scans and evaluate CAD risk within minutes – a process that traditionally requires two to four hours of specialist analysis.
SENSE will be deployed at three major healthcare institutions – NHCS, National University Hospital, and Tan Tock Seng Hospital – in the third quarter of 2025. This implementation represents a crucial step forward in addressing CAD, which currently accounts for nearly one-third of cardiovascular-related deaths in Singapore.
Using a computational strategy that allows them to combine information from many large datasets, MIT researchers have identified several new potential targets for Alzheimer’s disease.
In the future, autonomous delivery drones could independently assess whether their remaining battery charge is sufficient for upcoming deliveries. A team of researchers from Technical University of Darmstadt and the University of Sheffield, in collaboration with the French National Institute for Research in Digital Science and Technology (INRIA) and industry partner Ingeniarius Ltd, has developed a new method for energy-aware deployment planning. The approach enables each drone to learn what orders it is capable of fulfilling even when not knowing its own battery health . It is shown to reduce delivery times and increase the number of processed orders compared to conventional approaches.