Blog

AlphaFold 3 predicts the structure and interactions of all of life’s molecules. May 08, 2024Introducing AlphaFold 3, a new AI model developed by Google DeepMind and Isomorphic Labs. By accurately predicting the structure of proteins, DNA, RNA, ligands and more, and how they interact, we hope it will transform our understanding of the biological world and drug discovery. Inside every plant, animal and human cell are billions of molecular machines. They’re made up of proteins, DNA and other molecules, but no single piece works on its own. Only by seeing how they interact together, across millions of types of combinations, can we start to truly understand life’s processes.In a paper published in Nature, we introduce AlphaFold 3, a revolutionary model that can predict the structure and interactions of all life’s molecules with unprecedented accuracy. For the interactions of proteins with other molecule types we see at least a 50% improvement compared with existing prediction methods, and for some important categories of interaction we have doubled prediction accuracy.We hope AlphaFold 3 will help transform our understanding of the biological world and drug discovery. Scientists can access the majority of its capabilities, for free, through our newly launched AlphaFold Server, an easy-to-use research tool. To build on AlphaFold 3’s potential for drug design, Isomorphic Labs is already collaborating with pharmaceutical companies to apply it to real-world drug design challenges and, ultimately, develop new life-changing treatments for patients.Our new model builds on the foundations of AlphaFold 2, which in 2020 made a fundamental breakthrough in protein structure prediction. So far, millions of researchers globally have used AlphaFold 2 to make discoveries in areas including malaria vaccines, cancer treatments and enzyme design. AlphaFold has been cited more than 20,000 times and its scientific impact recognized through many prizes, most recently the Breakthrough Prize in Life Sciences. AlphaFold 3 takes us beyond proteins to a broad spectrum of biomolecules. This leap could unlock more transformative science, from developing biorenewable materials and more resilient crops, to accelerating drug design and genomics research. &nbsp;<figure id="figure-0liec3x1jcgmlvyckfsh" class=""><img id="0liec3x1jcgmlvyckfsh" src="https://cdn.durable.co/blocks/25l7QgtUCKvfmQg3MDs2vBIApOxgeE4LOqRVpmECrw8O7AuzzxPNZDdz9iwJWLd2.png" alt="books on brown wooden shelf" title="books on brown wooden shelf" class="object-cover bg-gray-100 rounded-sm md:rounded-md lg:rounded-lg" style="aspect-ratio: 16/9; object-position: 20% 50%;" data-corners="default" data-aspect="16:9" data-caption="" data-positionx="20%" data-positiony="50%" data-fullwidth="false" data-mce-src="https://cdn.durable.co/blocks/25l7QgtUCKvfmQg3MDs2vBIApOxgeE4LOqRVpmECrw8O7AuzzxPNZDdz9iwJWLd2.png" data-mce-style="aspect-ratio: 16/9; object-position: 20% 50%;"></figure> 7PNM - Spike protein of a common cold virus (Coronavirus OC43): AlphaFold 3’s structural prediction for a spike protein (blue) of a cold virus as it interacts with antibodies (turquoise) and simple sugars (yellow), accurately matches the true structure (gray). The animation shows the protein interacting with an antibody, then a sugar. Advancing our knowledge of such immune-system processes helps better understand coronaviruses, including COVID-19, raising possibilities for improved treatments.How AlphaFold 3 reveals life’s molecules Given an input list of molecules, AlphaFold 3 generates their joint 3D structure, revealing how they all fit together. It models large biomolecules such as proteins, DNA and RNA, as well as small molecules, also known as ligands — a category encompassing many drugs. Furthermore, AlphaFold 3 can model chemical modifications to these molecules which control the healthy functioning of cells, that when disrupted can lead to disease.AlphaFold 3’s capabilities come from its next-generation architecture and training that now covers all of life’s molecules. At the core of the model is an improved version of our Evoformer module — a deep learning architecture that underpinned AlphaFold 2’s incredible performance. After processing the inputs, AlphaFold 3 assembles its predictions using a diffusion network, akin to those found in AI image generators. The diffusion process starts with a cloud of atoms, and over many steps converges on its final, most accurate molecular structure.AlphaFold 3’s predictions of molecular interactions surpass the accuracy of all existing systems. As a single model that computes entire molecular complexes in a holistic way, it’s uniquely able to unify scientific insights.<figure id="figure-st67f3h31dlvycp8n1" class=""><img id="st67f3h31dlvycp8n1" src="https://cdn.durable.co/blocks/dxtCOpY9rlEmHivErmPcu4ZUblsSWurZETifPMSJeqneeeWI2BvwByUj2oWBkCbo.png" alt="boy in green sweater writing on white paper" title="boy in green sweater writing on white paper" class="object-cover bg-gray-100 rounded-sm md:rounded-md lg:rounded-lg" style="aspect-ratio: 16/9; object-position: center center;" data-corners="default" data-aspect="16:9" data-caption="" data-positionx="undefined" data-positiony="undefined" data-fullwidth="false" data-mce-src="https://cdn.durable.co/blocks/dxtCOpY9rlEmHivErmPcu4ZUblsSWurZETifPMSJeqneeeWI2BvwByUj2oWBkCbo.png" data-mce-style="aspect-ratio: 16/9; object-position: center center;"></figure>7R6R - DNA binding protein: AlphaFold 3’s prediction for a molecular complex featuring a protein (blue) bound to a double helix of DNA (pink) is a near-perfect match to the true molecular structure discovered through painstaking experiments (gray).Leading drug discovery at Isomorphic Labs AlphaFold 3 creates capabilities for drug design with predictions for molecules commonly used in drugs, such as ligands and antibodies, that bind to proteins to change how they interact in human health and disease.AlphaFold 3 achieves unprecedented accuracy in predicting drug-like interactions, including the binding of proteins with ligands and antibodies with their target proteins. AlphaFold 3 is 50% more accurate than the best traditional methods on the PoseBusters benchmark without needing the input of any structural information, making AlphaFold 3 the first AI system to surpass physics-based tools for biomolecular structure prediction. The ability to predict antibody-protein binding is critical to understanding aspects of the human immune response and the design of new antibodies — a growing class of therapeutics.Using AlphaFold 3 in combination with a complementary suite of in-house AI models, Isomorphic Labs is working on drug design for internal projects as well as with pharmaceutical partners. Isomorphic Labs is using AlphaFold 3 to accelerate and improve the success of drug design — by helping understand how to approach new disease targets, and developing novel ways to pursue existing ones that were previously out of reach.AlphaFold Server: A free and easy-to-use research tool<figure id="figure-y67iqapl5olvycr8qt" class=""><img id="y67iqapl5olvycr8qt" src="https://cdn.durable.co/blocks/2dwUZ1daT20Lvl4tcQPezwsB6eSJKZgCZiGvAsRNqWsaJS6v2uNqc0ruBj5D0ZdF.png" alt="person writing on brown wooden table near white ceramic mug" title="person writing on brown wooden table near white ceramic mug" class="object-cover bg-gray-100 rounded-sm md:rounded-md lg:rounded-lg" style="aspect-ratio: 16/9; object-position: center center;" data-corners="default" data-aspect="16:9" data-caption="" data-positionx="undefined" data-positiony="undefined" data-fullwidth="false" data-mce-src="https://cdn.durable.co/blocks/2dwUZ1daT20Lvl4tcQPezwsB6eSJKZgCZiGvAsRNqWsaJS6v2uNqc0ruBj5D0ZdF.png" data-mce-style="aspect-ratio: 16/9; object-position: center center;"></figure>8AW3 - RNA modifying protein: AlphaFold 3’s prediction for a molecular complex featuring a protein (blue), a strand of RNA (purple), and two ions (yellow) closely matches the true structure (gray). This complex is involved with the creation of other proteins — a cellular process fundamental to life and health. Google DeepMind’s newly launched AlphaFold Server is the most accurate tool in the world for predicting how proteins interact with other molecules throughout the cell. It is a free platform that scientists around the world can use for non-commercial research. With just a few clicks, biologists can harness the power of AlphaFold 3 to model structures composed of proteins, DNA, RNA and a selection of ligands, ions and chemical modifications.AlphaFold Server helps scientists make novel hypotheses to test in the lab, speeding up workflows and enabling further innovation. Our platform gives researchers an accessible way to generate predictions, regardless of their access to computational resources or their expertise in machine learning.Experimental protein-structure prediction can take about the length of a PhD and cost hundreds of thousands of dollars. Our previous model, AlphaFold 2, has been used to predict hundreds of millions of structures, which would have taken hundreds of millions of researcher-years at the current rate of experimental structural biology. "With AlphaFold Server, it’s not only about predicting structures anymore, it’s about generously giving access: allowing researchers to ask daring questions and accelerate discoveries." Céline Bouchoux The Francis Crick Institute &nbsp; Sharing the power of AlphaFold 3 responsibly With each AlphaFold release, we’ve sought to understand the broad impact of the technology, working together with the research and safety community. We take a science-led approach and have conducted extensive assessments to mitigate potential risks and share the widespread benefits to biology and humanity.Building on the external consultations we carried out for AlphaFold 2, we’ve now engaged with more than 50 domain experts, in addition to specialist third parties, across biosecurity, research and industry, to understand the capabilities of successive AlphaFold models and any potential risks. We also participated in community-wide forums and discussions ahead of AlphaFold 3’s launch.AlphaFold Server reflects our ongoing commitment to share the benefits of AlphaFold, including our free database of 200 million protein structures. We’ll also be expanding our free AlphaFold education online course with EMBL-EBI and partnerships with organizations in the Global South to equip scientists with the tools they need to accelerate adoption and research, including on underfunded areas such as neglected diseases and food security. We’ll continue to work with the scientific community and policy makers to develop and deploy AI technologies responsibly.Opening up the future of AI-powered cell biology <figure id="figure-k9qts0ak6lvyctbuo" class=""><img id="k9qts0ak6lvyctbuo" src="https://cdn.durable.co/blocks/17Vq9TdkM9w6nbjv21Nqw8j2Gj2ksx9skoTGfKCpZrRAw1sYb93AHDSwliTzwe18.png" alt="MacBook Pro near white open book" title="MacBook Pro near white open book" class="object-cover bg-gray-100 rounded-sm md:rounded-md lg:rounded-lg selected" style="aspect-ratio: 16/9; object-position: center center;" data-corners="default" data-aspect="16:9" data-caption="" data-positionx="undefined" data-positiony="undefined" data-fullwidth="false" data-mce-src="https://cdn.durable.co/blocks/17Vq9TdkM9w6nbjv21Nqw8j2Gj2ksx9skoTGfKCpZrRAw1sYb93AHDSwliTzwe18.png" data-mce-style="aspect-ratio: 16/9; object-position: center center;" data-mce-selected="1"></figure>7BBV - Enzyme: AlphaFold 3’s prediction for a molecular complex featuring an enzyme protein (blue), an ion (yellow sphere) and simple sugars (yellow), along with the true structure (gray). This enzyme is found in a soil-borne fungus (Verticillium dahliae) that damages a wide range of plants. Insights into how this enzyme interacts with plant cells could help researchers develop healthier, more resilient crops. AlphaFold 3 brings the biological world into high definition. It allows scientists to see cellular systems in all their complexity, across structures, interactions and modifications. This new window on the molecules of life reveals how they’re all connected and helps understand how those connections affect biological functions — such as the actions of drugs, the production of hormones and the health-preserving process of DNA repair.The impacts of AlphaFold 3 and our free AlphaFold Server will be realized through how they empower scientists to accelerate discovery across open questions in biology and new lines of research. We’re just beginning to tap into AlphaFold 3’s potential and can’t wait to see what the future holds.... End&nbsp;

AlphaFold 3 predicts the structure and interactions of all of life’s molecules

<h2>Introduction</h2>In recent years, the healthcare industry has been undergoing a transformative shift towards leveraging cutting-edge technologies to address not only physical health but also mental and emotional well-being. One pressing issue that has come to the forefront is the loneliness epidemic, affecting individuals of all ages and backgrounds. However, with the emergence of generative AI technology, there is newfound hope for tackling this pervasive problem in innovative ways.<h3>The Impact of Loneliness on Health</h3>Loneliness is more than just a feeling of isolation; it can have serious implications for both mental and physical health. Studies have shown that chronic loneliness is associated with a higher risk of developing conditions such as depression, anxiety, cardiovascular diseases, and even premature death.<h3>Understanding Generative AI</h3>Generative AI, a subset of artificial intelligence, is revolutionizing the way we approach healthcare by enabling machines to create content autonomously. This technology has the potential to generate personalized interventions that cater to individuals' unique needs and preferences, making it a powerful tool in combating loneliness.<figure id="figure-wwmuwzy8wplvla3zh8"><img id="wwmuwzy8wplvla3zh8" src="https://images.unsplash.com/photo-1617992799119-5d0ed1fb9379?crop=entropy&amp;cs=tinysrgb&amp;fit=max&amp;fm=jpg&amp;ixid=M3wyNjI5NjF8MHwxfHNlYXJjaHwxNnx8Z2VuZXJhdGl2ZSUyMEFJJTIwbG9uZWxpbmVzc3xlbnwwfHx8fDE3MTQ0MTQ0NzB8MA&amp;ixlib=rb-4.0.3&amp;q=80&amp;w=1080" alt="generative AI loneliness" class="object-cover bg-gray-100 rounded-sm md:rounded-md lg:rounded-lg" style="aspect-ratio: 16/9;" data-mce-src="https://images.unsplash.com/photo-1617992799119-5d0ed1fb9379?crop=entropy&amp;cs=tinysrgb&amp;fit=max&amp;fm=jpg&amp;ixid=M3wyNjI5NjF8MHwxfHNlYXJjaHwxNnx8Z2VuZXJhdGl2ZSUyMEFJJTIwbG9uZWxpbmVzc3xlbnwwfHx8fDE3MTQ0MTQ0NzB8MA&amp;ixlib=rb-4.0.3&amp;q=80&amp;w=1080" data-mce-style="aspect-ratio: 16/9;"></figure><h3>Personalized Interventions</h3>One of the key advantages of harnessing generative AI in healthcare is its ability to create tailored interventions for individuals struggling with loneliness. By analyzing vast amounts of data, including social interactions, interests, and communication patterns, AI algorithms can generate personalized recommendations for fostering meaningful connections.<h3>Virtual Companionship</h3>Generative AI can also be utilized to develop virtual companionship programs that provide individuals with round-the-clock support and companionship. These AI-powered companions can engage in conversations, offer emotional support, and even facilitate social interactions, helping to alleviate feelings of loneliness and isolation.<figure id="figure-e03yhfmfpmlvla41mv"><img id="e03yhfmfpmlvla41mv" src="https://images.unsplash.com/photo-1644428239740-b8295bfef42f?crop=entropy&amp;cs=tinysrgb&amp;fit=max&amp;fm=jpg&amp;ixid=M3wyNjI5NjF8MHwxfHNlYXJjaHw5fHx2aXJ0dWFsJTIwY29tcGFuaW9uc2hpcCUyMGxvbmVsaW5lc3N8ZW58MHx8fHwxNzE0NDE0NDcyfDA&amp;ixlib=rb-4.0.3&amp;q=80&amp;w=1080" alt="virtual companionship loneliness" class="object-cover bg-gray-100 rounded-sm md:rounded-md lg:rounded-lg" style="aspect-ratio: 16/9;" data-mce-src="https://images.unsplash.com/photo-1644428239740-b8295bfef42f?crop=entropy&amp;cs=tinysrgb&amp;fit=max&amp;fm=jpg&amp;ixid=M3wyNjI5NjF8MHwxfHNlYXJjaHw5fHx2aXJ0dWFsJTIwY29tcGFuaW9uc2hpcCUyMGxvbmVsaW5lc3N8ZW58MHx8fHwxNzE0NDE0NDcyfDA&amp;ixlib=rb-4.0.3&amp;q=80&amp;w=1080" data-mce-style="aspect-ratio: 16/9;"></figure><h3>Community Building</h3>Another exciting application of generative AI in tackling the loneliness epidemic is in community building. By analyzing social networks and identifying individuals who may be at risk of loneliness, AI algorithms can facilitate the creation of supportive communities where members can connect, share experiences, and provide mutual support.<h3>Empowering Healthcare Providers</h3>Generative AI technology not only benefits individuals directly but also empowers healthcare providers to deliver more personalized and effective care. By leveraging AI-generated insights and recommendations, healthcare professionals can tailor their interventions to better address the underlying causes of loneliness and improve patient outcomes.<h3>Ethical Considerations</h3>While the potential of generative AI in combating the loneliness epidemic is vast, it is crucial to consider the ethical implications of using such technology in healthcare. Safeguarding data privacy, ensuring transparency in AI decision-making processes, and maintaining human oversight are essential to uphold ethical standards and protect individuals' well-being.<h3>The Future of Healthcare</h3>As we continue to harness the power of generative AI to tackle the loneliness epidemic, we are witnessing a paradigm shift in the way we approach mental health and well-being. By combining technological innovation with human empathy, we have the opportunity to revolutionize healthcare and create a more connected and supportive society for all.<figure id="figure-mb09ubmreplvla462b"><img id="mb09ubmreplvla462b" src="https://images.unsplash.com/photo-1579684385127-1ef15d508118?crop=entropy&amp;cs=tinysrgb&amp;fit=max&amp;fm=jpg&amp;ixid=M3wyNjI5NjF8MHwxfHNlYXJjaHwxOHx8ZnV0dXJlJTIwb2YlMjBoZWFsdGhjYXJlfGVufDB8fHx8MTcxNDQxNDQ3OHww&amp;ixlib=rb-4.0.3&amp;q=80&amp;w=1080" alt="future of healthcare" class="object-cover bg-gray-100 rounded-sm md:rounded-md lg:rounded-lg" style="aspect-ratio: 16/9;" data-mce-src="https://images.unsplash.com/photo-1579684385127-1ef15d508118?crop=entropy&amp;cs=tinysrgb&amp;fit=max&amp;fm=jpg&amp;ixid=M3wyNjI5NjF8MHwxfHNlYXJjaHwxOHx8ZnV0dXJlJTIwb2YlMjBoZWFsdGhjYXJlfGVufDB8fHx8MTcxNDQxNDQ3OHww&amp;ixlib=rb-4.0.3&amp;q=80&amp;w=1080" data-mce-style="aspect-ratio: 16/9;"></figure>

Revolutionising Healthcare: Harnessing Generative AI to Tackle the Loneliness Epidemic

Polypharmacy, or the concurrent use of five or more medications, is common in older adults and increases the risk of adverse drug interactions. While deprescribing unnecessary drugs can combat this risk, the decision-making process can be complex and time-consuming. Increasingly, there is a need for effective polypharmacy management tools that can support short-staffed primary care practitioners.In a new study, researchers from the Mass General Brigham MESH Incubator found that ChatGPT, a generative artificial intelligence (AI) chatbot, showed promise as a tool to manage polypharmacy and deprescription. These findings, published April 18th in the Journal of Medical Systems, demonstrate the first use case of AI models in medicine management.To evaluate its utility, the investigators provided ChatGPT with different clinical scenarios and asked it a set of decision-making questions. Each scenario featured the same elderly patient taking a mixture of medications but included variations in cardiovascular disease history (CVD) and degree of impairment in activities of daily living (ADL).When asked yes or no questions about reducing prescribed drugs, ChatGPT consistently recommended deprescribing medications in patients without a history of CVD. However, it was more cautious when overlying CVD was introduced, and more likely to keep the patient’s medication regimen unchanged. In both cases, the researchers observed that ADL impairment severity did not seem to affect decision outcomes.The team also noted that ChatGPT had a tendency to disregard pain and favored deprescribing pain medications over other drug types like statins or antihypertensives. In addition, ChatGPT responses varied when presented with the same scenario in new chat sessions — which the authors suggest could reflect inconsistency in commonly reported clinical deprescribing trends on which the model was trained.More than 40 percent of older adults meet the criteria for polypharmacy. The rate of seniors on Medicare seeing more specialists on their care teams has increased in recent years, leaving primary care providers to oversee medication management. An effective AI tool could help aid this practice, according to the researchers.&nbsp;“Our study provides the first use case of ChatGPT as a clinical support tool for medication management,” said senior corresponding author Marc Succi, MD, Associate Chair of Innovation and Commercialization at Mass General Brigham Radiology and Executive Director of the MESH Incubator. “While caution should be taken to increase accuracy of such models, AI-assisted polypharmacy management could help alleviate the increasing burden on general practitioners. Further research with specifically trained AI tools may significantly enhance the care of aging patients.”Arya Rao, lead author, MESH researcher and Harvard Medical student, added “Our findings suggest that AI-based tools can play an important role in ensuring safe medication practices for older adults; it is imperative that we continue to refine these tools to account for the complexities of medical decision-making.”Read more in the Journal of Medical Systems.

Study Finds ChatGPT Shows Promise as Medication Management Tool, Could Help Improve Geriatric Health Care

NEWS 21 March 2024 Google AI could soon use a person’s cough to diagnose disease Machine-learning system trained on millions of human audio clips shows promise for detecting COVID-19 and tuberculosis. By Mariana Lenharo <figure id="figure-p6fe6b3s57lur44iyu"><img id="p6fe6b3s57lur44iyu" src="https://cdn.durable.co/blocks/15M2ZWOpgNsMqODyjSIslp5sq384SEhPOK4f3DSjr4tmbnQoWYP0sx9i2fPYaEYn.png" alt="person holding pencil near laptop computer" title="person holding pencil near laptop computer" class="object-cover bg-gray-100 rounded-sm md:rounded-md lg:rounded-lg" style="aspect-ratio: 16/9; object-position: undefined undefined;" data-corners="default" data-aspect="16:9" data-caption="" data-positionx="undefined" data-positiony="undefined" data-fullwidth="false" data-mce-src="https://cdn.durable.co/blocks/15M2ZWOpgNsMqODyjSIslp5sq384SEhPOK4f3DSjr4tmbnQoWYP0sx9i2fPYaEYn.png" data-mce-style="aspect-ratio: 16/9; object-position: undefined undefined;"></figure> &nbsp;The field of audiomics combines artificial-intelligence tools with human sounds, such as coughs, to evaluate health.Credit: Getty A team led by Google scientists has developed a machine-learning tool that can help to detect and monitor health conditions by evaluating noises such as coughing and breathing. The artificial intelligence (AI) system1, trained on millions of audio clips of human sounds, might one day be used by physicians to diagnose diseases including COVID-19 and tuberculosis and to assess how well a person’s lungs are functioning. &nbsp;AI detects eye disease and risk of Parkinson’s from retinal images This is not the first time a research group has explored using sound as a biomarker for disease. The concept gained traction during the COVID-19 pandemic, when scientists discovered that it was possible to detect the respiratory disease through a person’s cough2.What’s new about the Google system — called Health Acoustic Representations (HeAR) — is the massive data set that it was trained on, and the fact that it can be fine-tuned to perform multiple tasks.The researchers, who reported the tool earlier this month in a preprint1 that has not yet been peer reviewed, say it’s too early to tell whether HeAR will become a commercial product. For now, the plan is to give interested researchers access to the model so that they can use it in their own investigations. “Our goal as part of Google Research is to spur innovation in this nascent field,” says Sujay Kakarmath, a product manager at Google in New York City who worked on the project.How to train your model Most AI tools being developed in this space are trained on audio recordings — for example, of coughs — that are paired with health information about the person who made the sounds. For example, the clips might be labelled to indicate that the person had bronchitis at the time of the recording. The tool comes to associate features of the sounds with the data label, in a training process called supervised learning. Google AI has better bedside manner than human doctors — and makes better diagnoses “In medicine, traditionally, we have been using a lot of supervised learning, which is great because you have a clinical validation,” says Yael Bensoussan, a laryngologist at the University of South Florida in Tampa. “The downside is that it really limits the data sets that you can use, because there is a lack of annotated data sets out there.”Instead, the Google researchers used self-supervised learning, which relies on unlabelled data. Through an automated process, they extracted more than 300 million short sound clips of coughing, breathing, throat clearing and other human sounds from publicly available YouTube videos.Each clip was converted into a visual representation of sound called a spectrogram. Then the researchers blocked segments of the spectrograms to help the model learn to predict the missing portions. This is similar to how the large language model that underlies chatbot ChatGPT was taught to predict the next word in a sentence after being trained on myriad examples of human text. Using this method, the researchers created what they call a foundation model, which they say can be adapted for many tasks.An efficient learner In the case of HeAR, the Google team adapted it to detect COVID-19, tuberculosis and characteristics such as whether a person smokes. Because the model was trained on such a broad range of human sounds, to fine-tune it, the researchers only had to feed it very limited data sets labelled with these diseases and characteristics.On a scale where 0.5 represents a model that performs no better than a random prediction and 1 represents a model that makes an accurate prediction each time, HeAR scored 0.645 and 0.710 for COVID-19 detection, depending on which data set it was tested on — a better performance than existing models trained on speech data or general audio. For tuberculosis, the score was 0.739. &nbsp;An AI revolution is brewing in medicine. What will it look like? The fact that the original training data were so diverse — with varying sound quality and human sources — also means that the results are generalizable, Kakarmath says.Ali Imran, an engineer at the University of Oklahoma in Tulsa, says that the sheer volume of data used by Google lends significance to the research. “It gives us the confidence that this is a reliable tool,” he says.Imran leads the development of an app named AI4COVID-19, which has shown promise at distinguishing COVID-19 coughs from other types of cough3. His team plans to apply for approval from the US Food and Drug Administration (FDA) so that the app can eventually move to market; he is currently seeking funding to conduct the necessary clinical trials. So far, no FDA-approved tool provides diagnosis through sounds.The field of health acoustics, or ‘audiomics’, is promising, Bensoussan says. “Acoustic science has existed for decades. What’s different is that now, with AI and machine learning, we have the means to collect and analyse a lot of data at the same time.” She co-leads a research consortium focused on exploring voice as a biomarker to track health.“There’s an immense potential not only for diagnosis, but also for screening” and monitoring, she says. “We can’t repeat scans or biopsies every week. So that’s why voice becomes a really important biomarker for disease monitoring,” she adds. “It’s not invasive, and it’s low resource.” Nature 628, 19-20 (2024)doi: https://doi.org/10.1038/d41586-024-00869-0References Baur, S. et al. Preprint at arXiv https://doi.org/10.48550/arXiv.2403.02522 (2024). Santosh, K. C., Rasmussen, N., Mamun, M. &amp; Aryal, S. PeerJ. Comput. Sci. 8, e958 (2022).Article PubMed Google Scholar&nbsp; Imran, A. et al. Inform. Med. Unlocked 20, 100378 (2020).Article PubMed Google Scholar

Google AI has better bedside manner than human doctors — and makes better diagnoses

HEALTH CARE Who pays when AI steers your doctor wrong? The adoption of artificial intelligence in health care is reviving the debate over medical malpractice reform — with a new twist. &nbsp;<figure id="figure-u3ymqj093dlur3akku"><img id="u3ymqj093dlur3akku" src="https://cdn.durable.co/blocks/2fkhX3tbtZVF2uTD6mxuHLBuD2DkVXkdZhfUxzems1v2awQoM87NqWuMKNa2DRNN.jpg" alt="silhouette of people standing on hill" title="silhouette of people standing on hill" class="object-cover bg-gray-100 rounded-sm md:rounded-md lg:rounded-lg" style="aspect-ratio: 16/9; object-position: center center;" data-corners="default" data-aspect="16:9" data-caption="" data-positionx="undefined" data-positiony="undefined" data-fullwidth="false" data-mce-src="https://cdn.durable.co/blocks/2fkhX3tbtZVF2uTD6mxuHLBuD2DkVXkdZhfUxzems1v2awQoM87NqWuMKNa2DRNN.jpg" data-mce-style="aspect-ratio: 16/9; object-position: center center;"></figure>&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; Doctors, tech leaders and lawmakers all hope to have a say in how the medical malpractice system adapts to the AI era. | AFP via Getty ImagesBy DANIEL PAYNE03/24/2024 07:00 AM EDT &nbsp; &nbsp; &nbsp; &nbsp; Doctors using new artificial intelligence tools to help them diagnose and treat their patients say they wish Congress would provide some clarity on a big unanswered question: Who pays if AI makes a mistake?Advancements in AI promise to improve care, but only if doctors trust the systems and are protected from liability, according to the country’s leading physicians’ group, the American Medical Association. “We’re seeing lawsuits already,” AMA President Jesse Ehrenfeld said at a recent meeting on doctors’ lobbying priorities, adding the issue so far has been “a little bit [like] building the plane as you’re flying it.” &nbsp; At stake is more than just millions in medical malpractice payouts, over a dozen health, legal and tech leaders told POLITICO. Judges, lawmakers and regulators are shaping what a medical system infused with artificial intelligence owes patients, both in terms of quality care and the right to recompense if something goes wrong.It won’t happen without a fight. Health tech companies and some hospitals say that doctors making the final call in care are ultimately responsible for their decisions.The AMA has in one way reinforced the point. The group refers to AI as “augmented intelligence” because it says doctors must never rely on it entirely.The explosion of AI tools in health care is reviving the once heated debate over medical malpractice. But it’s not clear that the old battle lines, in which doctors and Republicans pushed for liability limits, and Democrats and plaintiffs’ attorneys pushed back, will hold. In the years since the “med-mal” debate went dormant, doctors’ politics have trended Democratic and the tech giants building the tools have run afoul of both parties. That suggests there could be an opening to give the doctors what they want.There’s no legislation to do that yet, but lawmakers have begun looking into it. Rep. Greg Murphy (R-N.C.), a urologist and co-chair of the GOP Doctors Caucus in the House, recently wrote to Food and Drug Administration Commissioner Robert Califf to ask for Califf’s take and to suggest an approach. &nbsp; Rep. Vern Buchanan (R-Fla.), chair of the Ways and Means Health Subcommittee, told the American Medical Association he sees a role for Congress.“We’re going to look at that and see if we can’t make a big difference,” he said of liability reform.Until Congress does, legal experts said judges will set the parameters of the new system.“Courts are going to need to evolve some of their existing rules [and] doctrines,” said Michelle Mello, a Stanford health law scholar who has testified on the matter before Congress.In the meantime, care providers, tech companies and patients find themselves in an uncertain, even risky, situation.Safety vs. absolution The medical malpractice debate used to center on a GOP list of “tort reform” proposals to cap damage awards, limit plaintiffs’ attorney fees, and shorten the window during which patients could bring suits.Republicans were rarely successful in Congress, but many state legislatures imposed restrictions.Now, the policy decisions about AI liability are less about how much compensation a patient gets, but who bears the blame for a mistake.Murphy says AI could reshape the medical malpractice debate in Washington, so he’s taking an early stab at steering it.In his letter to Califf he suggested a “safe harbor” that would apply to doctors and the AI products they use if both join a surveillance program tracking patient outcomes.“There are going to be great precedents set here on the national level,” Murphy told POLITICO. &nbsp;MOST READ &nbsp; &nbsp; &nbsp; The polls are suggesting a huge shift in the electorate. Are they right? Trump promises a ‘deal’ on abortion that will please everyone. It likely doesn’t exist. Navy cancels ship briefings after damning internal report Democrats fear Netanyahu may have undermined Biden’s image among voters What Liberals Get Wrong About ‘White Rural Rage’ — Almost Everything&nbsp; But Murphy hasn’t proposed a bill, nor has Califf responded to his inquiry.Trial lawyers say they’ll fight if Murphy proceeds.“When we give people absolution from responsibility, it means everyday people bear the brunt of it,” said Sean Domnick, president of the American Association for Justice, a group that represents trial lawyers.AI tools make it “easier to just click, click, click,” he said. “We have to guard against that.”The government’s take-it-slow approach to regulation of the new AI tools is adding legal uncertainty, Mello said.Currently, the FDA regulates some AI-powered medical devices, but its approach to the latest version of the tech is still forming.Unregulated tools could be more vulnerable to lawsuits, according to Mello, because they aren’t protected by the “preemption doctrine,” which bars some claims against regulated devices on the theory that they’re known to be safe enough and effective enough to have received FDA clearance.So how federal agencies like the FDA decide to regulate the latest AI tools could play a role in determining when software makers could be held accountable for errors.It could also prompt some makers of AI products to seek regulatory approval.“If you’re explicitly saying, ‘I’m definitely not a medical device,’ well, now you don’t get to benefit from the preemption doctrine,” Mello said.The legal landscape The case law suggests doctors aren’t likely to have similar avenues to avoid lawsuits.When patients get substandard care because an intake form is incomplete, medical literature is incorrect or a drugmaker doesn’t adequately warn about its risks, courts have rarely let doctors off the hook. &nbsp; In some cases, doctors have followed allegedly errant software suggestions for which drugs to prescribe and how to treat patients — which plaintiffs said harmed them. When sued, courts still took the cases, despite doctors’ efforts to blame the tool.“The group that I worry about most is physicians,” Mello said. “Physicians are the ones that kind of get left holding the bag.”That’s the case for a reason, some health tech companies say. Diagnostic software is like GPS, in their view: It’s up to the driver to stay on the road, no matter what instructions are given.“In that room, there’s only one licensed health care provider,” said Dr. C.K. Wang, chief medical officer at COTA, a health data company working to advance AI models. “I’m liable for the decisions made under my license. Therefore, I do advise clinicians to really truly understand the technology that they’re using.”Making the situation especially thorny for doctors, they also could open themselves up to litigation if they eschew AI.Legal scholars foresee plaintiffs arguing that doctors were negligent because they didn’t use the best tools available to them.Doctors are contemplating a host of other questions, too: What should a physician do if a patient wants to follow the advice of an algorithm instead of their doctor?In what scenarios should doctors trust themselves over a machine?Insuring the unknown There’s big money at stake in the answers — and implications for how quickly doctors adopt AI tools, the specialities they choose, and the places they decide to live.After a big dip during the Covid pandemic, malpractice payouts have spiked and returned to more than $3 billion in 2022, according to statistics collected by the National Practitioner Data Bank. &nbsp; The AMA said last year that many doctors have incurred double-digit percentage increases in their malpractice insurance premiums for four years running.If the upward trend in premiums continues, the AMA said, it could have implications for where doctors choose to practice and what specialities they pick.The AMA noted that premiums vary considerably by speciality. In Philadelphia, internists paid $31,909 for an annual policy in 2022, while general surgeons paid $105,013 and OB/GYNs paid $185,565.And fees varied markedly by region, in part because state laws can affect the risk of high-cost litigation. OB/GYNs in Los Angeles, for instance, paid $49,804, while those in Miami paid $226,224.Insurance companies are taking note of the AI revolution, but making no forecasts on how it will affect premiums.“It’s uncharted territory,” said Melissa Menard, executive vice president of the health care division of CAC Specialty.Insurers need to look at risks from AI that could include more bias, less privacy and new cybersecurity threats, she said.But if the technology, on the whole, improves care, it could also reduce the chances a doctor is sued for malpractice — and possibly reduce insurance costs.Amid the uncertainty, the core principles of practicing medicine responsibly should be front-and-center, lawyers, health leaders and tech companies agree.“We always go back to the standard of care,” Menard said.But that standard, legal and medical experts admit, could change with AI, especially as its adoption continues to grow.Even so, some health systems are intent on using AI as a tool to support and better current standards. &nbsp; “We’re using AI like a stethoscope,” said John Couris, president and CEO of the Florida Health Sciences Center, emphasizing the need to balance excitement about the technology with level-headedness. “It’s a tool to help augment the care that they’re providing, but our doctors and our nurses are the ultimate authority on what gets done or doesn’t get done to a patient.”Other health providers aren’t as calm, including Dr. Wendy Dean, president and co-founder of Moral Injury of Healthcare, an organization that advocates for physicians’ well-being.“The confidence with which AI posits its conclusions,” she said, “makes it really hard as a human to say, ‘Wait a minute, I need to question this.’”

AI in Healthcare: Who Pays When Things Go Wrong?

The medical industry plays a critical role in the health and well-being of individuals worldwide. From developing life-saving drugs to creating innovative medical devices, the industry is constantly evolving to meet the needs of patients. However, with the advancements in technology and the increasing complexity of medical products, ensuring their safety and efficacy has become more important than ever.<h2>The Importance of Product Testing</h2>Product testing is a vital step in the development and manufacturing process of medical products. It involves subjecting the product to rigorous evaluation to determine its safety, effectiveness, and quality. This testing is essential to identify any potential risks or flaws that could compromise patient safety or the product's performance.<h3>Ensuring Patient Safety</h3>One of the primary reasons for product testing in the medical industry is to ensure patient safety. Medical products, such as drugs and devices, are designed to diagnose, treat, or prevent diseases. If these products are not thoroughly tested, they could pose serious risks to patients, leading to adverse reactions or even fatalities.<figure id="figure-v518hplmailsytymwy"><img id="v518hplmailsytymwy" src="https://images.unsplash.com/photo-1599493758267-c6c884c7071f?crop=entropy&amp;cs=tinysrgb&amp;fit=max&amp;fm=jpg&amp;ixid=M3wyNjI5NjF8MHwxfHNlYXJjaHwxfHxtZWRpY2FsJTIwdGVzdGluZ3xlbnwwfHx8fDE3MDg3MDM1MjV8MA&amp;ixlib=rb-4.0.3&amp;q=80&amp;w=1080" alt="medical testing" class="object-cover bg-gray-100 rounded-sm md:rounded-md lg:rounded-lg" style="aspect-ratio: 16/9;" data-mce-src="https://images.unsplash.com/photo-1599493758267-c6c884c7071f?crop=entropy&amp;cs=tinysrgb&amp;fit=max&amp;fm=jpg&amp;ixid=M3wyNjI5NjF8MHwxfHNlYXJjaHwxfHxtZWRpY2FsJTIwdGVzdGluZ3xlbnwwfHx8fDE3MDg3MDM1MjV8MA&amp;ixlib=rb-4.0.3&amp;q=80&amp;w=1080" data-mce-style="aspect-ratio: 16/9;"></figure><h3>Evaluating Effectiveness</h3>In addition to safety, product testing also evaluates the effectiveness of medical products. It determines whether the product delivers the intended results and meets the expectations of healthcare professionals and patients. By testing the product in real-world scenarios, manufacturers can gather valuable data on its performance and make necessary improvements if needed.<h2>The Role of Regulatory Authorities</h2>Regulatory authorities, such as the Food and Drug Administration (FDA) in the United States, play a crucial role in ensuring the safety and efficacy of medical products. These authorities set stringent guidelines and regulations that manufacturers must comply with before their products can be approved for use.<h3>Preclinical Testing</h3>Before a medical product can be tested on humans, it undergoes preclinical testing. This phase involves laboratory and animal studies to assess the product's safety and effectiveness. It helps identify any potential risks or side effects before moving on to human trials.<figure id="figure-3j8s7oppelsytyoms"><img id="3j8s7oppelsytyoms" src="https://images.unsplash.com/photo-1583912267382-49a82d19bd94?crop=entropy&amp;cs=tinysrgb&amp;fit=max&amp;fm=jpg&amp;ixid=M3wyNjI5NjF8MHwxfHNlYXJjaHwxOHx8cHJlY2xpbmljYWwlMjB0ZXN0aW5nfGVufDB8fHx8MTcwODcwMzUyN3ww&amp;ixlib=rb-4.0.3&amp;q=80&amp;w=1080" alt="preclinical testing" class="object-cover bg-gray-100 rounded-sm md:rounded-md lg:rounded-lg" style="aspect-ratio: 16/9;" data-mce-src="https://images.unsplash.com/photo-1583912267382-49a82d19bd94?crop=entropy&amp;cs=tinysrgb&amp;fit=max&amp;fm=jpg&amp;ixid=M3wyNjI5NjF8MHwxfHNlYXJjaHwxOHx8cHJlY2xpbmljYWwlMjB0ZXN0aW5nfGVufDB8fHx8MTcwODcwMzUyN3ww&amp;ixlib=rb-4.0.3&amp;q=80&amp;w=1080" data-mce-style="aspect-ratio: 16/9;"></figure><h3>Clinical Trials</h3>Clinical trials are the final stage of product testing and involve testing the product on human subjects. These trials are conducted in multiple phases, each with a specific objective. They gather data on the product's safety, dosage, effectiveness, and any potential side effects. Only after successful completion of clinical trials can a medical product be approved for use.<h2>Ensuring Quality Control</h2>Product testing also plays a crucial role in maintaining quality control in the medical industry. By subjecting products to rigorous testing, manufacturers can identify and rectify any manufacturing or design flaws that could impact the product's performance or reliability.<h3>Post-Market Surveillance</h3>Even after a medical product is approved and available on the market, product testing continues through post-market surveillance. This involves monitoring the product's performance, safety, and any potential adverse events reported by healthcare professionals or patients. It helps identify any unforeseen risks or issues that may arise over time.<figure id="figure-hb3mwkm948lsytypzf"><img id="hb3mwkm948lsytypzf" src="https://images.unsplash.com/photo-1567613747256-9f97205d23d2?crop=entropy&amp;cs=tinysrgb&amp;fit=max&amp;fm=jpg&amp;ixid=M3wyNjI5NjF8MHwxfHNlYXJjaHwzfHxxdWFsaXR5JTIwY29udHJvbHxlbnwwfHx8fDE3MDg3MDM1Mjl8MA&amp;ixlib=rb-4.0.3&amp;q=80&amp;w=1080" alt="quality control" class="object-cover bg-gray-100 rounded-sm md:rounded-md lg:rounded-lg" style="aspect-ratio: 16/9;" data-mce-src="https://images.unsplash.com/photo-1567613747256-9f97205d23d2?crop=entropy&amp;cs=tinysrgb&amp;fit=max&amp;fm=jpg&amp;ixid=M3wyNjI5NjF8MHwxfHNlYXJjaHwzfHxxdWFsaXR5JTIwY29udHJvbHxlbnwwfHx8fDE3MDg3MDM1Mjl8MA&amp;ixlib=rb-4.0.3&amp;q=80&amp;w=1080" data-mce-style="aspect-ratio: 16/9;"></figure><h2>Conclusion</h2>Product testing is an essential aspect of the medical industry to ensure the safety, efficacy, and quality of medical products. It plays a vital role in protecting patient safety, evaluating effectiveness, and maintaining quality control. By adhering to strict testing protocols and regulations, manufacturers can develop and deliver products that meet the highest standards of safety and efficacy.

Blog

AlphaFold 3 predicts the structure and interactions of al...

Revolutionising Healthcare: Harnessing Generative AI to T...

Study Finds ChatGPT Shows Promise as Medication Managemen...

Google AI has better bedside manner than human doctors — ...

AI in Healthcare: Who Pays When Things Go Wrong?

Ensuring Safety and Efficacy: The Vital Role of Product T...