AI in health care

Artificial intelligence (AI) is advancing rapidly and is already starting to transform cancer research and care. Here, the authors outline how AI could be incorporated into liver cancer management, highlighting areas with academic, commercial and clinical potential, as well as ongoing progress and pitfalls.

Increase in clinical digital data is propelling the development and application of artificial intelligence methods in histopathology. In this Review, machine learning algorithms and models and their clinical use cases are discussed, highlighting the computational and operational challenges in the field.

Deep learning is a powerful technique with great potential for the analysis and interpretation of rheumatological images. To successfully use deep learning, rheumatologists should understand the tasks involved in image processing and the potential confounders and limitations that can affect the analysis of clinical data.

In this Roadmap, Föllmer et al. summarize the evidence for the application of artificial intelligence (AI) technology to the imaging of vulnerable plaques in coronary arteries and discuss the current and future approaches to addressing the limitations of AI-guided coronary plaque imaging, such as bias, uncertainty and generalizability.

In this Perspective, the authors highlight the ethical challenges of adopting artificial intelligence (AI) in urology and its influence on daily practice. Ethical principles for the application of AI in health care and urology are proposed to improve and oversee the use of such technologies.

Advances in artificial intelligence (AI) are changing endoscopy and gastrointestinal surgery, including computer-assisted detection and diagnosis, computer-aided navigation, robot-assisted intervention and automated reporting. This Perspective introduces the role of AI in computer-assisted interventions in gastroenterology with insights on regulatory aspects and the challenges ahead.

Advances in computational omics technologies are enabling access to the hidden diversity of natural products, and artificial intelligence approaches are facilitating key steps in harnessing the therapeutic potential of such compounds, including biological activity prediction. This article discusses synergies between these fields to effectively identify drug candidates from the plethora of molecules produced by nature, and how to address the challenges in realizing the potential of these synergies.

Emerging evidence suggests that artificial intelligence (AI)-enabled decision support systems may have an important role in the diagnosis, prognosis and treatment of kidney diseases. This Review provides an overview of AI fundamentals and the state of the art of AI-enabled decision support systems in nephrology.

In this Review, Friedman and colleagues summarize the use of artificial intelligence-enhanced electrocardiography in the detection of cardiovascular disease in at-risk populations, discuss its implications for clinical decision-making in patients with cardiovascular disease and critically appraise potential limitations and unknowns.

In this Review, Sermesant and colleagues discuss the applications of artificial intelligence (AI) in cardiovascular imaging and explain how pre-existing clinical knowledge can be included in AI methods to increase robustness. They also discuss the limitations of AI approaches in cardiovascular imaging and how they can be overcome.

In this Review, the authors discuss how machine learning can improve sperm selection, focusing on the near-term, achievable machine learning methods based on available assessments of selected sperm and the areas in which the most considerable gains are possible.

This Review explains core concepts in artificial intelligence (AI) and machine learning for endocrinologists. AI applications in endocrine cancer diagnostics are highlighted as well as research challenges and future directions for the field.

In this Review, the authors introduce the application of artificial intelligence, in particular machine learning and deep learning, for improving multiple stages of MRI, including acquisition, processing and post-processing steps, for studying osteoarthritis.

In this Review, the authors provide an introduction to machine learning and discuss the use of this approach in rheumatic autoimmune inflammatory diseases, including the classification of patients based on medical records or molecular characteristics, identification of novel biomarkers or drug repurposing candidates and prediction of disease progression or onset.

Prognostication of outcome across multiple cancers and prediction of response to various treatment modalities are among the next generation of challenges that artificial intelligence (AI) tools can solve using radiology images. The authors of this Perspective describe the evolution of AI-based approaches in oncology imaging and address the path to their adoption as decision-support tools in the clinic.

Advances in cardiovascular monitoring technologies have resulted in an influx of consumer-targeted wearable sensors that have the potential to detect numerous heart conditions. In this Review, Krittanawong and colleagues describe processes involved in biosignal acquisition and analysis of cardiovascular monitors, as well as their associated ethical, regulatory and legal challenges.

Several big data ‘omics’ studies have analysed hepatocellular carcinoma (HCC). This Review describes omics studies in HCC and their potential in drug discovery and as candidate biomarkers. The application of emerging new artificial intelligence methods in HCC drug discovery is also discussed.

Large-scale datasets of increasing size and complexity are being produced in the microbiome and oncology field. This Perspective discusses the potential to harness gut microbiome analysis, big data and machine learning in cancer, and the potential and limitations with this approach.

In this Review, the authors describe the latest developments in the use of machine learning to interrogate neurodegenerative disease-related datasets. They discuss applications of machine learning to diagnosis, prognosis and therapeutic development, and the challenges involved in analysing health-care data.

Imaging-based assessment is becoming increasingly important in the management of acute stroke, but processing and interpretation of images in clinical practice is challenging. In this Review, Parsons and colleagues explore the potential of artificial intelligence to provide treatment decision support.

The possible uses of artificial intelligence (AI) in radiation oncology are diverse and wide ranging. Herein, the authors discuss the potential applications of AI at each step of the radiation oncology workflow, which might improve the efficiency and overall quality of radiation therapy for patients with cancer. The authors also describe the associated challenges and provide their perspective on how AI platforms might change the roles of radiation oncology medical professionals.

Machine learning (ML) is revolutionizing and reshaping health care, and computer-based systems can be trained to perform complex tasks in bioinformatics, medical imaging and medical robotics. In this Review, Goldenberg et al. consider ML in the management of prostate cancer, with growing applications in diagnostic imaging, surgical interventions, skills training and assessment, digital pathology and genomics.

Machine learning has been applied to numerous stages in the drug discovery pipeline. Here, Vamathevan and colleagues discuss the most useful techniques and how machine learning can promote data-driven decision making in drug discovery and development. They highlight major hurdles in the field, such as the required data characteristics for applying machine learning, which will need to be solved as machine learning matures.

Artificial intelligence (AI) tools are increasingly being applied in drug discovery. This article presents the views of a group of international experts on the ‘grand challenges’ in small-molecule drug discovery with AI, including obtaining appropriate data sets, generating new hypotheses, optimizing in a multi-objective manner, reducing cycle times and changing the research culture.

In 2023, large language models demonstrated potential for use in rheumatology to accurately suggest diagnoses and provide empathetic patient education. However, the propensity of this technology to generate misleading information continues to pose risks. Balancing innovation with physician guidance is essential.

The next generation of artificial intelligence (AI)-enabled nephrology will leverage generalist models that link diverse multimodal patient data with the linguistic and emergent capabilities of large language models. In 2023, advances in AI that linked novel unstructured data with physiological and clinical characteristics moved the field closer to realizing this vision.

Conversational agents (CAs) are computer programs designed to engage in human-like conversations with users. They are increasingly used in digital health applications, for example medical history taking. CAs have potential to facilitate health-care processes when designed carefully, considering quality aspects and are integrated into health-care processes.

In gastroenterology, ChatGPT and large language models (LLMs) can assist clinicians in various tasks but also have several shortcomings. Although LLMs have great potential to assist clinicians in health care, they should be used as a tool to support, rather than replace, human expertise.

Artificial intelligence (AI) has rapidly become one of the most important and transformative technologies of our time, with applications in virtually every field and industry. Among these applications, academic writing is one of the areas that has experienced perhaps the most rapid development and uptake of AI-based tools and methodologies. We argue that use of AI-based tools for scientific writing should widely be adopted.

The COVID-19 pandemic has disrupted surgical training worldwide, and reconstructive urology training has been neglected at the expense of more urgent life-saving procedures. To help address this problem, virtual reality must become a fundamental training aid in modern reconstructive urology surgery education.

Artificial intelligence has already revolutionized various fields in medicine and research. Due to the complex and interconnected nature of the endocrine system, it is an ideal area to further exploit and maximize the potential benefits of artificial intelligence.

An artificial intelligence-based tool can turn low-resolution clinical MRI scans into high-resolution 3D objects suitable for research studies. The new approach opens up the possibility of secondary analysis of large clinical MRI datasets to answer disease-relevant questions, although further work to automate scan annotation will be required.

Deep brain stimulation (DBS) is a well-established approach for treating movement disorders such as Parkinson disease, dystonia and essential tremor. However, the outcomes are variable, and researchers are now exploring artificial intelligence-based strategies to help improve DBS procedures.

Artificial intelligence has emerged as a powerful tool for predicting protein structure. This technology is now being applied to improve our understanding of protein aggregation in neurodegenerative and other neurological disorders, and could potentially improve disease management by enabling precision medicine.

Artificial intelligence-based tools have the potential to transform health care, enabling faster and more accurate diagnosis, personalized treatment plans, new therapeutic approaches and effective disease monitoring. Artificial intelligence shows particular promise for the management of rare neurological disorders by augmenting knowledge and facilitating the sharing of expertise among physicians.

Electronic health records (EHRs) contain enormous amounts of real-world data that could inform researchers, doctors and patients about many aspects of rheumatology. However, EHRs are not yet fully utilized, mainly because automatic data extraction is difficult. Several studies in 2022 highlight the feasibility and clinical utility of computer-assisted EHR analysis.

Artificial intelligence and machine learning have the potential to make cancer care more accessible, efficient, cost-effective and personalized. However, meticulously planned prospective deployment strategies are required to validate the performance of these technologies in real-world clinical settings and overcome the human trust barrier.

Deep learning can mine clinically useful information from histology. In gastrointestinal and liver cancer, such algorithms can predict survival and molecular alterations. Once pathology workflows are widely digitized, these methods could be used as inexpensive biomarkers. However, clinical translation requires training interdisciplinary researchers in both programming and clinical applications.

The ability to predict how a patient might respond to a medication would shift treatment decisions away from trial and error and reduce disease-associated health and financial burdens. Machine learning approaches applied to genomic datasets offer great promise to deliver personalized medicine but their application must first be optimized.

Machine learning and high-throughput technologies hold promise for the classification, diagnosis and treatment of patients with rheumatic diseases, with the ultimate goal of precision medicine. Several studies in 2019 highlight the feasibility and clinical utility of using machine learning in rheumatology to stratify patients and/or predict treatment responses.

Idiopathic inflammatory myopathies (IIMs) are heterogeneous conditions, and the optimal way to classify patients and divide them into subgroups remains unclear. Could machine learning techniques be the answer to the problem of defining homogeneous disease phenotypes, enabling stratified treatment approaches and the formulation of future IIM classification criteria?

Developing novel technologies to discriminate malignant tissue from nonmalignant structures and thereby facilitate safe, complete tumour resection is a major priority for advancing oncological neurosurgery. Herein, we discuss a recently reported innovation involving stimulated Raman spectroscopy of intraoperative tissue samples and data interpretation with artificial intelligence, as well as the implications of this approach for neurosurgical oncology.

Artificial intelligence (AI) has the potential to change many aspects of health-care practice. Two newly published trials explore the potential applications of AI to improve polyp detection and mucosal visualization in gastrointestinal endoscopy — both show the benefits of AI to improve detection in gastrointestinal endoscopy.

Artificial intelligence (AI) holds promise for cardiovascular medicine but is limited by a lack of large, heterogeneous and granular data sets. Blockchain provides secure interoperability between siloed stakeholders and centralized data sources. We discuss integration of blockchain with AI for data-centric analysis and information flow, its current limitations and potential cardiovascular applications.