Medical forecasting.

"AI-Powered Forecasting" was recently on the cover of Science, highlighting a new deep learning model for much faster and more accurate weather forecasting. Known as GraphCast, it outperformed the gold-standard system and had an accuracy of 99.7% for tropospheric predictions, the most important forecasting region that is closest to Earth's surface. Better warnings for extreme weather events such as hurricanes and cyclones will help save lives. The parallel in medicine is forecasting specific, actionable, high risk for individuals to prevent diseases or severe acute events. But we don't have a gold standard for predicting health outcomes. That is hopefully about to change.

Three-year outcomes of post-acute sequelae of COVID-19.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection causes post-acute sequelae of coronavirus disease 2019 (COVID-19) (PASC) in many organ systems. Risks of these sequelae have been characterized up to 2 years after infection, but longer-term follow-up is limited. Here we built a cohort of 135,161 people with SARS-CoV-2 infection and 5,206,835 controls from the US Department of Veterans Affairs who were followed for 3 years to estimate risks of death and PASC. Among non-hospitalized individuals, the increased risk of death was no longer present after the first year of infection, and risk of incident PASC declined over the 3 years but still contributed 9.6 (95% confidence interval (CI): 0.4-18.7) disability-adjusted life years (DALYs) per 1,000 persons in the third year. Among hospitalized individuals, risk of death declined but remained significantly elevated in the third year after infection (incidence rate ratio: 1.29 (95% CI: 1.19-1.40)). Risk of incident PASC declined over the 3 years, but substantial residual risk remained in the third year, leading to 90.0 (95% CI: 55.2-124.8) DALYs per 1,000 persons. Altogether, our findings show reduction of risks over time, but the burden of mortality and health loss remains in the third year among hospitalized individuals.

Artificial intelligence in surgery.

Artificial intelligence (AI) is rapidly emerging in healthcare, yet applications in surgery remain relatively nascent. Here we review the integration of AI in the field of surgery, centering our discussion on multifaceted improvements in surgical care in the preoperative, intraoperative and postoperative space. The emergence of foundation model architectures, wearable technologies and improving surgical data infrastructures is enabling rapid advances in AI interventions and utility. We discuss how maturing AI methods hold the potential to improve patient outcomes, facilitate surgical education and optimize surgical care. We review the current applications of deep learning approaches and outline a vision for future advances through multimodal foundation models.

Randomised controlled trials evaluating artificial intelligence in clinical practice: a scoping review.

This scoping review of randomised controlled trials on artificial intelligence (AI) in clinical practice reveals an expanding interest in AI across clinical specialties and locations. The USA and China are leading in the number of trials, with a focus on deep learning systems for medical imaging, particularly in gastroenterology and radiology. A majority of trials (70 [81%] of 86) report positive primary endpoints, primarily related to diagnostic yield or performance; however, the predominance of single-centre trials, little demographic reporting, and varying reports of operational efficiency raise concerns about the generalisability and practicality of these results. Despite the promising outcomes, considering the likelihood of publication bias and the need for more comprehensive research including multicentre trials, diverse outcome measures, and improved reporting standards is crucial. Future AI trials should prioritise patient-relevant outcomes to fully understand AI's true effects and limitations in health care.

Transforming the cardiometabolic disease landscape: Multimodal AI-powered approaches in prevention and management.

The rise of artificial intelligence (AI) has revolutionized various scientific fields, particularly in medicine, where it has enabled the modeling of complex relationships from massive datasets. Initially, AI algorithms focused on improved interpretation of diagnostic studies such as chest X-rays and electrocardiograms in addition to predicting patient outcomes and future disease onset. However, AI has evolved with the introduction of transformer models, allowing analysis of the diverse, multimodal data sources existing in medicine today. Multimodal AI holds great promise in more accurate disease risk assessment and stratification as well as optimizing the key driving factors in cardiometabolic disease: blood pressure, sleep, stress, glucose control, weight, nutrition, and physical activity. In this article we outline the current state of medical AI in cardiometabolic disease, highlighting the potential of multimodal AI to augment personalized prevention and treatment strategies in cardiometabolic disease.

The hospital at home in the USA: current status and future prospects.

The annual cost of hospital care services in the US has risen to over $1 trillion despite relatively worse health outcomes compared to similar nations. These trends accentuate a growing need for innovative care delivery models that reduce costs and improve outcomes. HaH-a program that provides patients acute-level hospital care at home-has made significant progress over the past two decades. Technological advancements in remote patient monitoring, wearable sensors, health information technology infrastructure, and multimodal health data processing have contributed to its rise across hospitals. More recently, the COVID-19 pandemic brought HaH into the mainstream, especially in the US, with reimbursement waivers that made the model financially acceptable for hospitals and payors. However, HaH continues to face serious challenges to gain widespread adoption. In this review, we evaluate the peer-reviewed evidence and discuss the promises, challenges, and what it would take to tap into the future potential of HaH.

AI-Enhanced Reconstruction of the 12-Lead Electrocardiogram via 3-Leads with Accurate Clinical Assessment.

The 12-lead electrocardiogram (ECG) is an integral component to the diagnosis of a multitude of cardiovascular conditions. It is performed using a complex set of skin surface electrodes, limiting its use outside traditional clinical settings. We developed an artificial intelligence algorithm, trained over 600,000 clinically acquired ECGs, to explore whether fewer leads as input are sufficient to reconstruct a full 12-lead ECG. Two limb leads (I and II) and one precordial lead (V3) were required to generate a reconstructed synthetic 12-lead ECG highly correlated with the original ECG. An automatic algorithm for detection of acute myocardial infarction (MI) performed similarly for original and reconstructed ECGs (AUC=0.94). When interpreted by cardiologists, reconstructed ECGs achieved an accuracy of 81.4±5.0% in identifying ST elevation MI, comparable with the original 12-lead ECGs (accuracy 84.6±4.6%). These results will impact development efforts to innovate ECG acquisition methods with simplified tools in non-specialized settings.

Toward the eradication of medical diagnostic errors.

The medical community does not broadcast the problem, but there are many studies that have reinforced a serious issue with diagnostic errors. A recent study concluded: "We estimate that nearly 800,000 Americans die or are permanently disabled by diagnostic errors each year." Diagnostic errors are inaccurate assessments of a patient's root cause of illness, such as missing a heart attack or infection or assigning the wrong diagnosis of pneumonia when the correct one is pulmonary embolism. Despite ever-increasing use of medical imaging and laboratory tests intended to promote diagnostic accuracy, there is nothing to suggest improvement since the report by the National Academies of Sciences, Engineering and Medicine in 2015, which provided a conservative estimate that 5% of adults experience a diagnostic error each year, and that most people will experience at least one in their lifetime.

A foundation model for generalizable disease detection from retinal images.

Medical artificial intelligence (AI) offers great potential for recognizing signs of health conditions in retinal images and expediting the diagnosis of eye diseases and systemic disorders1. However, the development of AI models requires substantial annotation and models are usually task-specific with limited generalizability to different clinical applications2. Here, we present RETFound, a foundation model for retinal images that learns generalizable representations from unlabelled retinal images and provides a basis for label-efficient model adaptation in several applications. Specifically, RETFound is trained on 1.6 million unlabelled retinal images by means of self-supervised learning and then adapted to disease detection tasks with explicit labels. We show that adapted RETFound consistently outperforms several comparison models in the diagnosis and prognosis of sight-threatening eye diseases, as well as incident prediction of complex systemic disorders such as heart failure and myocardial infarction with fewer labelled data. RETFound provides a generalizable solution to improve model performance and alleviate the annotation workload of experts to enable broad clinical AI applications from retinal imaging.

As artificial intelligence goes multimodal, medical applications multiply.

Machines don't have eyes, but you wouldn't know that if you followed the progression of deep learning models for accurate interpretation of medical images, such as x-rays, computed tomography (CT) and magnetic resonance imaging (MRI) scans, pathology slides, and retinal photos. Over the past several years, there has been a torrent of studies that have consistently demonstrated how powerful "machine eyes" can be, not only compared with medical experts but also for detecting features in medical images that are not readily discernable by humans. For example, a retinal scan is rich with information that people can't see, but machines can, providing a gateway to multiple aspects of human physiology, including blood pressure; glucose control; risk of Parkinson's, Alzheimer's, kidney, and hepatobiliary diseases; and the likelihood of heart attacks and strokes. As a cardiologist, I would not have envisioned that machine interpretation of an electrocardiogram would provide information about the individual's age, sex, anemia, risk of developing diabetes or arrhythmias, heart function and valve disease, kidney, or thyroid conditions. Likewise, applying deep learning to a pathology slide of tumor tissue can also provide insight about the site of origin, driver mutations, structural genomic variants, and prognosis. Although these machine vision capabilities for medical image interpretation may seem impressive, they foreshadow what is potentially far more expansive terrain for artificial intelligence (AI) to transform medicine. The big shift ahead is the ability to transcend narrow, unimodal tasks, confined to images, and broaden machine capabilities to include text and speech, encompassing all input modes, setting the foundation for multimodal AI.

Retinal Optical Coherence Tomography Features Associated With Incident and Prevalent Parkinson Disease.

BACKGROUND AND OBJECTIVES: Cadaveric studies have shown disease-related neurodegeneration and other morphological abnormalities in the retina of individuals with Parkinson disease (PD); however, it remains unclear whether this can be reliably detected with in vivo imaging. We investigated inner retinal anatomy, measured using optical coherence tomography (OCT), in prevalent PD and subsequently assessed the association of these markers with the development of PD using a prospective research cohort. METHODS: This cross-sectional analysis used data from 2 studies. For the detection of retinal markers in prevalent PD, we used data from AlzEye, a retrospective cohort of 154,830 patients aged 40 years and older attending secondary care ophthalmic hospitals in London, United Kingdom, between 2008 and 2018. For the evaluation of retinal markers in incident PD, we used data from UK Biobank, a prospective population-based cohort where 67,311 volunteers aged 40-69 years were recruited between 2006 and 2010 and underwent retinal imaging. Macular retinal nerve fiber layer (mRNFL), ganglion cell-inner plexiform layer (GCIPL), and inner nuclear layer (INL) thicknesses were extracted from fovea-centered OCT. Linear mixed-effects models were fitted to examine the association between prevalent PD and retinal thicknesses. Hazard ratios for the association between time to PD diagnosis and retinal thicknesses were estimated using frailty models. RESULTS: Within the AlzEye cohort, there were 700 individuals with prevalent PD and 105,770 controls (mean age 65.5 ± 13.5 years, 51.7% female). Individuals with prevalent PD had thinner GCIPL (-2.12 µm, 95% CI -3.17 to -1.07, p = 8.2 × 10-5) and INL (-0.99 µm, 95% CI -1.52 to -0.47, p = 2.1 × 10-4). The UK Biobank included 50,405 participants (mean age 56.1 ± 8.2 years, 54.7% female), of whom 53 developed PD at a mean of 2,653 ± 851 days. Thinner GCIPL (hazard ratio [HR] 0.62 per SD increase, 95% CI 0.46-0.84, p = 0.002) and thinner INL (HR 0.70, 95% CI 0.51-0.96, p = 0.026) were also associated with incident PD. DISCUSSION: Individuals with PD have reduced thickness of the INL and GCIPL of the retina. Involvement of these layers several years before clinical presentation highlight a potential role for retinal imaging for at-risk stratification of PD.

The imperative for regulatory oversight of large language models (or generative AI) in healthcare.

The rapid advancements in artificial intelligence (AI) have led to the development of sophisticated large language models (LLMs) such as GPT-4 and Bard. The potential implementation of LLMs in healthcare settings has already garnered considerable attention because of their diverse applications that include facilitating clinical documentation, obtaining insurance pre-authorization, summarizing research papers, or working as a chatbot to answer questions for patients about their specific data and concerns. While offering transformative potential, LLMs warrant a very cautious approach since these models are trained differently from AI-based medical technologies that are regulated already, especially within the critical context of caring for patients. The newest version, GPT-4, that was released in March, 2023, brings the potentials of this technology to support multiple medical tasks; and risks from mishandling results it provides to varying reliability to a new level. Besides being an advanced LLM, it will be able to read texts on images and analyze the context of those images. The regulation of GPT-4 and generative AI in medicine and healthcare without damaging their exciting and transformative potential is a timely and critical challenge to ensure safety, maintain ethical standards, and protect patient privacy. We argue that regulatory oversight should assure medical professionals and patients can use LLMs without causing harm or compromising their data or privacy. This paper summarizes our practical recommendations for what we can expect from regulators to bring this vision to reality.