Nature medicine
-
Asymptomatic left ventricular dysfunction (ALVD) is present in 3-6% of the general population, is associated with reduced quality of life and longevity, and is treatable when found1-4. An inexpensive, noninvasive screening tool for ALVD in the doctor's office is not available. We tested the hypothesis that application of artificial intelligence (AI) to the electrocardiogram (ECG), a routine method of measuring the heart's electrical activity, could identify ALVD. ⋯ When tested on an independent set of 52,870 patients, the network model yielded values for the area under the curve, sensitivity, specificity, and accuracy of 0.93, 86.3%, 85.7%, and 85.7%, respectively. In patients without ventricular dysfunction, those with a positive AI screen were at 4 times the risk (hazard ratio, 4.1; 95% confidence interval, 3.3 to 5.0) of developing future ventricular dysfunction compared with those with a negative screen. Application of AI to the ECG-a ubiquitous, low-cost test-permits the ECG to serve as a powerful screening tool in asymptomatic individuals to identify ALVD.
-
Here we present deep-learning techniques for healthcare, centering our discussion on deep learning in computer vision, natural language processing, reinforcement learning, and generalized methods. We describe how these computational techniques can impact a few key areas of medicine and explore how to build end-to-end systems. Our discussion of computer vision focuses largely on medical imaging, and we describe the application of natural language processing to domains such as electronic health record data. Similarly, reinforcement learning is discussed in the context of robotic-assisted surgery, and generalized deep-learning methods for genomics are reviewed.
-
Big data has become the ubiquitous watch word of medical innovation. The rapid development of machine-learning techniques and artificial intelligence in particular has promised to revolutionize medical practice from the allocation of resources to the diagnosis of complex diseases. ⋯ We discuss, among other topics, how best to conceive of health privacy; the importance of equity, consent, and patient governance in data collection; discrimination in data uses; and how to handle data breaches. We close by sketching possible ways forward for the regulatory system.
-
Randomized Controlled Trial
A patient-level pooled analysis of treatment-shortening regimens for drug-susceptible pulmonary tuberculosis.
Tuberculosis kills more people than any other infectious disease. Three pivotal trials testing 4-month regimens failed to meet non-inferiority margins; however, approximately four-fifths of participants were cured. Through a pooled analysis of patient-level data with external validation, we identify populations eligible for 4-month treatment, define phenotypes that are hard to treat and evaluate the impact of adherence and dosing strategy on outcomes. ⋯ Four-month regimens were non-inferior in participants with minimal disease defined by <2+ sputum smear grade or non-cavitary disease. A hard-to-treat phenotype, defined by high smear grades and cavitation, may require durations >6 months to cure all. Regimen duration can be selected in order to improve outcomes, providing a stratified medicine approach as an alternative to the 'one-size-fits-all' treatment currently used worldwide.
-
Chimeric antigen receptor (CAR) T cells targeting CD19 mediate potent effects in relapsed and/or refractory pre-B cell acute lymphoblastic leukemia (B-ALL), but antigen loss is a frequent cause of resistance to CD19-targeted immunotherapy. CD22 is also expressed in most cases of B-ALL and is usually retained following CD19 loss. We report results from a phase 1 trial testing a new CD22-targeted CAR (CD22-CAR) in 21 children and adults, including 17 who were previously treated with CD19-directed immunotherapy. ⋯ Relapses were associated with diminished CD22 site density that likely permitted CD22+ cell escape from killing by CD22-CAR T cells. These results are the first to establish the clinical activity of a CD22-CAR in B-ALL, including leukemia resistant to anti-CD19 immunotherapy, demonstrating potency against B-ALL comparable to that of CD19-CAR at biologically active doses. Our results also highlight the critical role played by antigen density in regulating CAR function.