0
selected
-
1.
Optimizing warfarin dosing for patients with atrial fibrillation using machine learning.
Petch, J, Nelson, W, Wu, M, Ghassemi, M, Benz, A, Fatemi, M, Di, S, Carnicelli, A, Granger, C, Giugliano, R, et al
Scientific reports. 2024;(1):4516
Abstract
While novel oral anticoagulants are increasingly used to reduce risk of stroke in patients with atrial fibrillation, vitamin K antagonists such as warfarin continue to be used extensively for stroke prevention across the world. While effective in reducing the risk of strokes, the complex pharmacodynamics of warfarin make it difficult to use clinically, with many patients experiencing under- and/or over- anticoagulation. In this study we employed a novel implementation of deep reinforcement learning to provide clinical decision support to optimize time in therapeutic International Normalized Ratio (INR) range. We used a novel semi-Markov decision process formulation of the Batch-Constrained deep Q-learning algorithm to develop a reinforcement learning model to dynamically recommend optimal warfarin dosing to achieve INR of 2.0-3.0 for patients with atrial fibrillation. The model was developed using data from 22,502 patients in the warfarin treated groups of the pivotal randomized clinical trials of edoxaban (ENGAGE AF-TIMI 48), apixaban (ARISTOTLE) and rivaroxaban (ROCKET AF). The model was externally validated on data from 5730 warfarin-treated patients in a fourth trial of dabigatran (RE-LY) using multilevel regression models to estimate the relationship between center-level algorithm consistent dosing, time in therapeutic INR range (TTR), and a composite clinical outcome of stroke, systemic embolism or major hemorrhage. External validation showed a positive association between center-level algorithm-consistent dosing and TTR (R2 = 0.56). Each 10% increase in algorithm-consistent dosing at the center level independently predicted a 6.78% improvement in TTR (95% CI 6.29, 7.28; p < 0.001) and a 11% decrease in the composite clinical outcome (HR 0.89; 95% CI 0.81, 1.00; p = 0.015). These results were comparable to those of a rules-based clinical algorithm used for benchmarking, for which each 10% increase in algorithm-consistent dosing independently predicted a 6.10% increase in TTR (95% CI 5.67, 6.54, p < 0.001) and a 10% decrease in the composite outcome (HR 0.90; 95% CI 0.83, 0.98, p = 0.018). Our findings suggest that a deep reinforcement learning algorithm can optimize time in therapeutic range for patients taking warfarin. A digital clinical decision support system to promote algorithm-consistent warfarin dosing could optimize time in therapeutic range and improve clinical outcomes in atrial fibrillation globally.
-
2.
Cardiovascular Risk Stratification in Diabetic Retinopathy via Atherosclerotic Pathway in COVID-19/Non-COVID-19 Frameworks Using Artificial Intelligence Paradigm: A Narrative Review.
Munjral, S, Maindarkar, M, Ahluwalia, P, Puvvula, A, Jamthikar, A, Jujaray, T, Suri, N, Paul, S, Pathak, R, Saba, L, et al
Diagnostics (Basel, Switzerland). 2022;(5)
Abstract
Diabetes is one of the main causes of the rising cases of blindness in adults. This microvascular complication of diabetes is termed diabetic retinopathy (DR) and is associated with an expanding risk of cardiovascular events in diabetes patients. DR, in its various forms, is seen to be a powerful indicator of atherosclerosis. Further, the macrovascular complication of diabetes leads to coronary artery disease (CAD). Thus, the timely identification of cardiovascular disease (CVD) complications in DR patients is of utmost importance. Since CAD risk assessment is expensive for low-income countries, it is important to look for surrogate biomarkers for risk stratification of CVD in DR patients. Due to the common genetic makeup between the coronary and carotid arteries, low-cost, high-resolution imaging such as carotid B-mode ultrasound (US) can be used for arterial tissue characterization and risk stratification in DR patients. The advent of artificial intelligence (AI) techniques has facilitated the handling of large cohorts in a big data framework to identify atherosclerotic plaque features in arterial ultrasound. This enables timely CVD risk assessment and risk stratification of patients with DR. Thus, this review focuses on understanding the pathophysiology of DR, retinal and CAD imaging, the role of surrogate markers for CVD, and finally, the CVD risk stratification of DR patients. The review shows a step-by-step cyclic activity of how diabetes and atherosclerotic disease cause DR, leading to the worsening of CVD. We propose a solution to how AI can help in the identification of CVD risk. Lastly, we analyze the role of DR/CVD in the COVID-19 framework.
-
3.
Nutrition, atherosclerosis, arterial imaging, cardiovascular risk stratification, and manifestations in COVID-19 framework: a narrative review.
Munjral, S, Ahluwalia, P, Jamthikar, AD, Puvvula, A, Saba, L, Faa, G, Singh, IM, Chadha, PS, Turk, M, Johri, AM, et al
Frontiers in bioscience (Landmark edition). 2021;(11):1312-1339
Abstract
Background: Atherosclerosis is the primary cause of the cardiovascular disease (CVD). Several risk factors lead to atherosclerosis, and altered nutrition is one among those. Nutrition has been ignored quite often in the process of CVD risk assessment. Altered nutrition along with carotid ultrasound imaging-driven atherosclerotic plaque features can help in understanding and banishing the problems associated with the late diagnosis of CVD. Artificial intelligence (AI) is another promisingly adopted technology for CVD risk assessment and management. Therefore, we hypothesize that the risk of atherosclerotic CVD can be accurately monitored using carotid ultrasound imaging, predicted using AI-based algorithms, and reduced with the help of proper nutrition. Layout: The review presents a pathophysiological link between nutrition and atherosclerosis by gaining a deep insight into the processes involved at each stage of plaque development. After targeting the causes and finding out results by low-cost, user-friendly, ultrasound-based arterial imaging, it is important to (i) stratify the risks and (ii) monitor them by measuring plaque burden and computing risk score as part of the preventive framework. Artificial intelligence (AI)-based strategies are used to provide efficient CVD risk assessments. Finally, the review presents the role of AI for CVD risk assessment during COVID-19. Conclusions: By studying the mechanism of low-density lipoprotein formation, saturated and trans fat, and other dietary components that lead to plaque formation, we demonstrate the use of CVD risk assessment due to nutrition and atherosclerosis disease formation during normal and COVID times. Further, nutrition if included, as a part of the associated risk factors can benefit from atherosclerotic disease progression and its management using AI-based CVD risk assessment.
-
4.
Vibro-magnetometry: theoretical aspects and simulations.
Carneiro, AO, Baffa, O, Silva, GT, Fatemi, M
IEEE transactions on ultrasonics, ferroelectrics, and frequency control. 2009;(5):1065-73
Abstract
In this article, we introduce the theory for vibro-magnetometry (VM) with computational simulations for an idealized experimental setup. A method based on acoustic radiation force and magnetic measurement has been proposed for interrogating the mechanical vibrations of a target immersed in fluid medium. In this method, ultrasound radiation is used to exert a low-frequency (in kHz range) force on a rigid magnetized target immersed in a viscoelastic medium. In response, the target vibrates sinusoidally in a pattern determined by viscoelastic properties of the medium. The magnetic field resulting from target vibration is related to both the ultrasonic and low-frequency (kHz range) mechanical characteristics of the medium. We report the relationship between the magnetic field signal and the incident ultrasonic pressure field in terms of the mechanical parameters of the medium. Simulations were conducted to demonstrate a simple approach based on using amplitude-modulated ultrasound to generate a dynamic acoustic radiation force on a magnetic target. The magnetic field generated by vibration of this target is then obtained and used to estimate the radiation-force-induced displacement as a function of time. It was observed that the intensity of the dynamic component of the magnetic field caused by the acoustic excitation is high enough to be registered by a conventional magnetic sensor. When a low stress is applied on a reactive magnetic target embedded in the medium, the subsequent oscillating magnetic field is measured by a dedicated magnetic sensor, yielding the applicable mechanical information of the host medium. The proposed methodology presents a powerful tool for evaluation of acoustic radiation force as well as the mechanical properties of soft materials.