RESUMEN
Pharmaceutical agents in oncology currently have high attrition rates from early to late phase clinical trials. Recent advances in computational methods, notably causal artificial intelligence, and availability of rich clinico-genomic databases have made it possible to simulate the efficacy of cancer drug protocols in diverse patient populations, which could inform and improve clinical trial design. Here, we review the current and potential use of in silico trials and causal AI to increase the efficacy and safety of traditional clinical trials. We conclude that in silico trials using causal AI approaches can simulate control and efficacy arms, inform patient recruitment and regimen titrations, and better enable subgroup analyses critical for precision medicine.
Asunto(s)
Antineoplásicos/uso terapéutico , Inteligencia Artificial , Ensayos Clínicos como Asunto , Simulación por Computador , Genómica , Neoplasias/tratamiento farmacológico , Medicina de Precisión , Proyectos de Investigación , Antineoplásicos/efectos adversos , Biomarcadores de Tumor/genética , Toma de Decisiones Clínicas , Humanos , Terapia Molecular Dirigida , Neoplasias/genética , Neoplasias/metabolismo , Neoplasias/patologíaRESUMEN
While the past decade has seen meaningful improvements in clinical outcomes for multiple myeloma patients, a subset of patients does not benefit from current therapeutics for unclear reasons. Many gene expression-based models of risk have been developed, but each model uses a different combination of genes and often involves assaying many genes making them difficult to implement. We organized the Multiple Myeloma DREAM Challenge, a crowdsourced effort to develop models of rapid progression in newly diagnosed myeloma patients and to benchmark these against previously published models. This effort lead to more robust predictors and found that incorporating specific demographic and clinical features improved gene expression-based models of high risk. Furthermore, post-challenge analysis identified a novel expression-based risk marker, PHF19, which has recently been found to have an important biological role in multiple myeloma. Lastly, we show that a simple four feature predictor composed of age, ISS, and expression of PHF19 and MMSET performs similarly to more complex models with many more gene expression features included.
Asunto(s)
Biomarcadores de Tumor/metabolismo , Ensayos Clínicos como Asunto/estadística & datos numéricos , Proteínas de Unión al ADN/metabolismo , Epigénesis Genética , Regulación Neoplásica de la Expresión Génica , Modelos Estadísticos , Mieloma Múltiple/patología , Factores de Transcripción/metabolismo , Biomarcadores de Tumor/genética , Ciclo Celular , Proliferación Celular , Proteínas de Unión al ADN/genética , Bases de Datos Factuales , Conjuntos de Datos como Asunto , Humanos , Mieloma Múltiple/genética , Mieloma Múltiple/metabolismo , Factores de Transcripción/genética , Células Tumorales CultivadasRESUMEN
In the past decade, low-field NMR relaxation and diffusion measurements in grossly inhomogeneous fields have been used to characterize pore size distribution of porous media. Estimation of these distributions from the measured magnetization data plays a central role in the inference of insitu petro-physical and fluid properties such as porosity, permeability, and hydrocarbon viscosity. In general, inversion of the relaxation and/or diffusion distribution from NMR data is a non-unique and ill-conditioned problem. It is often solved in the literature by finding the smoothest relaxation distribution that fits the measured data by use of regularization. In this paper, estimation of these distributions is further constrained by linear functionals of the measurement that can be directly estimated from the measured data. These linear functionals include Mellin, Fourier-Mellin, and exponential Haar transforms that provide moments, porosity, and tapered areas of the distribution, respectively. The addition of these linear constraints provides more accurate estimates of the distribution in terms of a reduction in bias and variance in the estimates. The resulting distribution is also more stable in that it is less sensitive to regularization. Benchmarking of this algorithm on simulated data sets shows a reduction of artefacts often seen in the distributions and, in some cases, there is an increase of resolution in the features of the T(2) distribution. This algorithm can be applied to data obtained from a variety of pulse sequences including CPMG, inversion and saturation recovery and diffusion editing, as well as pulse sequences often deployed down-hole.
RESUMEN
In the past decade, low-field NMR relaxation and diffusion measurements in grossly inhomogeneous fields have been used to characterize properties of porous media, e.g., porosity and permeability. Pulse sequences such as CPMG, inversion and saturation recovery as well as diffusion editing have been used to estimate distribution functions of relaxation times and diffusion. Linear functionals of these distribution functions have been used to predict petro-physical and fluid properties like permeability, viscosity, fluid typing, etc. This paper describes an analysis method using integral transforms to directly compute linear functionals of the distributions of relaxation times and diffusion without first computing the distributions from the measured magnetization data. Different linear functionals of the distribution function can be obtained by choosing appropriate kernels in the integral transforms. There are two significant advantages of this approach over the traditional algorithm involving inversion of the distribution function from the measured data. First, it is a direct linear transform of the data. Thus, in contrast to the traditional analysis which involves inversion of an ill-conditioned, non-linear problem, the estimates from this new method are more accurate. Second, the uncertainty in the linear functional can be obtained in a straight-forward manner as a function of the signal-to-noise ratio (SNR) in the measured data. We demonstrate the performance of this method on simulated data.
RESUMEN
This paper provides a theoretical basis to directly estimate moments of transverse relaxation time T(2) from measured CPMG data in grossly inhomogeneous fields. These moments
RESUMEN
This paper describes a new method for computing moments of the transverse relaxation time T(2) from measured CPMG data. This new method is based on Mellin transform of the measured data and its time-derivatives. The Mellin transform can also be used to compute the cumulant generating function of lnT(2). The moments of relaxation time T(2) and lnT(2) are related to petro-physical and fluid properties of hydrocarbons in porous media. The performance of the new algorithm is demonstrated on simulated data and compared to results from the traditional inverse Laplace transform. Analytical expressions are also derived for uncertainties in these moments in terms of the signal-to-noise ratio of the data.
Asunto(s)
Algoritmos , Espectroscopía de Resonancia Magnética/estadística & datos numéricos , Simulación por Computador , Hidrocarburos/química , Modelos Estadísticos , PorosidadRESUMEN
This paper develops, within a general framework that is applicable to rather arbitrary electromagnetic and acoustic remote sensing systems, a theory of time-reversal "MUltiple Signal Classification" (MUSIC)-based imaging of extended (nonpoint-like) scatterers (targets). The general analysis applies to arbitrary remote sensing geometry and sheds light onto how the singular system of the scattering matrix relates to the geometrical and propagation characteristics of the entire transmitter-target-receiver system and how to use this effect for imaging. All the developments are derived within exact scattering theory which includes multiple scattering effects. The derived time-reversal MUSIC methods include both interior sampling, as well as exterior sampling (or enclosure) approaches. For presentation simplicity, particular attention is given to the time-harmonic case where the informational wave modes employed for target interrogation are purely spatial, but the corresponding generalization to broadband fields is also given. This paper includes computer simulations illustrating the derived theory and algorithms.
Asunto(s)
Acústica , Algoritmos , Campos Electromagnéticos , Interpretación de Imagen Asistida por Computador/métodos , Imagenología Tridimensional/métodos , Reconocimiento de Normas Patrones Automatizadas/métodos , Inteligencia Artificial , Aumento de la Imagen/métodos , Reproducibilidad de los Resultados , Dispersión de Radiación , Sensibilidad y EspecificidadRESUMEN
This paper derives, in the exact framework of multiple scattering theory for point targets, a noniterative analytical formula for the nonlinear inversion of the target scattering strengths from the scattering or response matrix that can be applied after the target positions have been estimated in a previous step via, e.g., time-reversal multiple signal classification or another approach. The new formula provides a noniterative analytical alternative to the iterative numerical solution approach for the same problem presented in a recent paper [A. J. Devaney, E. A. Marengo, and F. K. Gruber, "Time-reversal-based imaging and inverse scattering of multiply scattering point targets," J. Acoust. Soc. Am. 118, 3129-3138 (2005)]. The two methods (noniterative versus iterative) are comparatively investigated with two numerical examples.