Digital twins employ mathematical and computational models to virtually represent a physical object (e.g., planes and human organs), predict the behavior of the object, and enable decision-making to optimize the future behavior of the object. While digital twins have been widely used in engineering for decades, their applications to oncology are only just emerging. Due to advances in experimental techniques quantitatively characterizing cancer, as well as advances in the mathematical and computational sciences, the notion of building and applying digital twins to understand tumor dynamics and personalize the care of cancer patients has been increasingly appreciated. In this review, we present the opportunities and challenges of applying digital twins in clinical oncology, with a particular focus on integrating medical imaging with mechanism-based, tissue-scale mathematical modeling. Specifically, we first introduce the general digital twin framework and then illustrate existing applications of image-guided digital twins in healthcare. Next, we detail both the imaging and modeling techniques that provide practical opportunities to build patient-specific digital twins for oncology. We then describe the current challenges and limitations in developing image-guided, mechanism-based digital twins for oncology along with potential solutions. We conclude by outlining five fundamental questions that can serve as a roadmap when designing and building a practical digital twin for oncology and attempt to provide answers for a specific application to brain cancer. We hope that this contribution provides motivation for the imaging science, oncology, and computational communities to develop practical digital twin technologies to improve the care of patients battling cancer.

Models using clinical and MRI-based radiomic features identified from ductal carcinoma in situ lesions improved prediction of disease upstaging at surgery compared with standard clinical information alone.

This study characterizes the error that results when performing quantitative analysis of abbreviated dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) data of the breast with the Standard Kety–Tofts (SKT) model and its Patlak variant. More specifically, we used simulations and patient data to determine the accuracy with which abbreviated time course data could reproduce the pharmacokinetic parameters, K^trans (volume transfer constant) and v_e (extravascular/extracellular volume fraction), when compared to the full time course data. SKT analysis of simulated abbreviated time courses (ATCs) based on the imaging parameters from two available datasets (collected with a 3T MRI scanner) at a temporal resolution of 15 s (N = 15) and 7.23 s (N = 15) found a concordance correlation coefficient (CCC) greater than 0.80 for ATCs of length 3.0 and 2.5 min, respectively, for the K^trans parameter. Analysis of the experimental data found that at least 90% of patients met this CCC cut-off of 0.80 for the ATCs of the aforementioned lengths. Patlak analysis of experimental data found that 80% of patients from the 15 s resolution dataset and 90% of patients from the 7.27 s resolution dataset met the 0.80 CCC cut-off for ATC lengths of 1.25 and 1.09 min, respectively. This study provides evidence for both the feasibility and potential utility of performing a quantitative analysis of abbreviated breast DCE-MRI in conjunction with acquisition of current standard-of-care high resolution scans without significant loss of information in the community setting.

The purpose of this work is to implement physics-based regularization as a stopping condition in tuning an untrained deep neural network for reconstructing MR images from accelerated data. The ConvDecoder neural network was trained with a physics-based regularization term incorporating the spoiled gradient echo equation that describes variable-flip angle (VFA) data. Fully-sampled VFA k-space data were retrospectively accelerated by factors of R={8,12,18,36} and reconstructed with ConvDecoder (CD), ConvDecoder with the proposed regularization (CD+r), locally low-rank (LR) reconstruction, and compressed sensing with L1-wavelet regularization (L1). Final images from CD+r training were evaluated at the \emph{argmin} of the regularization loss; whereas the CD, LR, and L1 reconstructions were chosen optimally based on ground truth data. The performance measures used were the normalized root-mean square error, the concordance correlation coefficient (CCC), and the structural similarity index (SSIM). The CD+r reconstructions, chosen using the stopping condition, yielded SSIMs that were similar to the CD (p=0.47) and LR SSIMs (p=0.95) across R and that were significantly higher than the L1 SSIMs (p=0.04). The CCC values for the CD+r T1 maps across all R and subjects were greater than those corresponding to the L1 (p=0.15) and LR (p=0.13) T1 maps, respectively. For R > 12 (<4.2 minutes scan time), L1 and LR T1 maps exhibit a loss of spatially refined details compared to CD+r. We conclude that the use of an untrained neural network together with a physics-based regularization loss shows promise as a measure for determining the optimal stopping point in training without relying on fully-sampled ground truth data.