@article{warrier_rutter_flores_2022, title={Multitask neural networks for predicting bladder pressure with time series data}, volume={72}, ISSN={["1746-8108"]}, DOI={10.1016/j.bspc.2021.103298}, abstractNote={Multitask learning (MTL) can improve accuracy over vanilla neural networks in modeling population level time series data. This can be accomplished by assigning the prediction for each individual in the population as a separate task, thereby leveraging the heterogeneity of population level data. Here, we investigate a novel approach by training recurrent neural networks (RNNs) in a multitask setting. We apply this new methodology to experimental data for predicting bladder pressure, and then bladder contractions, from an external urethral sphincter electromyograph (EUS EMG) signal. We found that the multitask models make more accurate individual level predictions than their single tasking counterparts. We observed that, for bladder pressure prediction, either incorporating multitask learning or RNN structure generalized best to out of sample test data and multitasking RNNs had high out of sample correlation coefficients. These results suggest that MTL models could be used to leverage heterogeneous population time series data for making individualized predictions. From these bladder pressure predictions, we predicted the onset of bladder contractions. Our results indicate that the MTL RNN model was superior in both intra- and inter-individual bladder contraction predictions as measured by sensitivity (85.7%), specificity (98.7%) and precision (73.5%).}, journal={BIOMEDICAL SIGNAL PROCESSING AND CONTROL}, author={Warrier, Sangeeta and Rutter, Erica M. and Flores, Kevin B.}, year={2022}, month={Feb} }
@misc{everett_flores_henscheid_lagergren_larripa_li_nardini_nguyen_pitman_rutter_2020, title={A tutorial review of mathematical techniques for quantifying tumor heterogeneity}, volume={17}, ISSN={["1551-0018"]}, DOI={10.3934/mbe.2020207}, abstractNote={ https://github.com/jtnardin/Tumor-Heterogeneity/ so that this study may serve as a tutorial for the mathematical modelling community. This review article was a product of a Tumor Heterogeneity Working Group as part of the 2018–2019 Program on Statistical, Mathematical, and Computational Methods for Precision Medicine which took place at the Statistical and Applied Mathematical Sciences Institute.]]> }, number={4}, journal={MATHEMATICAL BIOSCIENCES AND ENGINEERING}, author={Everett, Rebecca and Flores, Kevin B. and Henscheid, Nick and Lagergren, John and Larripa, Kamila and Li, Ding and Nardini, John T. and Nguyen, Phuong T. T. and Pitman, E. Bruce and Rutter, Erica M.}, year={2020}, pages={3660–3709} }
@article{lagergren_nardini_michael lavigne_rutter_flores_2020, title={Learning partial differential equations for biological transport models from noisy spatio-temporal data}, volume={476}, ISSN={["1471-2946"]}, DOI={10.1098/rspa.2019.0800}, abstractNote={We investigate methods for learning partial differential equation (PDE) models from spatiotemporal data under biologically realistic levels and forms of noise. Recent progress in learning PDEs from data have used sparse regression to select candidate terms from a denoised set of data, including approximated partial derivatives. We analyze the performance in utilizing previous methods to denoise data for the task of discovering the governing system of partial differential equations (PDEs). We also develop a novel methodology that uses artificial neural networks (ANNs) to denoise data and approximate partial derivatives. We test the methodology on three PDE models for biological transport, i.e., the advection-diffusion, classical Fisher-KPP, and nonlinear Fisher-KPP equations. We show that the ANN methodology outperforms previous denoising methods, including finite differences and polynomial regression splines, in the ability to accurately approximate partial derivatives and learn the correct PDE model.}, number={2234}, journal={PROCEEDINGS OF THE ROYAL SOCIETY A-MATHEMATICAL PHYSICAL AND ENGINEERING SCIENCES}, author={Lagergren, John H. and Nardini, John T. and Michael Lavigne, G. and Rutter, Erica M. and Flores, Kevin B.}, year={2020}, month={Feb} }
@article{rutter_lagergren_flores_2018, title={Automated Object Tracing for Biomedical Image Segmentation Using a Deep Convolutional Neural Network}, volume={11073}, ISBN={["978-3-030-00936-6"]}, ISSN={["1611-3349"]}, DOI={10.1007/978-3-030-00937-3_78}, abstractNote={Convolutional neural networks (CNNs) have been used for fast and accurate segmentation of medical images. In this paper, we present a novel methodology that uses CNNs for segmentation by mimicking the human task of tracing object boundaries. The architecture takes as input a patch of an image with an overlay of previously traced pixels and the output predicts the coordinates of the next m pixels to be traced. We also consider a CNN architecture that leverages the output from another semantic segmentation CNN, e.g., U-net, as an auxiliary image channel. To initialize the trace path in an image, we use either locations identified as object boundaries with high confidence from a semantic segmentation CNN or a short manually traced path. By iterating the CNN output, our method continues the trace until it intersects with the beginning of the path. We show that our network is more accurate than the state-of-the-art semantic segmentation CNN on microscopy images from the ISBI cell tracking challenge. Moreover, our methodology provides a natural platform for performing human-in-the-loop segmentation that is more accurate than CNNs alone and orders of magnitude faster than manual segmentation.}, journal={MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2018, PT IV}, author={Rutter, Erica M. and Lagergren, John H. and Flores, Kevin B.}, year={2018}, pages={686–694} }
@article{rutter_banks_flores_2018, title={Estimating intratumoral heterogeneity from spatiotemporal data}, volume={77}, ISSN={["1432-1416"]}, DOI={10.1007/s00285-018-1238-6}, abstractNote={Glioblastoma multiforme (GBM) is a malignant brain cancer with a tendency to both migrate and proliferate. We propose modeling GBM with heterogeneity in cell phenotypes using a random differential equation version of the reaction-diffusion equation, where the parameters describing diffusion (D) and proliferation ([Formula: see text]) are random variables. We investigate the ability to perform the inverse problem to recover the probability distributions of D and [Formula: see text] using the Prohorov metric, for a variety of probability distribution functions. We test the ability to perform the inverse problem for noisy synthetic data. We then examine the predicted effect of treatment, specifically, chemotherapy, when assuming such a heterogeneous population and compare with predictions from a homogeneous cell population model.}, number={6-7}, journal={JOURNAL OF MATHEMATICAL BIOLOGY}, author={Rutter, E. M. and Banks, H. T. and Flores, K. B.}, year={2018}, month={Dec}, pages={1999–2022} }
@article{pell_phan_rutter_chowell_kuang_2018, title={Simple multi-scale modeling of the transmission dynamics of the 1905 plague epidemic in Bombay}, volume={301}, ISSN={["1879-3134"]}, DOI={10.1016/j.mbs.2018.04.003}, abstractNote={The first few disease generations of an infectious disease outbreak is the most critical phase to implement control interventions. The lack of accurate data and information during the early transmission phase hinders the application of complex compartmental models to make predictions and forecasts about important epidemic quantities. Thus, simpler models are often times better tools to understand the early dynamics of an outbreak particularly in the context of limited data. In this paper we mechanistically derive and fit a family of logistic models to spatial-temporal data of the 1905 plague epidemic in Bombay, India. We systematically compare parameter estimates, reproduction numbers, model fit, and short-term forecasts across models at different spatial resolutions. At the same time, we also assess the presence of sub-exponential growth dynamics at different spatial scales and investigate the role of spatial structure and data resolution (district level data and city level data) using simple structured models. Our results for the 1905 plague epidemic in Bombay indicates that it is possible for the growth of an epidemic in the early phase to be sub-exponential at sub-city level, while maintaining near exponential growth at an aggregated city level. We also show that the rate of movement between districts can have a significant effect on the final epidemic size.}, journal={MATHEMATICAL BIOSCIENCES}, author={Pell, Bruce and Phan, Tin and Rutter, Erica M. and Chowell, Gerardo and Kuang, Yang}, year={2018}, month={Jul}, pages={83–92} }
@article{stepien_rutter_kuang_2018, title={TRAVELING WAVES OF A GO-OR-GROW MODEL OF GLIOMA GROWTH}, volume={78}, ISSN={["1095-712X"]}, DOI={10.1137/17m1146257}, abstractNote={Glioblastoma multiforme is a deadly brain cancer in which tumor cells excessively proliferate and migrate. The first mathematical models of the spread of gliomas featured reaction-diffusion equatio...}, number={3}, journal={SIAM JOURNAL ON APPLIED MATHEMATICS}, author={Stepien, Tracy L. and Rutter, Erica M. and Kuang, Yang}, year={2018}, pages={1778–1801} }
@article{mandi_rutter_payne_2017, title={Effects of non-physiological blood pressure artefacts on cerebral autoregulation}, volume={47}, ISSN={["1873-4030"]}, DOI={10.1016/j.medengphy.2017.06.007}, abstractNote={Cerebral autoregulation refers to the brain's regulation mechanisms that aim to maintain the cerebral blood flow approximately constant. It is often assessed by the autoregulation index (ARI). ARI uses arterial blood pressure and cerebral blood flow velocity time series to produce a ten-scale index of autoregulation performance (0 denoting the absence of and 9 the strongest autoregulation). Unfortunately, data are rarely free from various artefacts. Here, we consider four of the most common non-physiological blood pressure artefacts (saturation, square wave, reduced pulse pressure and impulse) and study their effects on ARI for a range of different artefact sizes. We show that a sufficiently large saturation and square wave always result in ARI reaching the maximum value of 9. The pulse pressure reduction and impulse artefact lead to more diverse behaviour. Finally, we characterized the critical size of artefacts, defined as the minimum artefact size that, on average, leads to a 10% deviation of ARI.}, journal={MEDICAL ENGINEERING & PHYSICS}, author={Mandi, Adam and Rutter, Erica M. and Payne, Stephen J.}, year={2017}, month={Sep}, pages={218–221} }
@article{rutter_stepien_anderies_plasencia_woolf_scheck_turner_liu_frakes_kodibagkar_et al._2017, title={Mathematical Analysis of Glioma Growth in a Murine Model}, volume={7}, ISSN={["2045-2322"]}, DOI={10.1038/s41598-017-02462-0}, abstractNote={Abstract Five immunocompetent C57BL/6-cBrd/cBrd/Cr (albino C57BL/6) mice were injected with GL261-luc2 cells, a cell line sharing characteristics of human glioblastoma multiforme (GBM). The mice were imaged using magnetic resonance (MR) at five separate time points to characterize growth and development of the tumor. After 25 days, the final tumor volumes of the mice varied from 12 mm 3 to 62 mm 3 , even though mice were inoculated from the same tumor cell line under carefully controlled conditions. We generated hypotheses to explore large variances in final tumor size and tested them with our simple reaction-diffusion model in both a 3-dimensional (3D) finite difference method and a 2-dimensional (2D) level set method. The parameters obtained from a best-fit procedure, designed to yield simulated tumors as close as possible to the observed ones, vary by an order of magnitude between the three mice analyzed in detail. These differences may reflect morphological and biological variability in tumor growth, as well as errors in the mathematical model, perhaps from an oversimplification of the tumor dynamics or nonidentifiability of parameters. Our results generate parameters that match other experimental in vitro and in vivo measurements. Additionally, we calculate wave speed, which matches with other rat and human measurements.}, journal={SCIENTIFIC REPORTS}, author={Rutter, Erica M. and Stepien, Tracy L. and Anderies, Barrett J. and Plasencia, Jonathan D. and Woolf, Eric C. and Scheck, Adrienne C. and Turner, Gregory H. and Liu, Qingwei and Frakes, David and Kodibagkar, Vikram and et al.}, year={2017}, month={May} }