2022

1. Mixed Effects Machine Learning Models for Colon Cancer Metastasis Prediction using Spatially Localized Immuno-Oncology Markers

   Levy, Joshua J., Bobak, Carly A., Nasir-Moin, Mustafa, Veziroglu, Eren M., Palisoul, Scott M., Barney, Rachael E., Salas, Lucas A., Christensen, Brock C., Tsongalis, Gregory J., and Vaickus, Louis J.

   Pacific Symposium on Biocomputing. Pacific Symposium on Biocomputing 2022

   Abs

   Spatially resolved characterization of the transcriptome and proteome promises to provide further clarity on cancer pathogenesis and etiology, which may inform future clinical practice through classifier development for clinical outcomes. However, batch effects may potentially obscure the ability of machine learning methods to derive complex associations within spatial omics data. Profiling thirty-five stage three colon cancer patients using the GeoMX Digital Spatial Profiler, we found that mixed-effects machine learning (MEML) methods† may provide utility for overcoming significant batch effects to communicate key and complex disease associations from spatial information. These results point to further exploration and application of MEML methods within the spatial omics algorithm development life cycle for clinical deployment.

2. Comparison of machine-learning algorithms for the prediction of current procedural terminology (CPT) codes from pathology reports

   Levy, Joshua, Vattikonda, Nishitha, Haudenschild, Christian, Christensen, Brock, and Vaickus, Louis

   Journal of Pathology Informatics 2022

2021

1. Video-Based Deep Learning to Detect Dyssynergic Defecation with 3D High-Definition Anorectal Manometry

   Levy, Joshua J., Navas, Christopher M., Chandra, Joan A., Christensen, Brock C., Vaickus, Louis J., Curley, Michael, Chey, William D., Baker, Jason R., and Shah, Eric D.

   bioRxiv Dec 2021

   Abs

   BACKGROUND AND AIMS Evaluation for dyssynergia is the most common reason that gastroenterologists refer patients for anorectal manometry, because dyssynergia is amenable to biofeedback by physical therapists. High-definition anorectal manometry (3D-HDAM) is a promising technology to evaluate anorectal physiology, but adoption remains limited by its sheer complexity. We developed a 3D-HDAM deep learning algorithm to evaluate for dyssynergia. METHODS Spatial-temporal data were extracted from consecutive 3D-HDAM studies performed between 2018-2020 at a tertiary institution. The technical procedure and gold standard definition of dyssynergia were based on the London consensus, adapted to the needs of 3D-HDAM technology. Three machine learning models were generated: (1) traditional machine learning informed by conventional anorectal function metrics, (2) deep learning, and (3) a hybrid approach. Diagnostic accuracy was evaluated using bootstrap sampling to calculate area-under-the-curve (AUC). To evaluate overfitting, models were validated by adding 502 simulated defecation maneuvers with diagnostic ambiguity. RESULTS 302 3D-HDAM studies representing 1,208 simulated defecation maneuvers were included (average age 55.2 years; 80.5% women). The deep learning model had comparable diagnostic accuracy (AUC=0.91 [95% confidence interval 0.89-0.93]) to traditional (AUC=0.93[0.92-0.95]) and hybrid (AUC=0.96[0.94-0.97]) predictive models in training cohorts. However, the deep learning model handled ambiguous tests more cautiously than other models; the deep learning model was more likely to designate an ambiguous test as inconclusive (odds ratio=4.21[2.78-6.38]) versus traditional/hybrid approaches. CONCLUSIONS By considering complex spatial-temporal information beyond conventional anorectal function metrics, deep learning on 3D-HDAM technology may enable gastroenterologists to reliably identify and manage dyssynergia in broader practice.

2. DNA 5-hydroxymethylcytosine in pediatric central nervous system tumors may impact tumor classification and is a positive prognostic marker

   Azizgolshani, Nasim, Petersen, Curtis L., Chen, Youdinghuan, Levy, Joshua J., Salas, Lucas A., Perreard, Laurent, Nguyen, Lananh N., and Christensen, Brock C.

   Clinical Epigenetics Sep 2021
3. Journey across epidemiology’s third variables: an anesthesiologist’s guide for successfully navigating confounding, mediation, and effect modification

   Levy, Joshua, Lebeaux, Rebecca, Christensen, Brock, Tosteson, Tor, and Bryan, Yvon

   Regional Anesthesia & Pain Medicine May 2021
4. Using Satellite Images and Deep Learning to Identify Associations Between County-Level Mortality and Residential Neighborhood Features Proximal to Schools: A Cross-Sectional Study

   Levy, Joshua J., Lebeaux, Rebecca M., Hoen, Anne G., Christensen, Brock C., Vaickus, Louis J., and MacKenzie, Todd A.

   Frontiers in Public Health May 2021

   Abs

   What is the relationship between mortality and satellite images as elucidated through the use of Convolutional Neural Networks?Background: Following a century of increase, life expectancy in the United States has stagnated and begun to decline in recent decades. Using satellite images and street view images, prior work has demonstrated associations of the built environment with income, education, access to care, and health factors such as obesity. However, assessment of learned image feature relationships with variation in crude mortality rate across the United States has been lacking.Objective: We sought to investigate if county-level mortality rates in the U.S. could be predicted from satellite images.Methods: Satellite images of neighborhoods surrounding schools were extracted with the Google Static Maps application programming interface for 430 counties representing ~68.9% of the US population. A convolutional neural network was trained using crude mortality rates for each county in 2015 to predict mortality. Learned image features were interpreted using Shapley Additive Feature Explanations, clustered, and compared to mortality and its associated covariate predictors.Results: Predicted mortality from satellite images in a held-out test set of counties was strongly correlated to the true crude mortality rate (Pearson r = 0.72). Direct prediction of mortality using a deep learning model across a cross-section of 430 U.S. counties identified key features in the environment (e.g., sidewalks, driveways, and hiking trails) associated with lower mortality. Learned image features were clustered, and we identified 10 clusters that were associated with education, income, geographical region, race, and age.Conclusions: The application of deep learning techniques to remotely-sensed features of the built environment can serve as a useful predictor of mortality in the United States. Although we identified features that were largely associated with demographic information, future modeling approaches that directly identify image features associated with health-related outcomes have the potential to inform targeted public health interventions.

5. Configuring a federated network of real-world patient health data for multimodal deep learning prediction of health outcomes.

   Haudenschild, Christian, Vaickus, Louis, and Levy, Joshua

   Nov 2021
6. Development of Biologically Interpretable Multimodal Deep Learning Model for Cancer Prognosis Prediction

   Azher, Zarif L, Vaickus, Louis J, Salas, Lucas A, Christensen, Brock, and Levy, Joshua

   Nov 2021
7. Improving the Virtual Trichrome Assessment through Bridge Category Models

   Levy, Joshua, Bobak, Carly A, Azizgolshani, Nasim, Liu, Xiaoying, Ren, Bing, Lisovsky, Mikhail, Suriawinata, Arief A, Christensen, Brock, O’Malley, James, and Vaickus, Louis J

   Nov 2021
8. Estimating the Inter- and Intra-Rater Reliability for NASH Fibrosis Staging in the Presence of Bridge Ordinal Ratings with Hierarchical Bridge Category Models

   Levy, Joshua, Bobak, Carly, Azizgolshani, Nasim, Liu, Xiaoying, Ren, Bing, Lisovsky, Mikhail, Suriawinata, Arief, Christensen, Brock, O’Malley, James, and Vaickus, Louis

   Oct 2021
9. MethylSPWNet and MethylCapsNet: Biologically Motivated Organization of DNAm Neural Networks, Inspired by Capsule Networks

   Levy, Joshua J., Chen, Youdinghuan, Azizgolshani, Nasim, Petersen, Curtis L., Titus, Alexander J., Moen, Erika L., Vaickus, Louis J., Salas, Lucas A., and Christensen, Brock C.

   npj Systems Biology and Applications Aug 2021

   Abs

   DNA methylation (DNAm) alterations have been heavily implicated in carcinogenesis and the pathophysiology of diseases through upstream regulation of gene expression. DNAm deep-learning approaches are able to capture features associated with aging, cell type, and disease progression, but lack incorporation of prior biological knowledge. Here, we present modular, user-friendly deep-learning methodology and software, MethylCapsNet and MethylSPWNet, that group CpGs into biologically relevant capsules—such as gene promoter context, CpG island relationship, or user-defined groupings—and relate them to diagnostic and prognostic outcomes. We demonstrate these models’ utility on 3,897 individuals in the classification of central nervous system (CNS) tumors. MethylCapsNet and MethylSPWNet provide an opportunity to increase DNAm deep-learning analyses’ interpretability by enabling a flexible organization of DNAm data into biologically relevant capsules.

10. Bridge Category Models: Development of Bayesian Modelling Procedures to Account for Bridge Ordinal Ratings for Disease Staging

    Levy, Joshua, Bobak, Carly A, Azizgolshani, Nasim, Andersen, Michael J, Suriawinata, Arief, Liu, Xiaoying, Lisovsky, Mikhail, Ren, Bing, Christensen, Brock, Vaickus, Louis, and O’Malley, James

    Aug 2021
11. Atypia of undetermined significance in thyroid cytology: Nuclear atypia and architectural atypia are associated with different molecular alterations and risks of malignancy

    Glass, Ryan E., Levy, Joshua J., Motanagh, Samaneh A., Vaickus, Louis J., and Liu, Xiaoying

    Cancer Cytopathology Aug 2021
12. MRI Screening after Silicone Implant Breast Surgery: Patient Survey of Adherence to FDA Recommendations

    Copeland-Halperin, Libby R., Wampler, Ariel T., Doughty, Hayden, Shank, Nina, Levy, Joshua J., Rada, Maria L. Reategui Via, and Rosen, Joseph M.

    Plastic and Reconstructive Surgery Aug 2021
13. Evaluating the mediating effects of perceived vulnerability to disease in the relation between disgust and contamination-based OCD

    Brady, Robert E., Badour, Christal L., Arega, Enat A., Levy, Joshua J., and Adams, Thomas G.

    Journal of Anxiety Disorders Apr 2021
14. Topological Feature Extraction and Visualization of Whole Slide Images using Graph Neural Networks.

    Levy, Joshua, Haudenschild, Christian, Barwick, Clark, Christensen, Brock, and Vaickus, Louis

    Pacific Symposium on Biocomputing. Pacific Symposium on Biocomputing Apr 2021
15. Emerging Diagnostic and Investigative Technologies: Validation of Deep Learning Technologies for Dna Methylation and Histopathology

    Levy, Joshua

    Pacific Symposium on Biocomputing. Pacific Symposium on Biocomputing Apr 2021
16. Artificial Intelligence in Anatomic Pathology

    Levy, Joshua J., and Vaickus, Louis J.

    Advances in Molecular Pathology Nov 2021

2020

1. A large-scale internal validation study of unsupervised virtual trichrome staining technologies on nonalcoholic steatohepatitis liver biopsies

   Levy, Joshua J., Azizgolshani, Nasim, Andersen, Michael J., Suriawinata, Arief, Liu, Xiaoying, Lisovsky, Mikhail, Ren, Bing, Bobak, Carly A., Christensen, Brock C., and Vaickus, Louis J.

   Modern Pathology Dec 2020
2. Don’t dismiss logistic regression: the case for sensible extraction of interactions in the era of machine learning

   Levy, Joshua J., and O’Malley, A. James

   BMC Medical Research Methodology Jun 2020
3. Gradual polyploid genome evolution revealed by pan-genomic analysis of Brachypodium hybridum and its diploid progenitors

   Gordon, Sean P., Contreras-Moreira, Bruno, Levy, Joshua J., Djamei, Armin, Czedik-Eysenberg, Angelika, Tartaglio, Virginia S., Session, Adam, Martin, Joel, Cartwright, Amy, Katz, Andrew, Singan, Vasanth R., Goltsman, Eugene, Barry, Kerrie, Dinh-Thi, Vinh Ha, Chalhoub, Boulos, Diaz-Perez, Antonio, Sancho, Ruben, Lusinska, Joanna, Wolny, Elzbieta, Nibau, Candida, Doonan, John H., Mur, Luis A. J., Plott, Chris, Jenkins, Jerry, Hazen, Samuel P., Lee, Scott J., Shu, Shengqiang, Goodstein, David, Rokhsar, Daniel, Schmutz, Jeremy, Hasterok, Robert, Catalan, Pilar, and Vogel, John P.

   Nature Communications Jul 2020
4. MethylNet: an automated and modular deep learning approach for DNA methylation analysis

   Levy, Joshua J., Titus, Alexander J., Petersen, Curtis L., Chen, Youdinghuan, Salas, Lucas A., and Christensen, Brock C.

   BMC Bioinformatics Mar 2020
5. Estimating latent positions in social and biological networks using Graph Neural Networks in R with GCN4R

   Levy, Joshua, Bobak, Carly A, Christensen, Brock, Vaickus, Louis J, and O’Malley, A. James

   Nov 2020
6. Does the timing of postoperative showering impact infection rates? A systematic review and meta-analysis

   Copeland-Halperin, Libby R., Rada, Maria L. Reategui Via, Levy, Joshua, Shank, Nina, Funderburk, Christopher D., and Shin, Joseph H.

   Journal of Plastic, Reconstructive & Aesthetic Surgery Jul 2020
7. Investigating the potential for machine learning prediction of patient outcomes: a retrospective study of hospital acquired pressure injuries

   Levy, Joshua J., Lima, Jorge F., Miller, Megan W., Freed, Gary L., O’Malley, A. James, and Emeny, Rebecca T.

   Apr 2020
8. PathFlow-MixMatch for Whole Slide Image Registration: An Investigation of a Segment-Based Scalable Image Registration Method

   Levy, Joshua J, Jackson, Christopher R, Haudenschild, Christian C, Christensen, Brock C, and Vaickus, Louis J

   Mar 2020
9. PathFlowAI: A High-Throughput Workflow for Preprocessing, Deep Learning and Interpretation in Digital Pathology

   Levy, Joshua, Salas, Lucas A, Christensen, Brock C, Sriharan, Aravindhan, and Vaickus, Louis J

   Pacific Symposium on Biocomputing Mar 2020

   Abs

   The diagnosis of disease often requires analysis of a biopsy. Many diagnoses depend not only on the presence of certain features but on their location within the tissue. Recently, a number of deep learning diagnostic aids have been developed to classify digitized biopsy slides. Clinical workflows often involve processing of more than 500 slides per day. But, clinical use of deep learning diagnostic aids would require a preprocessing workflow that is cost-effective, flexible, scalable, rapid, interpretable, and transparent. Here, we present such a workflow, optimized using Dask and mixed precision training via APEX, capable of handling any patch-level or slide level classification and prediction problem. The workflow uses a flexible and fast preprocessing and deep learning analytics pipeline, incorporates model interpretation and has a highly storageefficient audit trail. We demonstrate the utility of this package on the analysis of a prototypical anatomic pathology specimen, liver biopsies for evaluation of hepatitis from a prospective cohort. The preliminary data indicate that PathFlowAI may become a cost-effective and time-efficient tool for clinical use of Artificial Intelligence (AI) algorithms.

10. Preliminary Evaluation of the Utility of Deep Generative Histopathology Image Translation at a Mid-sized NCI Cancer Center

    Levy, Joshua, Jackson, Christopher, Sriharan, Aravindhan, Christensen, Brock, and Vaickus, Louis

    In Proceedings of the 13th International Joint Conference on Biomedical Engineering Systems and Technologies Mar 2020

2019

1. PyMethylProcess—convenient high-throughput preprocessing workflow for DNA methylation data

   Levy, Joshua J, Titus, Alexander J, Salas, Lucas A, and Christensen, Brock C

   Bioinformatics Aug 2019
2. PolyCRACKER, a robust method for the unsupervised partitioning of polyploid subgenomes by signatures of repetitive DNA evolution.

   Levy, Joshua, Gordon, Sean, and Vogel, John

   BMC genomics Aug 2019

2018