John T. Nardini
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
John T. Nardini has a total of 34 co-authors. Here are the top co-authors by number of shared publications:
- Alexandria Volkening
Department of Mathematics, Purdue University, West Lafayette, IN 47907 USA
3 shared publications - David Matthew Bortz
University of Colorado Boulder
3 shared publications - Kevin B. Flores
Department of Mathematics, North Carolina State University, Raleigh, NC, USA
3 shared publications - Adam A. Malik
Affiliation Mathematical Sciences, Chalmers University of Technology and University of Gothenburg, Gothenburg, Sweden
2 shared publications - Cecilia Krona
Department of Immunology, Genetics and Pathology, Uppsala University, Uppsala, Sweden
2 shared publications Charles W. J. Pugh
Mathematical Institute University of Oxford Oxfordshire UK
2 shared publications
Latest Publications
- 7 August 2026
- Published by Society for Industrial & Applied Mathematics (SIAM) in SIAM Journal on Applied Dynamical Systems
- Vol. 25 (3) , 1981-2021
Quantifying Topological Features and Irregularities in Zebrafish Patterns Using the Sweeping-Plane Filtration
Complex patterns emerge across a wide range of biological systems. While patterns often exhibit remarkable robustness, variation and irregularity exist at multiple scales and can carry important information about the underlying agent interactions driving collective dynamics. Many methods for quantifying patterns focus on large-scale, characteristic features (such as stripe width or spot number), but questions remain on how to characterize messy patterns. In the case of cellular patterns that emerge during development, understanding where patterns are most susceptible to variability may help shed light on cell behavior and the tissue environment. Motivated by these challenges, we introduce methods based on topological data analysis to classify and quantify messy patterns. To compute persistent homology, our methods rely on a sweeping-plane filtration which, in comparison to the Vietoris–Rips filtration, is more rarely applied to self-organization. We demonstrate how results from the sweeping-plane filtration can be interpreted to quantify stripe patterns—with and without interruptions—by analyzing in silico zebrafish skin patterns, and we generate new predictions about which pattern features may be most robust or variable. Our work provides an automated framework for quantifying features and irregularities in complex patterns and highlights how different approaches to persistent homology can provide complementary insight.
- 7 August 2026
- Published by Society for Industrial & Applied Mathematics (SIAM) in SIAM Journal on Applied Dynamical Systems
- Vol. 25 (3) , 1981-2021
Quantifying Topological Features and Irregularities in Zebrafish Patterns Using the Sweeping-Plane Filtration
Complex patterns emerge across a wide range of biological systems. While patterns often exhibit remarkable robustness, variation and irregularity exist at multiple scales and can carry important information about the underlying agent interactions driving collective dynamics. Many methods for quantifying patterns focus on large-scale, characteristic features (such as stripe width or spot number), but questions remain on how to characterize messy patterns. In the case of cellular patterns that emerge during development, understanding where patterns are most susceptible to variability may help shed light on cell behavior and the tissue environment. Motivated by these challenges, we introduce methods based on topological data analysis to classify and quantify messy patterns. To compute persistent homology, our methods rely on a sweeping-plane filtration which, in comparison to the Vietoris–Rips filtration, is more rarely applied to self-organization. We demonstrate how results from the sweeping-plane filtration can be interpreted to quantify stripe patterns—with and without interruptions—by analyzing in silico zebrafish skin patterns, and we generate new predictions about which pattern features may be most robust or variable. Our work provides an automated framework for quantifying features and irregularities in complex patterns and highlights how different approaches to persistent homology can provide complementary insight.
- Open Access
- 22 April 2026
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 22 (4) , e1014161
Enhancing generalizability of model discovery across parameter space with multi-experiment equation learning for biological systems
Agent-based modeling (ABM) is a powerful tool for understanding self-organizing biological systems, but it is computationally intensive and often not analytically tractable. Equation learning (EQL) methods can derive continuum models from ABM data, but they typically require extensive simulations for each parameter set, raising concerns about generalizability. In this work, we extend EQL to Multi-experiment equation learning (ME-EQL) by introducing two methods: (i) one-at-a-time ME-EQL (OAT ME-EQL), which learns individual models for each parameter set and connects them via interpolation, and (ii) embedded structure ME-EQL (ES ME-EQL), which builds a unified model library across parameters. We demonstrate these methods by learning continuum models from a noisy birth–death mean-field model and from an on-lattice agent-based model of birth, death, and migration with spatial structure, often used to investigate cell biology experiments. We show that both methods significantly reduce the relative error in recovering parameters from agent-based simulations, with OAT ME-EQL offering better generalizability across parameter space. Our findings highlight the potential of equation learning from multiple experiments to enhance the generalizability and interpretability of learned models for complex biological systems.
- 22 April 2026
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 22 (4) , e1014161
Open AccessEnhancing generalizability of model discovery across parameter space with multi-experiment equation learning for biological systems
Agent-based modeling (ABM) is a powerful tool for understanding self-organizing biological systems, but it is computationally intensive and often not analytically tractable. Equation learning (EQL) methods can derive continuum models from ABM data, but they typically require extensive simulations for each parameter set, raising concerns about generalizability. In this work, we extend EQL to Multi-experiment equation learning (ME-EQL) by introducing two methods: (i) one-at-a-time ME-EQL (OAT ME-EQL), which learns individual models for each parameter set and connects them via interpolation, and (ii) embedded structure ME-EQL (ES ME-EQL), which builds a unified model library across parameters. We demonstrate these methods by learning continuum models from a noisy birth–death mean-field model and from an on-lattice agent-based model of birth, death, and migration with spatial structure, often used to investigate cell biology experiments. We show that both methods significantly reduce the relative error in recovering parameters from agent-based simulations, with OAT ME-EQL offering better generalizability across parameter space. Our findings highlight the potential of equation learning from multiple experiments to enhance the generalizability and interpretability of learned models for complex biological systems.
- 4 March 2026
- Published by Springer Nature in Bulletin of Mathematical Biology
- Vol. 88 (4) , 48
SSRCA: A Novel Machine Learning Pipeline to Perform Sensitivity Analysis for Agent-Based Models
Agent-based models (ABMs) are widely used in biology to understand how individual actions scale into emergent population behavior. Modelers employ sensitivity analysis (SA) algorithms to quantify input parameters’ impact on model outputs, however, it is hard to perform SA for ABMs due to their computational and complex nature. In this work, we develop the Simulate, Summarize, Reduce, Cluster, and Analyze (SSRCA) methodology, a machine-learning based pipeline designed to facilitate SA for ABMs. In particular, SSRCA can achieve the following tasks for ABMs: 1) identify sensitive model parameters, 2) reveal common output model patterns, and 3) determine which input parameter values generate these patterns. We use an example ABM of tumor spheroid growth to showcase how SSRCA identifies four common patterns from the ABM and the parameter regions that generate these outputs. Additionally, we compare the SA results between SSRCA and the popular Sobol’ Method and find that SSRCA’s identified sensitive parameters are robust to the choice of model descriptors while Sobol’s are not. This analysis could streamline data-driven tasks, such as parameter estimation, for ABMs by reducing parameter space. While we highlight these results with an ABM on tumor spheroid formation, the SSRCA Methodology is broadly applicable to biological ABMs.
- 4 March 2026
- Published by Springer Nature in Bulletin of Mathematical Biology
- Vol. 88 (4) , 48
SSRCA: A Novel Machine Learning Pipeline to Perform Sensitivity Analysis for Agent-Based Models
Agent-based models (ABMs) are widely used in biology to understand how individual actions scale into emergent population behavior. Modelers employ sensitivity analysis (SA) algorithms to quantify input parameters’ impact on model outputs, however, it is hard to perform SA for ABMs due to their computational and complex nature. In this work, we develop the Simulate, Summarize, Reduce, Cluster, and Analyze (SSRCA) methodology, a machine-learning based pipeline designed to facilitate SA for ABMs. In particular, SSRCA can achieve the following tasks for ABMs: 1) identify sensitive model parameters, 2) reveal common output model patterns, and 3) determine which input parameter values generate these patterns. We use an example ABM of tumor spheroid growth to showcase how SSRCA identifies four common patterns from the ABM and the parameter regions that generate these outputs. Additionally, we compare the SA results between SSRCA and the popular Sobol’ Method and find that SSRCA’s identified sensitive parameters are robust to the choice of model descriptors while Sobol’s are not. This analysis could streamline data-driven tasks, such as parameter estimation, for ABMs by reducing parameter space. While we highlight these results with an ABM on tumor spheroid formation, the SSRCA Methodology is broadly applicable to biological ABMs.
- Open Access
Quantifying topological features and irregularities in zebrafish patterns using the sweeping-plane filtration
Complex patterns emerge across a wide range of biological systems. While such patterns often exhibit remarkable robustness, variation and irregularity exist at multiple scales and can carry important information about the underlying agent interactions driving collective dynamics. Many methods for quantifying biological patterns focus on large-scale, characteristic features (such as stripe width or spot number), but questions remain on how to characterize messy patterns. In the case of cellular patterns that emerge during development or regeneration, understanding where patterns are most susceptible to variability may help shed light on cell behavior and the tissue environment. Motivated by these challenges, we introduce methods based on topological data analysis to classify and quantify messy patterns arising from agent-based interactions, by extracting meaningful biological interpretations from persistence barcode summaries. To compute persistent homology, our methods rely on a sweeping-plane filtration which, in comparison to the Vietoris--Rips filtration, is more rarely applied to biological systems. We demonstrate how results from the sweeping-plane filtration can be interpreted to quantify stripe patterns (with and without interruptions) by analyzing in silico zebrafish skin patterns, and we generate new quantitative predictions about which pattern features may be most robust or variable. Our work provides an automated framework for quantifying features and irregularities in spot and stripe patterns and highlights how different approaches to persistent homology can provide complementary insight into biological systems.
Open AccessQuantifying topological features and irregularities in zebrafish patterns using the sweeping-plane filtration
Complex patterns emerge across a wide range of biological systems. While such patterns often exhibit remarkable robustness, variation and irregularity exist at multiple scales and can carry important information about the underlying agent interactions driving collective dynamics. Many methods for quantifying biological patterns focus on large-scale, characteristic features (such as stripe width or spot number), but questions remain on how to characterize messy patterns. In the case of cellular patterns that emerge during development or regeneration, understanding where patterns are most susceptible to variability may help shed light on cell behavior and the tissue environment. Motivated by these challenges, we introduce methods based on topological data analysis to classify and quantify messy patterns arising from agent-based interactions, by extracting meaningful biological interpretations from persistence barcode summaries. To compute persistent homology, our methods rely on a sweeping-plane filtration which, in comparison to the Vietoris--Rips filtration, is more rarely applied to biological systems. We demonstrate how results from the sweeping-plane filtration can be interpreted to quantify stripe patterns (with and without interruptions) by analyzing in silico zebrafish skin patterns, and we generate new quantitative predictions about which pattern features may be most robust or variable. Our work provides an automated framework for quantifying features and irregularities in spot and stripe patterns and highlights how different approaches to persistent homology can provide complementary insight into biological systems.
- Open Access
A novel sensitivity analysis method for agent-based models stratifies in-silico tumor spheroid simulations
Agent-based models (ABMs) are widely used in biology to understand how individual actions scale into emergent population behavior. Modelers employ sensitivity analysis (SA) algorithms to quantify input parameters' impact on model outputs, however, it is hard to perform SA for ABMs due to their computational and complex nature. In this work, we develop the Simulate, Summarize, Reduce, Cluster, and Analyze (SSRCA) methodology, a machine-learning based pipeline designed to facilitate SA for ABMs. In particular, SSRCA can achieve the following tasks for ABMS: 1) identify sensitive model parameters, 2) reveal common output model patterns, and 3) determine which input parameter values generate these patterns. We use an example ABM of tumor spheroid growth to showcase how SSRCA provides similar SA results to the popular Sobol' Method while also identifying four common patterns from the ABM and the parameter regions that generate these outputs. This analysis could streamline data-driven tasks, such as parameter estimation, for ABMs by reducing parameter space. While we highlight these results with an ABM on tumor spheroid formation, the SSRCA methodology is broadly applicable to biological ABMs.
Open AccessA novel sensitivity analysis method for agent-based models stratifies in-silico tumor spheroid simulations
Agent-based models (ABMs) are widely used in biology to understand how individual actions scale into emergent population behavior. Modelers employ sensitivity analysis (SA) algorithms to quantify input parameters' impact on model outputs, however, it is hard to perform SA for ABMs due to their computational and complex nature. In this work, we develop the Simulate, Summarize, Reduce, Cluster, and Analyze (SSRCA) methodology, a machine-learning based pipeline designed to facilitate SA for ABMs. In particular, SSRCA can achieve the following tasks for ABMS: 1) identify sensitive model parameters, 2) reveal common output model patterns, and 3) determine which input parameter values generate these patterns. We use an example ABM of tumor spheroid growth to showcase how SSRCA provides similar SA results to the popular Sobol' Method while also identifying four common patterns from the ABM and the parameter regions that generate these outputs. This analysis could streamline data-driven tasks, such as parameter estimation, for ABMs by reducing parameter space. While we highlight these results with an ABM on tumor spheroid formation, the SSRCA methodology is broadly applicable to biological ABMs.
- Open Access
- 15 February 2025
- Published by Springer Nature in npj Systems Biology and Applications
- Vol. 11 (1) , 1-14
Mathematical modeling of multicellular tumor spheroids quantifies inter-patient and intra-tumor heterogeneity
In the study of brain tumors, patient-derived three-dimensional sphere cultures provide an important tool for studying emerging treatments. The growth of such spheroids depends on the combined effects of proliferation and migration of cells, but it is challenging to make accurate distinctions between increase in cell number versus the radial movement of cells. To address this, we formulate a novel model in the form of a system of two partial differential equations (PDEs) incorporating both migration and growth terms, and show that it more accurately fits our data compared to simpler PDE models. We show that traveling-wave speeds are strongly associated with population heterogeneity. Having fitted the model to our dataset we show that a subset of the cell lines are best described by a “Go-or-Grow”-type model, which constitutes a special case of our model. Finally, we investigate whether our fitted model parameters are correlated with patient age and survival.
- 15 February 2025
- Published by Springer Nature in npj Systems Biology and Applications
- Vol. 11 (1) , 1-14
Open AccessMathematical modeling of multicellular tumor spheroids quantifies inter-patient and intra-tumor heterogeneity
In the study of brain tumors, patient-derived three-dimensional sphere cultures provide an important tool for studying emerging treatments. The growth of such spheroids depends on the combined effects of proliferation and migration of cells, but it is challenging to make accurate distinctions between increase in cell number versus the radial movement of cells. To address this, we formulate a novel model in the form of a system of two partial differential equations (PDEs) incorporating both migration and growth terms, and show that it more accurately fits our data compared to simpler PDE models. We show that traveling-wave speeds are strongly associated with population heterogeneity. Having fitted the model to our dataset we show that a subset of the cell lines are best described by a “Go-or-Grow”-type model, which constitutes a special case of our model. Finally, we investigate whether our fitted model parameters are correlated with patient age and survival.
- 23 September 2024
- Published by Springer Nature in Bulletin of Mathematical Biology
- Vol. 86 (11) , 1-28
Forecasting and Predicting Stochastic Agent-Based Model Data with Biologically-Informed Neural Networks
Collective migration is an important component of many biological processes, including wound healing, tumorigenesis, and embryo development. Spatial agent-based models (ABMs) are often used to model collective migration, but it is challenging to thoroughly predict these models’ behavior throughout parameter space due to their random and computationally intensive nature. Modelers often coarse-grain ABM rules into mean-field differential equation (DE) models. While these DE models are fast to simulate, they suffer from poor (or even ill-posed) ABM predictions in some regions of parameter space. In this work, we describe how biologically-informed neural networks (BINNs) can be trained to learn interpretable BINN-guided DE models capable of accurately predicting ABM behavior. In particular, we show that BINN-guided partial DE (PDE) simulations can (1) forecast future spatial ABM data not seen during model training, and (2) predict ABM data at previously-unexplored parameter values. This latter task is achieved by combining BINN-guided PDE simulations with multivariate interpolation. We demonstrate our approach using three case study ABMs of collective migration that imitate cell biology experiments and find that BINN-guided PDEs accurately forecast and predict ABM data with a one-compartment PDE when the mean-field PDE is ill-posed or requires two compartments. This work suggests that BINN-guided PDEs allow modelers to efficiently explore parameter space, which may enable data-driven tasks for ABMs, such as estimating parameters from experimental data. All code and data from our study is available at https://github.com/johnnardini/Forecasting_predicting_ABMs.
- 23 September 2024
- Published by Springer Nature in Bulletin of Mathematical Biology
- Vol. 86 (11) , 1-28
Forecasting and Predicting Stochastic Agent-Based Model Data with Biologically-Informed Neural Networks
Collective migration is an important component of many biological processes, including wound healing, tumorigenesis, and embryo development. Spatial agent-based models (ABMs) are often used to model collective migration, but it is challenging to thoroughly predict these models’ behavior throughout parameter space due to their random and computationally intensive nature. Modelers often coarse-grain ABM rules into mean-field differential equation (DE) models. While these DE models are fast to simulate, they suffer from poor (or even ill-posed) ABM predictions in some regions of parameter space. In this work, we describe how biologically-informed neural networks (BINNs) can be trained to learn interpretable BINN-guided DE models capable of accurately predicting ABM behavior. In particular, we show that BINN-guided partial DE (PDE) simulations can (1) forecast future spatial ABM data not seen during model training, and (2) predict ABM data at previously-unexplored parameter values. This latter task is achieved by combining BINN-guided PDE simulations with multivariate interpolation. We demonstrate our approach using three case study ABMs of collective migration that imitate cell biology experiments and find that BINN-guided PDEs accurately forecast and predict ABM data with a one-compartment PDE when the mean-field PDE is ill-posed or requires two compartments. This work suggests that BINN-guided PDEs allow modelers to efficiently explore parameter space, which may enable data-driven tasks for ABMs, such as estimating parameters from experimental data. All code and data from our study is available at https://github.com/johnnardini/Forecasting_predicting_ABMs.
- Open Access
- 13 February 2024
- Published by Springer Nature in Research Square
Mathematical Modeling of Multicellular Tumor Spheroids Quantifies Inter-Patient and Inter-Tumor Heterogeneity
In the study of brain tumors, patient-derived three-dimensional sphere cultures provide an important tool for studying emerging treatments. The growth of such spheroids depends on the combined effects of proliferation and migration of cells, but it is often challenging to make accurate distinctions between increase in cell number versus the radial movement of cells. To address this, we formulate a novel model (which we refer to as the reaction-diffusion, advection-reaction-diffusion model) in the form of a system of two partial differential equations (PDEs) and show that it more accurately fits our data as compared to simpler PDE models, including the reaction-diffusion model that has been found to be clinically relevant. We propose a numerical approach for calculating travelling-wave speeds, which are shown to be strongly associated with population heterogeneity. Having fitted the model to our dataset we show that a subset of the cell lines are best described by a “Go-or-Grow”-type model, which constitutes a special case of our model. Finally, we investigate whether our fitted model parameters are correlated with patient age at diagnosis and survival. Multiple model parameters are correlated with patient age, and a single model parameter with patient survival. Our work improves on current modelling approaches, and provides a novel step towards understanding associations between mathematical model parameters and clinical data.
- 13 February 2024
- Published by Springer Nature in Research Square
Open AccessMathematical Modeling of Multicellular Tumor Spheroids Quantifies Inter-Patient and Inter-Tumor Heterogeneity
In the study of brain tumors, patient-derived three-dimensional sphere cultures provide an important tool for studying emerging treatments. The growth of such spheroids depends on the combined effects of proliferation and migration of cells, but it is often challenging to make accurate distinctions between increase in cell number versus the radial movement of cells. To address this, we formulate a novel model (which we refer to as the reaction-diffusion, advection-reaction-diffusion model) in the form of a system of two partial differential equations (PDEs) and show that it more accurately fits our data as compared to simpler PDE models, including the reaction-diffusion model that has been found to be clinically relevant. We propose a numerical approach for calculating travelling-wave speeds, which are shown to be strongly associated with population heterogeneity. Having fitted the model to our dataset we show that a subset of the cell lines are best described by a “Go-or-Grow”-type model, which constitutes a special case of our model. Finally, we investigate whether our fitted model parameters are correlated with patient age at diagnosis and survival. Multiple model parameters are correlated with patient age, and a single model parameter with patient survival. Our work improves on current modelling approaches, and provides a novel step towards understanding associations between mathematical model parameters and clinical data.
- 2 February 2023
- Published by Wiley in Microcirculation
- Vol. 30 (4) , e12799
Statistical and topological summaries aid disease detection for segmented retinal vascular images
Objective: Disease complications can alter vascular network morphology and disrupt tissue functioning. Microvascular diseases of the retina are assessed by visual inspection of retinal images, but this can be challenging when diseases exhibit silent symptoms or patients cannot attend in‐person meetings. We examine the performance of machine learning algorithms in detecting microvascular disease when trained on statistical and topological summaries of segmented retinal vascular images.Methods: We compute 13 separate descriptor vectors (5 statistical, 8 topological) to summarize the morphology of retinal vessel segmentation images and train support vector machines to predict each image's disease classification from the summary vectors. We assess the performance of each descriptor vector, using five‐fold cross validation to estimate their accuracy. We apply these methods to four datasets that were assembled from four existing data repositories; three datasets contain segmented retinal vascular images from one of the repositories, whereas the fourth “All” dataset combines images from four repositories.Results: Among the 13 total descriptor vectors considered, either a statistical Box‐counting descriptor vector or a topological Flooding descriptor vector achieves the highest accuracy levels. On the combined “All” dataset, the Box‐counting vector outperforms all other descriptors, including the topological Flooding vector which is sensitive to differences in the annotation styles between the different datasets.Conclusion: Our work represents a first step to establishing which computational methods are most suitable for identifying microvascular disease and assessing their current limitations. These methods could be incorporated into automated disease assessment tools.
- 2 February 2023
- Published by Wiley in Microcirculation
- Vol. 30 (4) , e12799
Statistical and topological summaries aid disease detection for segmented retinal vascular images
Objective: Disease complications can alter vascular network morphology and disrupt tissue functioning. Microvascular diseases of the retina are assessed by visual inspection of retinal images, but this can be challenging when diseases exhibit silent symptoms or patients cannot attend in‐person meetings. We examine the performance of machine learning algorithms in detecting microvascular disease when trained on statistical and topological summaries of segmented retinal vascular images.Methods: We compute 13 separate descriptor vectors (5 statistical, 8 topological) to summarize the morphology of retinal vessel segmentation images and train support vector machines to predict each image's disease classification from the summary vectors. We assess the performance of each descriptor vector, using five‐fold cross validation to estimate their accuracy. We apply these methods to four datasets that were assembled from four existing data repositories; three datasets contain segmented retinal vascular images from one of the repositories, whereas the fourth “All” dataset combines images from four repositories.Results: Among the 13 total descriptor vectors considered, either a statistical Box‐counting descriptor vector or a topological Flooding descriptor vector achieves the highest accuracy levels. On the combined “All” dataset, the Box‐counting vector outperforms all other descriptors, including the topological Flooding vector which is sensitive to differences in the annotation styles between the different datasets.Conclusion: Our work represents a first step to establishing which computational methods are most suitable for identifying microvascular disease and assessing their current limitations. These methods could be incorporated into automated disease assessment tools.
- Open Access
Statistical and Topological Summaries Aid Disease Detection for Segmented Retinal Vascular Images
Disease complications can alter vascular network morphology and disrupt tissue functioning. Diabetic retinopathy, for example, is a complication of types 1 and 2 diabetes mellitus that can cause blindness. Microvascular diseases are assessed by visual inspection of retinal images, but this can be challenging when diseases exhibit silent symptoms or patients cannot attend in-person meetings. We examine the performance of machine learning algorithms in detecting microvascular disease when trained on statistical and topological summaries of segmented retinal vascular images. We apply our methods to three publicly-available datasets and find that, among the 13 total descriptor vectors we consider, either a statistical Box-counting descriptor vector or a topological Flooding descriptor vector achieves the highest accuracy levels on these datasets. We then created a fourth dataset by merging several datasets: the Box-counting vector outperforms all descriptors on this dataset, including the topological Flooding vector which is sensitive to differences in the annotation styles within the combined dataset. Our work represents a first step to establishing which computational methods are most suitable for identifying microvascular disease as well as some of their current limitations. In the longer term, these methods could be incorporated into automated disease assessment tools.
Open AccessStatistical and Topological Summaries Aid Disease Detection for Segmented Retinal Vascular Images
Disease complications can alter vascular network morphology and disrupt tissue functioning. Diabetic retinopathy, for example, is a complication of types 1 and 2 diabetes mellitus that can cause blindness. Microvascular diseases are assessed by visual inspection of retinal images, but this can be challenging when diseases exhibit silent symptoms or patients cannot attend in-person meetings. We examine the performance of machine learning algorithms in detecting microvascular disease when trained on statistical and topological summaries of segmented retinal vascular images. We apply our methods to three publicly-available datasets and find that, among the 13 total descriptor vectors we consider, either a statistical Box-counting descriptor vector or a topological Flooding descriptor vector achieves the highest accuracy levels on these datasets. We then created a fourth dataset by merging several datasets: the Box-counting vector outperforms all descriptors on this dataset, including the topological Flooding vector which is sensitive to differences in the annotation styles within the combined dataset. Our work represents a first step to establishing which computational methods are most suitable for identifying microvascular disease as well as some of their current limitations. In the longer term, these methods could be incorporated into automated disease assessment tools.