The classification of endoscopy images with persistent homology

Dunaeva, O. A.; Edelsbrunner, Herbert; Lukyanov, A. D.; Machin, M.; Malkova, Daria; Kuvaev, Roman; Kashin, Sergey

doi:10.1016/j.patrec.2015.12.012

Cited by 28 publications

(12 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A previous study also applied TDA to detect the patterns of the microsurface structure of the gastrointestinal tract; images were classified according to their patterns into three groups with variable risk for cancer (oval, tubular, and irregular patterns with no, low, and high risk, respectively). Approximately 90% of the classification matches were performed by medical doctors 15 . Moreover, in cardiac image analysis, a study applying TDA to computed tomography images successfully extracted the shape of the trabeculae, the fine muscle columns on the ventricular walls which had been missed by previous methods 45 .…”

Section: Discussionmentioning

confidence: 99%

“…TDA is a collection of methods for identifying topological structures in data 11 and is now considered to be an effective tool to analyze various data in many areas including material science 12 , engineering 13 , and biology 14 . Moreover, TDA has also been applied in medicine to the quantification of tumor shapes [15][16][17] , finding patterns in genetic data of cancer patients 18 , and characterizing brain artery networks 19 . In dermatology, TDA has been applied to segmenting and classifying skin lesions [20][21][22][23] and quantifying the connectivity of epidermal cells 24 .…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Assessment of skin barrier function using skin images with topological data analysis

et al. 2020

View full text Add to dashboard Cite

Recent developments of molecular biology have revealed diverse mechanisms of skin diseases, and precision medicine considering these mechanisms requires the frequent objective evaluation of skin phenotypes. Transepidermal water loss (TEWL) is commonly used for evaluating skin barrier function; however, direct measurement of TEWL is time-consuming and is not convenient for daily clinical practice. Here, we propose a new skin barrier assessment method using skin images with topological data analysis (TDA). TDA enabled efficient identification of structural features from a skin image taken by a microscope. These features reflected the regularity of the skin texture. We found a significant correlation between the topological features and TEWL. Moreover, using the features as input, we trained machine-learning models to predict TEWL and obtained good accuracy (R2 = 0.524). Our results suggest that assessment of skin barrier function by topological image analysis is promising.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Assessment of skin barrier function using skin images with topological data analysis

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Each ROI processed as described above, corresponding to a 2D array of scaled pixel intensities, can be considered as a function f : I → [0, 1] that maps the pixel at the coordinate ( x , y ) to a grayscale value f ( x , y ) in the interval [0, 1] 27,28 . A filtration over the sublevel set L τ = { x : f ( x ) ≤ τ }, was computed using the R package TDA 29 .…”

Section: Methodsmentioning

confidence: 99%

Persistent Homology for the Quantitative Evaluation of Architectural Features in Prostate Cancer Histology

Lawson

Sholl

Brown

et al. 2019

Sci Rep

View full text Add to dashboard Cite

The current system for evaluating prostate cancer architecture is the Gleason grading system which divides the morphology of cancer into five distinct architectural patterns, labeled 1 to 5 in increasing levels of cancer aggressiveness, and generates a score by summing the labels of the two most dominant patterns. The Gleason score is currently the most powerful prognostic predictor of patient outcomes; however, it suffers from problems in reproducibility and consistency due to the high intra-observer and inter-observer variability amongst pathologists. In addition, the Gleason system lacks the granularity to address potentially prognostic architectural features beyond Gleason patterns. We evaluate prostate cancer for architectural subtypes using techniques from topological data analysis applied to prostate cancer glandular architecture. In this work we demonstrate the use of persistent homology to capture architectural features independently of Gleason patterns. Specifically, using persistent homology, we compute topological representations of purely graded prostate cancer histopathology images of Gleason patterns 3,4 and 5, and show that persistent homology is capable of clustering prostate cancer histology into architectural groups through a ranked persistence vector. Our results indicate the ability of persistent homology to cluster prostate cancer histopathology images into unique groups with dominant architectural patterns consistent with the continuum of Gleason patterns. In addition, of particular interest, is the sensitivity of persistent homology to identify specific sub-architectural groups within single Gleason patterns, suggesting that persistent homology could represent a robust quantification method for prostate cancer architecture with higher granularity than the existing semi-quantitative measures. The capability of these topological representations to segregate prostate cancer by architecture makes them an ideal candidate for use as inputs to future machine learning approaches with the intent of augmenting traditional approaches with topological features for improved diagnosis and prognosis.

show abstract

“…Later, it became clear that the simplicial complex language was a natural framework for explicitly representing biological and physical systems. For example, simplicial complexes have been used to represent neural recordings [87,54], classify images [205,56,67], and describe the mesoscale architecture of brain networks [203,204,169,193,161]. Even more recent work has focused on defining generative models to construct simplicial complexes with given topological features [52,51] and investigating dynamics that could take place upon nodes or higher-dimensional simplices [211,132].…”

Section: Simplicial Complexesmentioning

confidence: 99%

The Why, How, and When of Representations for Complex Systems

Torres¹,

Blevins²,

Bassett³

et al. 2021

SIAM Rev.

144

View full text Add to dashboard Cite

Complex systems, composed at the most basic level of units and their interactions, describe phenomena in a wide variety of domains, from neuroscience to computer science and economics. The wide variety of applications has resulted in two key challenges: the generation of many domain-specific strategies for complex systems analyses that are seldom revisited, and the compartmentalization of representation and analysis ideas within a domain due to inconsistency in complex systems language. In this work we propose basic, domain-agnostic language in order to advance toward a more cohesive vocabulary. We use this language to evaluate each step of the complex systems analysis pipeline, beginning with the system under study and data collected, then moving through different mathematical frameworks for encoding the observed data (i.e., graphs, simplicial complexes, and hypergraphs), and relevant computational methods for each framework. At each step we consider different types of dependencies; these are properties of the system that describe how the existence of an interaction among a set of units in a system may affect the possibility of the existence of another relation. We discuss how dependencies may arise and how they may alter the interpretation of results or the entirety of the analysis pipeline. We close with two real-world examples using coauthorship data and email communications data that illustrate how the system under study, the dependencies therein, the research question, and the choice of mathematical representation influence the results. We hope this work can serve as an opportunity for reflection for experienced complex systems scientists, as well as an introductory resource for new researchers.

show abstract

The classification of endoscopy images with persistent homology

Cited by 28 publications

References 6 publications

Assessment of skin barrier function using skin images with topological data analysis

Assessment of skin barrier function using skin images with topological data analysis

Persistent Homology for the Quantitative Evaluation of Architectural Features in Prostate Cancer Histology

The Why, How, and When of Representations for Complex Systems

Contact Info

Product

Resources

About