Virtual disease landscape using mechanics-informed machine learning: Application to esophageal disorders

Halder, Sourav; Yamasaki, Jun; Acharya, Shashank; Kou, Wenjun; Elisha, Guy; Carlson, Dustin A.; Kahrilas, Peter J.; Pandolfino, John E.; Patankar, Neelesh A.

doi:10.1016/j.artmed.2022.102435

Cited by 7 publications

(15 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This includes the calculation of the mechanics‐based parameters, prediction by MI‐VAE, and post‐processing. The only manual inputs required are the specification of the time window of sustained volumetric distention at 50 mL and 60 mL and the identification of time instants corresponding to the lower values of pressure in the pressure variations 20 . This allows the VDL to be readily used to plot new patient data and investigate its relative location with respect to nearby points.…”

Section: Discussionmentioning

confidence: 99%

“…The data as described in Table 1 was augmented to increase the dataset and generalizability of predictions. The augmentation details can be found in Halder et al 20 Note that since the MI-VAE is an unsupervised learning approach, the diagnosis labels were not used in the training and the resultant generation of the latent space. The labels were only used as a post-processing step after the VDL was generated through the MI-VAE latent space along with the other mechanics-based scaler parameters.…”

Section: Training Detailsmentioning

confidence: 99%

See 1 more Smart Citation

Enhancing Chicago Classification diagnoses with functional lumen imaging probe—mechanics (FLIP‐MECH)

Halder,

Yamasaki,

Liu

et al. 2024

Neurogastroenterology Motil

Self Cite

View full text Add to dashboard Cite

BackgroundEsophageal motility disorders can be diagnosed by either high‐resolution manometry (HRM) or the functional lumen imaging probe (FLIP) but there is no systematic approach to synergize the measurements of these modalities or to improve the diagnostic metrics that have been developed to analyze them. This work aimed to devise a formal approach to bridge the gap between diagnoses inferred from HRM and FLIP measurements using deep learning and mechanics.MethodsThe “mechanical health” of the esophagus was analyzed in 740 subjects including a spectrum of motility disorder patients and normal subjects. The mechanical health was quantified through a set of parameters including wall stiffness, active relaxation, and contraction pattern. These parameters were used by a variational autoencoder to generate a parameter space called virtual disease landscape (VDL). Finally, probabilities were assigned to each point (subject) on the VDL through linear discriminant analysis (LDA), which in turn was used to compare with FLIP and HRM diagnoses.ResultsSubjects clustered into different regions of the VDL with their location relative to each other (and normal) defined by the type and severity of dysfunction. The two major categories that separated best on the VDL were subjects with normal esophagogastric junction (EGJ) opening and those with EGJ obstruction. Both HRM and FLIP diagnoses correlated well within these two groups.ConclusionMechanics‐based parameters effectively estimated esophageal health using FLIP measurements to position subjects in a 3‐D VDL that segregated subjects in good alignment with motility diagnoses gleaned from HRM and FLIP studies.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Training Detailsmentioning

confidence: 99%

Enhancing Chicago Classification diagnoses with functional lumen imaging probe—mechanics (FLIP‐MECH)

Halder,

Yamasaki,

Liu

et al. 2024

Neurogastroenterology Motil

Self Cite

View full text Add to dashboard Cite

show abstract

“…In our previous studies, we developed a one dimensional (1D) model of a flow inside an elastic tube closed on both ends to imitate a flow inside a FLIP device ( Elisha et al, 2022a ; Elisha et al, 2021 ; Acharya et al, 2021c ; Halder et al, 2022c ). Our goal was to study the relation between tube properties, fluid properties, muscle activation pattern, and their effect on pressure and CSA of the tube.…”

Section: Methodsmentioning

confidence: 99%

“…Thus, understanding esophageal motor function and investigating the contractile behavior of esophageal smooth muscles can help in identifying esophageal disease progression and develop diagnostic tools. Extensive computational studies have been performed in recent years on the mechanics of esophageal transport ( Kou et al, 2017a ; Kou et al, 2017b ; Kou et al, 2018 ; Kou et al, 2017c ; Acharya et al, 2022 ; Halder et al, 2021 ; Elisha et al, 2022a ; Acharya et al, 2021c ; Acharya et al, 2021a ; Elisha et al, 2021 ; Liao et al, 2020 ) and emptying as well as the state of various esophageal diseases ( Acharya et al, 2021b ; Halder et al, 2022a ; Halder et al, 2022c ; Halder et al, 2022b ).…”

Section: Introductionmentioning

confidence: 99%

A mechanics-based perspective on the pressure-cross-sectional area loop within the esophageal body

et al. 2023

Self Cite

View full text Add to dashboard Cite

Introduction: Plotting the pressure-cross-sectional area (P-CSA) hysteresis loops within the esophagus during a contraction cycle can provide mechanistic insights into esophageal motor function. Pressure and cross-sectional area during secondary peristalsis can be obtained from the functional lumen imaging probe (FLIP). The pressure-cross-sectional area plots at a location within the esophageal body (but away from the sphincter) reveal a horizontal loop shape. The horizontal loop shape has phases that appear similar to those in cardiovascular analyses, whichinclude isometric and isotonic contractions followed by isometric and isotonic relaxations. The aim of this study is to explain the various phases of the pressurecross-sectional area hysteresis loops within the esophageal body.Materials and Methods: We simulate flow inside a FLIP device placed inside the esophagus lumen. We focus on three scenarios: long functional lumen imaging probe bag placed insidethe esophagus but not passing through the lower esophageal sphincter, long functional lumen imaging probe bag that crosses the lower esophageal sphincter, and a short functional lumen imaging probe bag placed in the esophagus body that does not pass through the lower esophageal sphincter.Results and Discussion: Horizontal P-CSA area loop pattern is robust and is reproduced in all three cases with only small differences. The results indicate that the horizontal loop pattern is primarily a product of mechanical conditions rather than any inherently different function of the muscle itself. Thus, the distinct phases of the loop can be explained solely based on mechanics.

show abstract

“…A systematic review of the various constitutive models of the esophagus and the other organs of the gastrointestinal tract was conducted by Patel et al (2022 ). In silico mechanics-based analyses ( Acharya et al, 2021b ; Halder et al, 2021 ; Halder et al, 2022b ) have also been performed on data obtained from various diagnostic devices to identify mechanics-based physiomarkers. Acharya et al (2021b ) used a mechanics-based approach to calculate the work carried out by the esophagus in opening the esophagogastric junction (EGJ) and the necessary work required to open the EGJ using data obtained from FLIP.…”

Section: Introductionmentioning

confidence: 99%

MRI-MECH: mechanics-informed MRI to estimate esophageal health

et al. 2023

Self Cite

View full text Add to dashboard Cite

Dynamic magnetic resonance imaging (MRI) is a popular medical imaging technique that generates image sequences of the flow of a contrast material inside tissues and organs. However, its application to imaging bolus movement through the esophagus has only been demonstrated in few feasibility studies and is relatively unexplored. In this work, we present a computational framework called mechanics-informed MRI (MRI-MECH) that enhances that capability, thereby increasing the applicability of dynamic MRI for diagnosing esophageal disorders. Pineapple juice was used as the swallowed contrast material for the dynamic MRI, and the MRI image sequence was used as input to the MRI-MECH. The MRI-MECH modeled the esophagus as a flexible one-dimensional tube, and the elastic tube walls followed a linear tube law. Flow through the esophagus was governed by one-dimensional mass and momentum conservation equations. These equations were solved using a physics-informed neural network. The physics-informed neural network minimized the difference between the measurements from the MRI and model predictions and ensured that the physics of the fluid flow problem was always followed. MRI-MECH calculated the fluid velocity and pressure during esophageal transit and estimated the mechanical health of the esophagus by calculating wall stiffness and active relaxation. Additionally, MRI-MECH predicted missing information about the lower esophageal sphincter during the emptying process, demonstrating its applicability to scenarios with missing data or poor image resolution. In addition to potentially improving clinical decisions based on quantitative estimates of the mechanical health of the esophagus, MRI-MECH can also be adapted for application to other medical imaging modalities to enhance their functionality.

show abstract

Virtual disease landscape using mechanics-informed machine learning: Application to esophageal disorders

Cited by 7 publications

References 58 publications

Enhancing Chicago Classification diagnoses with functional lumen imaging probe—mechanics (FLIP‐MECH)

Enhancing Chicago Classification diagnoses with functional lumen imaging probe—mechanics (FLIP‐MECH)

A mechanics-based perspective on the pressure-cross-sectional area loop within the esophageal body

MRI-MECH: mechanics-informed MRI to estimate esophageal health

Contact Info

Product

Resources

About