A deep-learning-based unsupervised model on esophageal manometry using variational autoencoder

Artificial Intelligence Tools for Improving Manometric Diagnosis of Esophageal Dysmotility

PURPOSE OF REVIEW

Artificial intelligence (AI) is a broad term that pertains to a computer's ability to mimic and sometimes surpass human intelligence in interpretation of large datasets. The adoption of AI in gastrointestinal motility has been slower compared to other areas such as polyp detection and interpretation of histopathology.

RECENT FINDINGS

Within esophageal physiologic testing, AI can automate interpretation of image-based tests, especially high resolution manometry (HRM) and functional luminal imaging probe (FLIP) studies. Basic tasks such as identification of landmarks, determining adequacy of the HRM study and identification from achalasia from non-achalasia patterns are achieved with good accuracy. However, existing AI systems compare AI interpretation to expert analysis rather than to clinical outcome from management based on AI diagnosis. The use of AI methods is much less advanced within the field of ambulatory reflux monitoring, where challenges exist in assimilation of data from multiple impedance and pH channels. There remains potential for replication of the AI successes within esophageal physiologic testing to HRM of the anorectum, and to innovative and novel methods of evaluating gastric electrical activity and motor function. The use of AI has tremendous potential to improve detection of dysmotility within the esophagus using esophageal physiologic testing, as well as in other regions of the gastrointestinal tract. Eventually, integration of patient presentation, demographics and alternate test results to individual motility test interpretation will improve diagnostic precision and prognostication using AI tools.

Artificial Intelligence and Anorectal Manometry: Automatic Detection and Differentiation of Anorectal Motility Patterns-A Proof-of-Concept Study

INTRODUCTION

Anorectal manometry (ARM) is the gold standard for the evaluation of anorectal functional disorders, prevalent in the population. Nevertheless, the accessibility to this examination is limited, and the complexity of data analysis and report is a significant drawback. This pilot study aimed to develop and validate an artificial intelligence model to automatically differentiate motility patterns of fecal incontinence (FI) from obstructed defecation (OD) using ARM data.

METHODS

We developed and tested multiple machine learning algorithms for the automatic interpretation of ARM data. Four models were tested: k-nearest neighbors, support vector machines, random forests, and gradient boosting (xGB). These models were trained using a stratified 5-fold strategy. Their performance was assessed after fine-tuning of each model's hyperparameters, using 90% of data for training and 10% of data for testing.

RESULTS

A total of 827 ARM examinations were used in this study. After fine-tuning, the xGB model presented an overall accuracy (84.6% ± 2.9%), similar to that of random forests (82.7% ± 4.8%) and support vector machines (81.0% ± 8.0%) and higher that of k-nearest neighbors (74.4% ± 3.8%). The xGB models showed the highest discriminating performance between OD and FI, with an area under the curve of 0.939.

DISCUSSION

The tested machine learning algorithms, particularly the xGB model, accurately differentiated between FI and OD manometric patterns. Subsequent development of these tools may optimize the access to ARM studies, which may have a significant impact on the management of patients with anorectal functional diseases.

Development of a Deep Learning-Based Epiglottis Obstruction Ratio Calculation System

Surgeons determine the treatment method for patients with epiglottis obstruction based on its severity, often by estimating the obstruction severity (using three obstruction degrees) from the examination of drug-induced sleep endoscopy images. However, the use of obstruction degrees is inadequate and fails to correspond to changes in respiratory airflow. Current artificial intelligence image technologies can effectively address this issue. To enhance the accuracy of epiglottis obstruction assessment and replace obstruction degrees with obstruction ratios, this study developed a computer vision system with a deep learning-based method for calculating epiglottis obstruction ratios. The system employs a convolutional neural network, the YOLOv4 model, for epiglottis cartilage localization, a color quantization method to transform pixels into regions, and a region puzzle algorithm to calculate the range of a patient's epiglottis airway. This information is then utilized to compute the obstruction ratio of the patient's epiglottis site. Additionally, this system integrates web-based and PC-based programming technologies to realize its functionalities. Through experimental validation, this system was found to autonomously calculate obstruction ratios with a precision of 0.1% (ranging from 0% to 100%). It presents epiglottis obstruction levels as continuous data, providing crucial diagnostic insight for surgeons to assess the severity of epiglottis obstruction in patients.

A fuzzy-based framework for diagnosing esophageal motility disorder using high-resolution manometry

In recent years, the high-resolution manometry (HRM) technique has been increasingly used to study esophageal and colonic pressurization and has become a standard routine for discovering mobility disorders. In addition to evolving guidelines for the interpretation of HRM like Chicago standard, some complexities, such as the dependency of normative reference values on the recording device and other external variables, still remain for medical professions. In this study, a decision support framework is developed to aid the diagnosis of esophageal mobility disorders based on HRM data. To abstract HRM data, Spearman correlation is employed to model the spatio-temporal dependencies of pressure values of HRM components and convolutional graph neural networks are then utilized to embed relation graphs to the features vector. In the decision-making stage, a novel Expert per Class Fuzzy Classifier (EPC-FC) is presented that employs the ensemble structure and contains expertized sub-classifiers for recognizing a specific disorder. Training sub-classifiers using the negative correlation learning method makes the EPC-FC highly generalizable. Meanwhile, separating the sub-classifiers of each class gives flexibility and interpretability to the structure. The suggested framework is evaluated on a dataset of 67 patients in 5 different classes recorded in Shariati Hospital. The average accuracy of 78.03% for a single swallow and 92.54% for subject-level is achieved for distinguishing mobility disorders. Moreover, compared with the other studies, the presented framework has an outstanding performance considering that it imposes no limits on the type of classes or HRM data. On the other hand, the EPC-FC outperforms other comparative classifiers such as SVM and AdaBoost not only in HRM diagnosis but also on other benchmark classification problems.

An artificial intelligence platform provides an accurate interpretation of esophageal motility from Functional Lumen Imaging Probe Panometry studies

BACKGROUND

Functional lumen imaging probe (FLIP) Panometry is performed at the time of sedated endoscopy and evaluates esophageal motility in response to distension. This study aimed to develop and test an automated artificial intelligence (AI) platform that could interpret FLIP Panometry studies.

METHODS

The study cohort included 678 consecutive patients and 35 asymptomatic controls that completed FLIP Panometry during endoscopy and high-resolution manometry (HRM). "True" study labels for model training and testing were assigned by experienced esophagologists per a hierarchical classification scheme. The supervised, deep learning, AI model generated FLIP Panometry heatmaps from raw FLIP data and based on convolutional neural networks assigned esophageal motility labels using a two-stage prediction model. Model performance was tested on a 15% held-out test set (n = 103); the remainder of the studies were utilized for model training (n = 610).

KEY RESULTS

"True" FLIP labels across the entire cohort included 190 (27%) "normal," 265 (37%) "not normal/not achalasia," and 258 (36%) "achalasia." On the test set, both the Normal/Not normal and the achalasia/not achalasia models achieved an accuracy of 89% (with 89%/88% recall, 90%/89% precision, respectively). Of 28 patients with achalasia (per HRM) in the test set, 0 were predicted as "normal" and 93% as "achalasia" by the AI model.

CONCLUSIONS

An AI platform provided accurate interpretation of FLIP Panometry esophageal motility studies from a single center compared with the impression of experienced FLIP Panometry interpreters. This platform may provide useful clinical decision support for esophageal motility diagnosis from FLIP Panometry studies performed at the time of endoscopy.

A survey on deep learning applied to medical images: from simple artificial neural networks to generative models

Deep learning techniques, in particular generative models, have taken on great importance in medical image analysis. This paper surveys fundamental deep learning concepts related to medical image generation. It provides concise overviews of studies which use some of the latest state-of-the-art models from last years applied to medical images of different injured body areas or organs that have a disease associated with (e.g., brain tumor and COVID-19 lungs pneumonia). The motivation for this study is to offer a comprehensive overview of artificial neural networks (NNs) and deep generative models in medical imaging, so more groups and authors that are not familiar with deep learning take into consideration its use in medicine works. We review the use of generative models, such as generative adversarial networks and variational autoencoders, as techniques to achieve semantic segmentation, data augmentation, and better classification algorithms, among other purposes. In addition, a collection of widely used public medical datasets containing magnetic resonance (MR) images, computed tomography (CT) scans, and common pictures is presented. Finally, we feature a summary of the current state of generative models in medical image including key features, current challenges, and future research paths.

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

Esophageal disorders are related to the mechanical properties and function of the esophageal wall. Therefore, to understand the underlying fundamental mechanisms behind various esophageal disorders, it is crucial to map mechanical behavior of the esophageal wall in terms of mechanics-based parameters corresponding to altered bolus transit and increased intrabolus pressure. We present a hybrid framework that combines fluid mechanics and machine learning to identify the underlying physics of various esophageal disorders (motility disorders, eosinophilic esophagitis, reflux disease, scleroderma esophagus) and maps them onto a parameter space which we call the virtual disease landscape (VDL). A one-dimensional inverse model processes the output from an esophageal diagnostic device called the functional lumen imaging probe (FLIP) to estimate the mechanical "health" of the esophagus by predicting a set of mechanics-based parameters such as esophageal wall stiffness, muscle contraction pattern and active relaxation of esophageal wall. The mechanics-based parameters were then used to train a neural network that consists of a variational autoencoder that generated a latent space and a side network that predicted mechanical work metrics for estimating esophagogastric junction motility. The latent vectors along with a set of discrete mechanics-based parameters define the VDL and formed clusters corresponding to specific esophageal disorders. The VDL not only distinguishes among disorders but also displayed disease progression over time. Finally, we demonstrated the clinical applicability of this framework for estimating the effectiveness of a treatment and tracking patients' condition after a treatment.

Automated Chicago Classification for Esophageal Motility Disorder Diagnosis Using Machine Learning

The goal of this paper is to provide a Machine Learning-based solution that can be utilized to automate the Chicago Classification algorithm, the state-of-the-art scheme for esophageal motility disease identification. First, the photos were preprocessed by locating the area of interest—the precise instant of swallowing. After resizing and rescaling the photos, they were utilized as input for the Deep Learning models. The InceptionV3 Deep Learning model was used to identify the precise class of the IRP. We used the DenseNet201 CNN architecture to classify the images into 5 different classes of swallowing disorders. Finally, we combined the results of the two trained ML models to automate the Chicago Classification algorithm. With this solution we obtained a top-1 accuracy and f1-score of 86% with no human intervention, automating the whole flow, from image preprocessing until Chicago classification and diagnosis.

Deep learning-based artificial intelligence model for identifying swallow types in esophageal high-resolution manometry

BACKGROUND

This study aimed to build and evaluate a deep learning, artificial intelligence (AI) model to automatically classify swallow types based on raw data from esophageal high-resolution manometry (HRM).

METHODS

HRM studies on patients with no history of esophageal surgery were collected including 1,741 studies with 26,115 swallows labeled by swallow type (normal, hypercontractile, weak-fragmented, failed, and premature) by an expert interpreter per the Chicago Classification. The dataset was stratified and split into train/validation/test datasets for model development. Long short-term memory (LSTM), a type of deep-learning AI model, was trained and evaluated. The overall performance and detailed per-swallow type performance were analyzed. The interpretations of the supine swallows in a single study were further used to generate an overall classification of peristalsis.

KEY RESULTS

The LSTM model for swallow type yielded accuracies from the train/validation/test datasets of 0.86/0.81/0.83. The model's interpretation for study-level classification of peristalsis yielded accuracy of 0.88 in the test dataset. Among model misclassification, 535/698 (77%) swallows and 25/35 (71%) studies were to adjacent categories, for example, normal to weak or normal to ineffective, respectively.

CONCLUSIONS AND INFERENCES

A deep-learning AI model can automatically and accurately identify the Chicago Classification swallow types and peristalsis classification from raw HRM data. While future work to refine this model and incorporate overall manometric diagnoses are needed, this study demonstrates the role that AI will serve in the interpretation and classification of esophageal HRM studies.

The present and future of gastroenterology and hepatology: an international SWOT analysis (the GASTROSWOT project)

GASTROSWOT is a strategic analysis of the current and projected states of the different subspecialties in gastroenterology that aims to provide guidance for research, clinical, and financial planning in gastroenterology. We executed a consensus-based international strengths, weaknesses, opportunities, and threats (SWOT) analysis. Four general coordinators, six field coordinators, and 12 experts participated in the study. SWOTs were provided for the following fields: neurogastroenterology, functional gastrointestinal disorders, and upper gastrointestinal diseases; inflammatory bowel disease; pancreatology and biliary diseases; endoscopy; gastrointestinal oncology; and hepatology. The GASTROSWOT analysis highlights the following in the current state of the field of gastroenterology: the incidence and complexity of several gastrointestinal diseases, including malignancies, are increasing; the COVID-19 pandemic has affected patient care on several levels; and with the advent of technical innovations in gastroenterology, a well trained workforce and strategic planning are required to optimise health-care utilisation. The analysis calls attention to the following in the future of gastroenterology: artificial intelligence and the use of big data will speed up discovery and smarter health-care provision in the field; the growth and diversification of gastroenterological specialties will improve specialised care for patients, but could promote fragmentation of care and health system inefficiencies; and furthermore, thoughtful planning is needed to reach an effective balance between the need for subspecialists and the value of general gastroenterology services.

A multi-stage machine learning model for diagnosis of esophageal manometry

High-resolution manometry (HRM) is the primary procedure used to diagnose esophageal motility disorders. Its manual interpretation and classification, including evaluation of swallow-level outcomes and then derivation of a study-level diagnosis based on Chicago Classification (CC), may be limited by inter-rater variability and inaccuracy of an individual interpreter. We hypothesized that an automatic diagnosis platform using machine learning and artificial intelligence approaches could be developed to accurately identify esophageal motility diagnoses. Further, a multi-stage modeling framework, akin to the step-wise approach of the CC, was utilized to leverage advantages of a combination of machine learning approaches including deep-learning models and feature-based models. Models were trained and tested using a dataset comprised of 1741 patients' HRM studies with CC diagnoses assigned by expert physician raters. In the swallow-level stage, three models based on convolutional neural networks (CNNs) were developed to predict swallow type and swallow pressurization (test accuracies of 0.88 and 0.93, respectively), and integrated relaxation pressure (IRP)(regression model with test error of 4.49 mmHg). At the study-level stage, model selection from families of the expert-knowledge-based rule models, xgboost models and artificial neural network(ANN) models were conducted. A simple model-agnostic strategy of model balancing motivated by Bayesian principles was utilized, which gave rise to model averaging weighted by precision scores. The averaged (blended) models and individual models were compared and evaluated, of which the best performance on test dataset is 0.81 in top-1 prediction, 0.92 in top-2 predictions. This is the first artificial-intelligence style model to automatically predict esophageal motility (CC) diagnoses from HRM studies using raw multi-swallow data and it achieved high accuracy. Thus, this proposed modeling framework could be broadly applied to assist with HRM interpretation in a clinical setting.

Integrated Relaxation Pressure Classification and Probe Positioning Failure Detection in High-Resolution Esophageal Manometry Using Machine Learning

High-resolution esophageal manometry is used for the study of esophageal motility disorders, with the help of catheters with up to 36 sensors. Color pressure topography plots are generated and analyzed and using the Chicago algorithm a final diagnosis is established. One of the main parameters in this algorithm is integrated relaxation pressure (IRP). The procedure is time consuming. Our aim was to firstly develop a machine learning based solution to detect probe positioning failure and to create a classifier to automatically determine whether the IRP is in the normal range or higher than the cut-off, based solely on the raw images. The first step was the preprocessing of the images, by finding the region of interest—the exact moment of swallowing. Afterwards, the images were resized and rescaled, so they could be used as input for deep learning models. We used the InceptionV3 deep learning model to classify the images as correct or failure in catheter positioning and to determine the exact class of the IRP. The accuracy of the trained convolutional neural networks was above 90% for both problems. This work is just the first step in fully automating the Chicago Classification, reducing human intervention.

Applications of Artificial Intelligence for the Diagnosis of Gastrointestinal Diseases

The development of convolutional neural networks has achieved impressive advances of machine learning in recent years, leading to an increasing use of artificial intelligence (AI) in the field of gastrointestinal (GI) diseases. AI networks have been trained to differentiate benign from malignant lesions, analyze endoscopic and radiological GI images, and assess histological diagnoses, obtaining excellent results and high overall diagnostic accuracy. Nevertheless, there data are lacking on side effects of AI in the gastroenterology field, and high-quality studies comparing the performance of AI networks to health care professionals are still limited. Thus, large, controlled trials in real-time clinical settings are warranted to assess the role of AI in daily clinical practice. This narrative review gives an overview of some of the most relevant potential applications of AI for gastrointestinal diseases, highlighting advantages and main limitations and providing considerations for future development.

Deep learning for tracing esophageal motility function over time

BACKGROUND AND OBJECTIVE

Esophageal high-resolution manometry (HRM) is widely performed to evaluate the representation of manometric features in patients for diagnosing normal esophageal motility and motility disorders. Clinicians commonly assess esophageal motility function using a scheme termed the Chicago classification, which is difficult, time-consuming and inefficient with large amounts of data.

METHODS

Deep learning is a promising approach for diagnosing disorders and has various attractive advantages. In this study, we effectively trace esophageal motility function with HRM by using a deep learning computational model, namely, EMD-DL, which leverages three-dimensional convolution (Conv3D) and bidirectional convolutional long-short-term-memory (BiConvLSTM) models. More specifically, to fully exploit wet swallowing information, we establish an efficient swallowing representation method by localizing manometric features and swallowing box regressions from HRM. Then, EMD-DL learns how to identify major motility disorders, minor motility disorders and normal motility. To the best of our knowledge, this is the first attempt to use Conv3D and BiConvLSTM to predict esophageal motility function over esophageal HRM.

RESULTS

Test experiments on HRM datasets demonstrated that the overall accuracy of the proposed EMD-DL model is 91.32% with 90.5% sensitivity and 95.87% specificity. By leveraging information across swallowing motor cycles, our model can rapidly recognize esophageal motility function better than a gastroenterologist and lays the foundation for accurately diagnosing esophageal motility disorders in real time.

CONCLUSIONS

This approach opens new avenues for detecting and identifying esophageal motility function, thereby facilitating more efficient computer-aided diagnosis in clinical practice.