A Procedure for the Automatic Analysis of High-Resolution Manometry Data to Support the Clinical Diagnosis of Esophageal Motility Disorders

Mixed attention ensemble for esophageal motility disorders classification

Esophageal motility disorders result from dysfunction of the lower esophageal sphincter and abnormalities in esophageal peristalsis, often presenting symptoms such as dysphagia, chest pain, or heartburn. High-resolution esophageal manometry currently serves as the primary diagnostic method for these disorders, but it has some shortcomings including technical complexity, high demands on diagnosticians, and time-consuming diagnostic process. Therefore, based on ensemble learning with a mixed voting mechanism and multi-dimensional attention enhancement mechanism, a classification model for esophageal motility disorders is proposed and named mixed attention ensemble(MAE) in this paper, which integrates four distinct base models, utilizing a multi-dimensional attention mechanism to extract important features and being weighted with a mixed voting mechanism. We conducted extensive experiments through exploring three different voting strategies and validating our approach on our proprietary dataset. The MAE model outperforms traditional voting ensembles on multiple metrics, achieving an accuracy of 98.48% while preserving a low parameter. The experimental results demonstrate the effectiveness of our method, providing valuable reference to pre-diagnosis for physicians.

Biomedical data analytics for better patient outcomes

Medical professionals today have access to immense amounts of data, which enables them to make decisions that enhance patient care and treatment efficacy. This innovative strategy can improve global health care by bridging the divide between clinical practice and medical research. This paper reviews biomedical developments aimed at improving patient outcomes by addressing three main questions regarding techniques, data sources and challenges. The review includes peer-reviewed articles from 2018 to 2023, found via systematic searches in PubMed, Scopus and Google Scholar. The results show diverse disease-specific applications. Challenges such as data quality and ethics are discussed, underscoring data analytics' potential for patient-focused health care. The review concludes that successful implementation requires addressing gaps, collaboration and innovation in biomedical science and data analytics.

Own Experience in the Use of Artificial Intelligence Technologies in the Diagnosis of Esophageal Achalasia

Aim: to demonstrate an artificial intelligence model that optimises the differential diagnosis of achalasia.

Material and methods. The study included 75 patients: 52 % men (mean age 44.5 ± 17.8 years) and 48 % women (mean age 45.6 ± 16.6 years,) with a preliminary diagnosis of achalasia. Patients were divided into four groups: type I, II, III achalasia and a group of patients whose results did not correspond to a diagnosis of achalasia according to HRM performed based on Chicago Classification version 4.0. On the basis of a set of data from 750 swallows and therefore 6750 manometric parameters, the artificial intelligence models DecisionTreeClassifier, RandomForestClassifier and CatBoostClassifier have been trained to provide a manometric diagnosis. The comparison criteria were the training time and the f1_score metric. The technical characteristics of the model (hyperparameters) were selected using the GridSearchCV method. The model with the best results was integrated into a web application.

Results. The RandomForestClassifier was chosen as the best performing model to compare. Its technical characteristics were “decision trees” and branching depth the number of which was 14 and 5 respectively. With a maximum possible value of 1.0, these hyperparameters achieved f1_score=0.91 in 27 seconds. The web application, developed on the basis of this model, is capable of analyzing manometric data and establishing one of three types of achalasia in patients. Alternatively, it can exclude the diagnosis of achalasia. The output of an image corresponding to the diagnosis is produced for each manometric type of the disease.

Conclusions. For the first time in Russia, a machine learning model based on high-resolution esophageal manometry data was developed at the V. Kh. Vasilenko Clinic of Internal Disease Propedeutics, Gastroenterology, and Hepatology of Sechenov University. The model has been applied to the creation of a web application which has the ability to substantiate the manometry diagnosis of patients. The Federal Service for Intellectual Property (Rospatent) issued a certificate of state registration of the computer program No. 2024665795 dated July 5, 2024. This artificial intelligence programme can be used in clinical practice as a medical decision support tool to optimize the process of differential diagnosis of achalasia and early detection of the disease, to determine the patient's prognosis and to select the method of further treatment.

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.

Mechanical Behavior of Subcutaneous and Visceral Abdominal Adipose Tissue in Patients with Obesity

The mechanical characterization of adipose tissues is important for various medical purposes, including plastic surgery and biomechanical applications, such as computational human body models for the simulation of surgical procedures or injury prediction, for example, in the evaluation of vehicle crashworthiness. In this context, the measurement of human subcutaneous adipose tissue (SAT) and visceral adipose tissue (VAT) mechanical properties in relation to subject characteristics may be really relevant. The aim of this work was to properly characterize the mechanical response of adipose tissues in patients with obesity. Then, the data were exploited to develop a reliable finite element model of the adipose tissues characterized by a constitutive material model that accounted for nonlinear elasticity and time dependence. Mechanical tests have been performed on both SAT and VAT specimens, which have been harvested from patients with severe obesity during standard laparoscopic sleeve gastrectomy intervention. The experimental campaign included indentation tests, which permitted us to obtain the initial/final indentation stiffnesses for each specimen. Statistical results revealed a higher statistical stiffness in SAT than in VAT, with an initial/final indentation stiffness of 1.65 (SD ± 0.29) N/30.30 (SD ± 20) N compared to 1.29 (SD ± 0.30) N/21.00 (SD ± 16) N. Moreover, the results showed that gender, BMI, and age did not significantly affect the stiffness. The experimental results were used in the identification of the constitutive parameters to be inserted in the constitutive material model. Such constitutive characterization of VAT and SAT mechanics can be the starting point for the future development of more accurate computational models of the human adipose tissue and, in general, of the human body for the optimization of numerous medical and biomechanical procedures and applications.

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.

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.

Utility of a new automated diagnostic program in high-resolution esophageal manometry

BACKGROUND

A new automated diagnostic program for high-resolution esophageal manometry (HREM) has been developed. This diagnostic program could detect locations of landmarks and could make final diagnoses automatically. However, the accuracy of the program is not known. The aim of this study was to evaluate the accuracy of the automated diagnostic program for HREM.

METHODS

A total of 445 studies were enrolled. An HREM system (Starlet®) was used, and esophageal motility was diagnosed using the Chicago classification v3.0. First, the locations of the upper esophageal sphincter, transition zone, lower esophageal sphincter, esophago-gastric junction, crural diaphragm and stomach were determined, and each swallow was checked manually. Then, the parameters of the Chicago classification were calculated using an analytic program of the Starlet, and diagnoses were made by three experts. Second, all study raw data were analyzed again by the automated diagnostic program. Diagnoses made by the program were compared to those made by experts to evaluate the accuracy of the diagnoses.

RESULTS

The new diagnostic program could identify the landmarks of each swallow, calculate the parameters and make a final diagnosis within 10 s. The diagnoses made by the automated diagnostic program were not matched to those made by experts in only 10 studies, and the overall accuracy of the new automated diagnostic program thus reached 97.8% (435/445).

CONCLUSIONS

The new automated diagnostic program for HREM is clinically useful in terms of high diagnostic accuracy and time-saving.

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

High-resolution manometry (HRM) is the primary method for diagnosing esophageal motility disorders and its interpretation and classification are based on variables (features) from data of each swallow. Modeling and learning the semantics directly from raw swallow data could not only help automate the feature extraction, but also alleviate the bias from pre-defined features. With more than 32-thousand raw swallow data, a generative model using the approach of variational auto-encoder (VAE) was developed, which, to our knowledge, is the first deep-learning-based unsupervised model on raw esophageal manometry data. The VAE model was reformulated to include different types of loss motivated by domain knowledge and tuned with different hyper-parameters. Training of the VAE model was found sensitive on the learning rate and hence the evidence lower bound objective (ELBO) was further scaled by the data dimension. Case studies showed that the dimensionality of latent space have a big impact on the learned semantics. In particular, cases with 4-dimensional latent variables were found to encode various physiologically meaningful contraction patterns, including strength, propagation pattern as well as sphincter relaxation. Cases with so-called hybrid L2 loss seemed to better capture the coherence of contraction/relaxation transition. Discriminating capability was further evaluated using simple linear discriminative analysis (LDA) on predicting swallow type and swallow pressurization, which yields clustering patterns consistent with clinical impression. The current work on modeling and understanding swallow-level data will guide the development of study-level models for automatic diagnosis as the next stage.