Overview 

Gastroesophageal reflux disease (GERD) is a prevalent digestive disorder that affects millions globally. Traditional diagnostic techniques, such as endoscopy and 24-hour pH monitoring, are often used to evaluate patients with GERD symptoms. However, many patients with normal endoscopy and pH measurements continue to experience symptoms and do not respond adequately to anti-secretory therapy.

Multichannel Intraluminal Impedance-pH (MII-pH) monitoring is an innovative technique that detects intraluminal bolus movement and identifies gastroesophageal refluxes (GER) without the use of radiation. First introduced by Silny et al. in 1991, MII-pH monitoring is now considered the gold standard for diagnosing GERD. This method not only assesses bolus transit but also determines the physical state of the bolus. Despite its clinical significance, there is limited biomedical research analyzing MII-pH data, highlighting a need for further software development in this area.

Data Acquisition 

For this dataset, 24-hour MII-pH monitoring was conducted on patients with normal endoscopy who did not respond to anti-secretory therapy. Data were archived at Noor and Hazrat-e-Ali Asghar Hospital in Isfahan, Iran. Each monitoring session lasted approximately 24 hours, utilizing a mobile recording device (Ohmega Impedance Ambulatory pH Meter; MMS, Enschede, Netherlands). 

The six-channel impedance-pH catheter was passed transnasally under topical anesthesia and positioned 5 cm above the lower esophageal sphincter (LES) to record pH at 5 cm and impedance at 3, 5, 7, 9, 15, and 17 cm proximal to the LES. Data from both the impedance channels and pH electrodes were stored on portable data recorders (Ohmega R; MMS B.V., Enschede, Netherlands) and processed using MMS database software for raw data format conversion. MII-pH data were collected at a sampling frequency of 50 Hz, then downloaded onto a computer.

We selected archived data from 26 patients of varying ages and sexes for this study. The MII signals allow for the identification of all types of GER, independent of pH measurements. To simplify the analysis and avoid multi-modal processing, the pH data were excluded, resulting in a dataset focused solely on unit impedance.

To ensure informative and balanced data, we selected at least four episodes for each 24-hour MII-pH signal, focusing on periods proximate to GER events. Each MII episode spans two minutes and contains swallows and at least one GER event. Characterizing GERs from MII data is crucial for diagnosing GERD. A total of 202 episodes were reviewed with the assistance of three experienced gastroenterologists to specify the characteristics of each GER event. To standardize the data, we normalized all episodes by dividing the impedance values of each channel by their corresponding L2 norm. A binary mask indicates GER events across all six channels.

Data Structure

This dataset was utilized in one of the challenges at the IAI_2023 event, with the corresponding paper available at [paper link]. The dataset is divided into three categories: training, testing, and hidden datasets. Each phase of the competition provided the dataset sequentially: first the training dataset, followed by the testing dataset, and finally, the hidden dataset was revealed during the live competition to validate the results.

The MII dataset comprises 202 episodes, including 208 GER events, collected from 26 patients. The training and testing sets originate from distinct individuals, while the hidden data was randomly sourced from any patient in the dataset. Each episode in the training and testing folders contains 6,001 rows, while each episode in the hidden folder contains 6,000 rows.

 To download the MII Dataset for Gastroesophageal Reflux Disease (GERD) please click hear 

Please reference the following paper if you would like to use any part of this dataset or method: