Tracheal sound analysis for detection of respiratory depressions in adult patients during cataract surgery under sedation analgesia
Increasingly, both surgical and non-surgical procedures—such as dental work, endoscopy, cosmetic surgery, and cataract surgery—are performed under sedation analgesia using a combination of sedative and narcotic drugs. These agents must be accurately titrated to meet individual patient needs, with close monitoring of their effects on respiratory function.
Common monitoring techniques, such as pulse oximetry, often have a considerable delay in detecting respiratory complications. Additionally, side-stream capnography has limited effectiveness for detecting respiratory depression due to challenges such as sampling errors, lumen obstruction by airway secretions, and frequent detachment from the patient’s airway. Therefore, direct monitoring of airway patency using auscultatory techniques is crucial during sedation analgesia.
Continuous monitoring of tracheal sounds with traditional or electronic stethoscopes can reliably and rapidly detect airway complications before they lead to serious issues. However, the overall efficacy of tracheal stethoscopes as airway monitors depends on the continuous listening to respiratory sounds by the anesthesia team. Continuous operator listening may not be practical, and interrupted auscultation can be associated with significant airway problems. Therefore, the development of real-time, automatic, and continuous techniques for tracheal sound monitoring and analysis is essential for effective airway monitoring during sedation analgesia.
Tracheal Sound Data Acquisition
Following approval from the institutional ethical committee and informed consent from the patients, we conducted a study involving 16 adults classified as ASA I and II, all scheduled for cataract surgery under sedation anesthesia. Patients with a history of respiratory diseases were excluded from the study.
Upon positioning on the operating table, all patients received supplemental oxygen via a mask. Monitoring included ECG, non-invasive blood pressure (NIBP), and pulse oximetry. Tracheal sound recording commenced one minute prior to the administration of sedative drugs using a C417 omni-directional condenser Lavalier microphone (AKG Acoustics, Vienna, Austria), secured over the suprasternal notch with double-sided adhesive tape.
Tracheal sound recording continued throughout the procedure at a sampling rate of 44,100 Hz. After the study's completion, the recorded sounds were analyzed by the anesthesiologist to identify periods of respiratory depression, defined as episodes of apnea, breath-holding, or airway obstruction lasting longer than 10 seconds.
Data Structure
In the initial phase of the challenge, competitors will have access to data from 11 subjects, along with their corresponding labels.
The dataset will include 11 .wav files.
The labels will be provided in separate .docx files (label_S1.docx belongs to S1.wav and so forth).
Each .docx file will contain the onset and offset timings of respiratory depression intervals.
Submission Requirements
The primary goal of the challenge is to utilize AI-based algorithms to analyze tracheal sound data for detecting areas indicative of respiratory depressions. Participants are required to submit the results of their algorithm in word file format, following the same structure and format as the provided labels. Each word file should correspond to the results for a specific subject, named accordingly (e.g., results_S1.docx).
1. Utilize AI-based algorithms: Competitors must use artificial intelligence algorithms to analyze tracheal sound data for detecting areas indicative of respiratory depressions.
2. Data preprocessing: Participants should articulate their approach to preprocess the provided tracheal sound data before applying the detection algorithms.
3. Algorithm development: Competitors are required to develop algorithms that can accurately detect and classify areas in the tracheal sound data corresponding to respiratory depressions.
4. Performance evaluation: Submissions must include a detailed description of the performance metrics used to evaluate the efficiency and accuracy of the algorithms in detecting respiratory depressions.
5. Documentation: Competitors should provide clear and detailed documentation of their algorithms, including the methodology used, implementation details, and any specific considerations taken into account during the development process.
6. Results presentation: Participants are expected to present the results of their algorithm, including visual representations of the detected areas in the tracheal sound data related to respiratory depressions.
Evaluation Policy
We will compute the following parameters for the received files compared to the ground-truth annotations.
Total number of TP, FP and FN events,
sensitivity and Positive Predictive Value
To download the MII Dataset for Gastroesophageal Reflux Disease (GERD), please click here
If you wish to use any part of this dataset, kindly reference the following paper:-