Artificial Intelligence and Machine Learning in Anesthesia

Depth of Anesthesia monitors helps individualize anesthesia by permitting accurate drug administration against the measured state of arousal of the patient.

About Our Product

We are offering a package that will help Anesthesiologists to get the best anesthesia dosage. Our product is a “Processor” which will help Anesthetic agents to get more accurate results and get an accurate DoA. The Processor will include hardware and an application (AI).

How it Works

For accurate monitoring of DoA, novel feature extraction is used to convert the EEG signal and combine with EMG signal reduction. Our machine learning based processor is used to facilitate the Closed Loop system. There are bound to be redundancies in the current system used by our clients, to be replaced by our system. We expect the efficiencies will outweigh the price paid to implement our Processor.

Accurate Monitoring

Accurate assessment of the depth of anesthesia and accurate monitoring contributes to tailoring drug administration to the individual patient and it is essential for intraoperative and postoperative patient’s health. A patient could be underdosed or overdosed because of equipment failure or error. On the contrary, inappropriate titration of the hypnotic components, leading to an excessive Depth of Anesthesia (DoA), might compromise patient outcomes.

Better Drug Administration

Depth of Anesthesia monitors might help to individualize anesthesia by permitting accurate drug administration against the measured state of arousal of the patient. Various depth of anesthesia monitors based on processed analysis of the EEG or mid-latency auditory-evoked potentials are commercially available as measures of anesthetic drug effect. However, not all of them are validated to the same extent. Depth of anesthesia monitoring becomes more compelling if it can be used to guide the clinician to the “sweet spot” where the anesthetic dose is sufficient to prevent awareness but not greater than needed.

N

Our Product Specifications

  • Ability to display and perform analysis and classification (all in Realtime mode).
  • Ability to display multichannel EEG signals up to 8 channels.
  • Ability to remove Baseline from EEG signals.
  • Ability to extract and display different frequency band signals from the main signal.
  • Ability to calculate PSD (Power Spectrum Density).
  • Ability to calculate 95% SEM features, Beta Ration, and power of different frequency bands.
  • Use fuzzy classifier and scoring system to detect DOA.
N

Our Competitive Advantages

  • Currently no similar product in the market such as our processor. 
  • Our processor can calculate accurate DoA estimation irrespective of the patient’s age and anesthetic drug.
  • For our hardware implementation of DOA estimation algorithms, will be programmed using VHDL.
  • The processor has proved to be capable of achieving a 100% accuracy for all DoA states.
  • The processor can be integrated into an existing setup, with minimal redundancies. 

Benefits of Our Solution

Assists the Preoperative Assessment

This includes a thorough surgical overview, medical history, physical exam, lab tests, and identification of specific cardiac and pulmonary risk factors, with the goal of reducing perioperative risks and improving outcomes. This is a crucial aspect of presurgical care that has evidence-based prognostic consequences. For example, postoperative lung complications can be predicted by pre-existing chronic lung disease, severe asthma, smoking status, and other relevant characteristics, allowing physicians to stratify patients into risk levels. Subsequently, anesthesiologists may opt to modify their anesthetic choice and dosage, or perhaps attempt to optimize the condition of the patient before proceeding with the surgery according to the characteristics obtained in the preoperative assessment.

Fine-tuning

Through the course of a procedure, an AI-based closed-loop system would make granular adjustments to the administered anesthetic in real-time according to changes in the DoA measured by the equipment, and addition of new drugs. This type of AI would prove to be most efficient for an anesthesiologist who would now be able to monitor other key aspects of the patient’s anesthetic condition during surgery.

Transfer of workload

The process of knowledge integration for risk stratification was traditionally the sole responsibility of the physician. Machine learning will make this task accurate, efficient and timely.

Consolidation of functions

AI may assist the physician in higher-order knowledge integration with the experience of thousands of medical procedures that a single person would not be able to integrate alone. AI also plays a crucial role in validating the robustness of its own outputs. In having access to limitless medical case studies, programs can be cross-trained from various data sets to test predictive accuracy.

Bottom-up approach

AI comes with the promise of self-sufficient and adaptable systems that can teach themselves through a bottom-up approach, one that is pre-emptively given medical information from previous surgeries and real-time data about the patient to form perception and malleable output that makes sense according to prior evidence and the current condition of the patient.

What is Our Solution

Learn More About Our Product

PROBLEM STATEMENT

Most of the alarm signs are seen on the monitor only after an event occurs in the surgical field. Many techniques and clinical indices such as blood pressure and heart rate have been used to indicate DoA. However, they are reactive measures and have drawbacks and are unreliable for assessing DoA. There have been many attempts to incorporate automation into the practice of anesthesiology, though very few have been fully successful. Fundamentally, these failures are due to the underlying complexity of anesthesia practice and the inability of a rule-based feedback loop to fully master it. There is potential to cause irreparable harm to the patient, by failing to detect and measure awareness and adjust the dose of Anesthesia in real-time. Our monitoring recommendations address the appropriate role for our technology in clinical practice which could prove more reliable at no extra cost to the administrator due to redundancy of the older machines.

CONSEQUENCES OF THE PROBLEM

The neural function is essential for cognition and to the hallmark of human existence. A permanent loss of neural function because of anesthesia complications during surgery is a major loss to the individual. If the mechanical or physiologic injury to these structures is suspected, neuromonitoring can alert the anesthesiologist and surgeon to allow modification of treatment strategies. Accidental awareness under general anesthesia (AAGA) is a potentially devastating complication due to inadequate depth of anesthesia. AAGA is estimated to occur in 0.2% of adults receiving general anesthesia and potentially greater in children. Total Intravenous Anesthesia (TIVA) has a particularly high risk of awareness, as there is no real-time measure like exhaled agent concentration to measure the anesthetic load in vivo. Depth of anesthesia monitoring may also be used to prevent excessively deep anesthesia, which may be associated with delayed emergence from anesthesia and increased risk of perioperative complications. Excessive deep anesthesia in high-risk patients (pre-existing neurocognitive disorders, cerebral vascular disease, frailty, etc.) is associated with the development of postoperative delirium and postoperative cognitive dysfunction.

PROBLEM ANALYSIS

Several metanalyses have shown that automated systems can be more effective than a human in attaining tight control within a specified range of target variables and can also result in lower doses of delivered anesthetic with reduced recovery time. The American Society of Anesthesiology has established guidelines and promotes minimum practice standards of monitoring cardiopulmonary function and perfusion. These include, observation of physiological parameters such as blood pressure (BP), heart rate (HR), body temperature, heart rate, and rhythm, pulse oxygen saturation (SpO2), and respiration rate to estimate Depth of Anesthesia and ventilation. Neuromonitoring techniques include cerebral oxygen monitoring, processed electroencephalography (pEEG), electromyography (EMG), evoked potentials (EP), somatosensory EP(SSEP), brainstem auditory EP, and visual EP. Due to the central nervous system (CNS) affected by the anesthetic drugs, the electroencephalogram (EEG) originating in CNS has been focused on by researchers. The EEG reflects the brain activities and contains lots of information about anesthesia, so it has been widely used to assess DoA. In recent decades, numerous EEG-based methods have been proposed to develop an index to assess the anesthetic drug effects during general anesthesia.

We are 24/7 available

Contact Us We Love to Hear From You.

Office Address

Spark Centre Head Office
Suite 300,
2 Simcoe Street South,
Oshawa,
L1H 8C1
Canada

We are 24/7 available

How Can We Help You?