Neurological emergencies such as acute stroke cause time-dependent brain damage. But emergency treatments for a certain kind of acute stroke caused by loss of blood flow to the brain ("ischemic stroke") such as intravenous 'clot busting' tissue plasminogen activator (rtPA) are rarely given, in large part because of the lack of timely diagnosis of the patient.
As it turns out, the neurologist is the limiting factor.
Faucetworks is developing the Faucet™ artificial intelligence diagnostic (AID), a medical device that is capability of diagnosing neurological emergencies such as acute stroke in the ambulance. Rapid diagnosis of acute stroke in the ambulance will improve patient routing to appropriate hospitals and reduce the time needed in the hospital to treat the patient with IV rtPA. This will reduce brain damage, improving the patient’s neurological condition while facilitating use of existing therapies and enabling use of new therapies.
Here is what the Faucet will look like:
Our Faucet AID will be a cloud-based, semi-autonomous artificial neural network endowed with natural language and visual recognition capabilities that can diagnose neurological emergencies by direct communication with, and examination of, the patient. Once the diagnosis is established, the Faucet AID will then direct emergency treatment measures and guide patient transport to an appropriate hospital. Patients who have uncertain diagnoses will be referred to on-call physicians for evaluation. However, the machine learning capability of the Faucet AID increases its diagnostic accuracy over time, allowing for service expansion .
HOW THE FAUCET WORKS
Direct patient interaction leads to diagnosis and treatment of neurological emergencies in the ambulance. Neurological emergencies that are currently treatable, such as ischemic stroke, must be distinguished from similar conditions ("stroke mimic" conditions). Physician backup is always available but becomes less-frequently needed over time as the Faucet improves its diagnostic accuracy.
Faucet's patient interface provides data to a series of computational algorithms trained for analysis and diagnostic decision making. The computational algorithms are proprietary and have been trained on multiple patient databases.
Stroke is the most common cause of disability and the second most common cause of death globally. In the US, nearly 800,000 cases of stroke occur each year, costing approximately $37 B in healthcare expenditures. Worldwide, there are more than 16.9 M cases of stroke annually of which 5.7 M prove fatal. Ischemic stroke, which accounts for 85% of all stroke, currently can be treated in the ambulance with the intravenous ‘clot busting’ medicine, tissue plasminogen activator, if only the patient diagnosis were available in that setting.
By delivering the patient to the hospital with the diagnosis of acute stroke established, the patient then only needs routine neuroimaging to establish the AIS diagnosis leading to IV rtPA treatment. Doing so eliminates the need for in-hospital physician evaluation, saving on-average 83 minutes of time to IV rtPA treatment.
This time saved translates into 37% increase in rtPA effectiveness and a 3-fold increase in the number of treated patients.
Because it is scalable, the Faucet AID can be used in any and all of the approximately 81,000 ambulances in the U.S. and 178,000 ambulances in Europe. Emergency departments in hospitals that lack adequate physician services may similarly benefit from the Faucet AID.