In general terms, the implementation of artificial intelligence (AI) benefits the approach and diagnosis of the patient being treated.
Healthcare management prior to the use of AI in the medical field was complex when talking about diseases with similar signs, difficult to diagnose and manage, which limited the healing process.
The mortality rates of cerebral vascular diseases (CVD) have reached high values.
Of the cerebral vascular diseases, it has caused more than 5 million deaths in the year with the highest recurrence (2005).
This is due to untimely diagnosis and poor management of patients at risk.
Artificial intelligence in medicine
The design of an AI is based on replicating and improving the understanding, reasoning and learning of human intelligence.
Collecting data and generating commands to be able to perform different tasks uniformly and without the need to be monitored.
Its function is relevant in medicine, since it is executed by means of the type of automatic learning with focus on the comparative process.
All the doctor needs to do is to enter the data to be analyzed, so that the AI can recognize and compare it with its own database.
Proceeding to identify the new information presented and at the same time provide an accurate result.
In addition, it is able to store the information in the program’s artificial neural network.
The more information collected in these networks, the greater the AI’s ability to correlate findings in new examinations.
This is something that cannot be adequately accomplished in traditional medical practice, due to the time it takes to analyze a diagnosis in person.
Implementation of Artificial Intelligence for the diagnosis of a BSS.
Artificial intelligences nowadays assist the radiologist and neurologist in order to assess the general state of the brain.
An example can be the identification of a cerebral vascular disease, of which there are 3 types of cerebral vascular diseases:
- Hemorrhagic: Occurs due to the rupture of a blood vessel in the cranial area.
It is identified in the presence of cranial trauma, aneurysms and uncontrolled arterial hypertension.
- Ischemic: due to clogging of the arterial flow to the central nervous system, mainly due to clot formation.
The narrowing of the vessels occurs due to calcification, fat accumulation and the presence of embolisms (generation of the clot in another area of the circulatory system).
- Transient: unlike the 2 previous ones, it tends to last minutes and does not generate permanent damage.
However, it is considered as a debut for those who have risk factors.
It is related to overweight, sedentary lifestyle and illicit drug use.
This disease has a higher recurrence in adults over 50 years of age.
The risk is elevated in people with heart disease, who have been diagnosed with hypertension or diabetes.
And in subjects with a family history of stroke, it usually develops within 24 hours or even less.
The presentation of clinical manifestations occurs before even exhibiting cerebral ischemia or hemorrhage, as in the case of transient CVE.
The identification of facial paralysis, limb numbness and disorganized speech are symptoms for prompt medical attention.
The imaging tests of choice (computed tomography – CT or magnetic resonance imaging – MRI) make it possible to search for the possible origin of a CVE.
In addition, the changes presented by the ASPECTS scale, where the middle cerebral artery (main artery supplying the brain) should be divided into 10 different sections, should be assessed.
One point is subtracted for each affected area depending on the areas through which it circulates.
Returning to artificial intelligences, imaging studies allow the elaboration of a result with the help of the aforementioned comparative process.
By analyzing the findings present in the brain CT scan, relying on the records already stored in its database and comparing them with those to be evaluated.
The program relies on stored images of BSS cases and information that clarifies the normal composition of the brain in order to confirm or deny a pathology.
It also allows to rule out diseases that exhibit similar signs as in the case of posterior encephalopathy syndrome.
That in the studies is manifested with edema at cerebral level, although there are similarities.
It is related to other risk factors, mainly preeclampsia, eclampsia and renal diseases.
In addition, it is able to differentiate the origin of the cerebral vascular event, whether hemorrhagic or ischemic.
A CT scan shows evidence of hemorrhage only, while MRI is more sensitive to finding both hemorrhagic and ischemic processes.
Other points to take into account in the diagnosis by artificial intelligence correspond to its capacity to prioritize those patients with a high risk of cerebral vascular diseases.
Even improving the quality of tomographic images presented in just 1 second.
Both processes considerably reduce patient examination time, which promotes early treatment.
It is relevant to know that the damage caused after cerebral vascular diseases can be irreparable.
The moment the patient presents an obstruction of blood flow or stroke, his neuronal function will no longer be optimal.
This leads to various complications, which may result in partial or complete paralysis, advanced cognitive impairment, personality and motor impairment, as well as possible loss of sensation.
Functioning of Artificial Intelligence in the study of brain imaging
Currently, there is e-Stroke Suite, which is a computer and mobile application designed by Brainomix.
By means of artificial intelligence, it contributes to the imaging diagnosis of the distinctive abnormalities presented in neurological disorders.
The service offered with this program allows the physician to facilitate the introduction to an efficient diagnosis and management in a short period of time.
It has several modalities, one of which is e-ASPECTS.
It allows evaluating and highlighting the degree of affectation and measures the volume of liquid present in a cerebral infarction automatically, it can detect occluded and calcified blood vessels.
In addition, the application offers the review of cerebral angiographies for the accurate finding of aneurysms in patients.
In addition, it offers the e-Stroke Cloud service, a cloud storage service (like OneDrive or Google Drive).
This makes available to medical centers that have e-Stroke Suite all the imaging diagnoses that were analyzed by the AI, allowing them to be viewed at any time.
Prognosis in the event of early diagnosis of cerebral vascular diseases
Early prediction of cerebral vascular disease allows treatment with thrombolytic drugs.
Where intravenous tissue plasminogen (TPA – IV) is usually the drug of choice because of its efficacy in dissolving clots, its effectiveness is greater if applied within the first 3 hours after symptom onset.
However, surgical treatment (thrombectomy) is indicated when the clot has already created an obstruction in the cerebral bloodstream and is causing motor disabilities.
Performing it within the first 6 hours reduces the degree of neurological involvement, in addition, before surgery TPA – IV is administered for a better management of the patient in the operating room.
The follow-up of these indications after the early finding allows the improvement of the patient who presented cerebral vascular diseases, avoiding most of the time serious damage to the brain cells and more importantly, decreases the time of hospital stay and accelerates the reincorporation to normal activities.
By means of AI, the Bayesian teaching method has begun to be applied to assess the probability of recovery of the stroke patient.
Based on brain, motor and reflex functioning, apart from being able to relate different complications that could occur over the years.
Allowing hospital centers to perform properly and in record time the diagnosis of cerebral vascular disease, which has an important impact on the treatment and prognosis of patients.
Written by: Ulises Bacilio
Follow us on Facebook