Since its inception, the development of artificial intelligence has been exposed to much scrutiny and even some mistrust from the scientific communities and especially the general public. However, the constant advances of AI tools have sought to overcome these obstacles to find solutions to the great problems of humanity.

In November 2018, the Duke University Health System Emergency Department launched “Sepsis Watch.” The tool was designed through deep learning to help professionals in the area detect the first signs of one of the leading causes of hospital death worldwide: infections and their overwhelming ability to wreak havoc on the human body.

The dreaded sepsis occurs when an infection triggers inflammation throughout the body, which can cause immediate -and multiple- organ failure. Fever, shortness of breath, low blood pressure, fast heartbeat, and mental confusion are just some of its symptoms. Although its effects are extremely harmful, the truth is that it can be treated with an early diagnosis. However, this is easier said than done since its early signs are often confused with other ailments.

Sepsis Watch is the product of three and a half years of development, during which medical records were digitized and 32 million data points were analyzed. Subsequently, the Duke University team focused on designing a simple interface so that the tool could be used in the form of an iPad app. The app checks each patient’s information and assigns them a rating based on their probability of developing the condition. Once a doctor confirms the diagnosis, an immediate treatment strategy is put in place.

The result is a drastic reduction in the deaths of patients from sepsis. Currently, the AI tool is part of a federally registered clinical trial. The preliminary results of which will be available by 2021.

VOYAGER: AI Tools solution for the health area made in Chile

Similar to the cases of death due to sepsis, arterial hypertension, Alzheimer’s, schizophrenia, retinitis pigmentosa, asthma and diabetes mellitus are pathologies with high mortality rates according to the WHO. Due to the complexity of their diagnosis, their treatment normally consists of rigid protocols, the results of which may vary from one patient to another.

VOYAGER, developed by UNIT, focuses on exponentially improving the management of these diseases, known as multifactorial. Through the use of artificial intelligence, the system is capable of processing data collected by voice interfaces to fully understand the status of each patient and perform predictive and automated monitoring of their treatment.

Similar to what happens with Sepsis Watch, this translates into more efficient diagnoses and identification of higher risk cases, directly impacting the fatality rate of these diseases. In concrete terms, VOYAGER’s goal is to reduce serious hospitalizations by 50% for those suffering from diabetes, cerebrovascular diseases, hypertension and even obesity, both in public and private health.

 

 

AvatarCarla Espinoza Gutiérrez

Journalist and Community Manager. I am passionate about delivering the right message for different kinds of audiences, always evolving and refreshing the tools, shape and tone of delivery.