Using AI to Save the Lives of Psychiatric Patients! Fujitsu Halves the Time Required to Diagnose Patients in Joint Field Trial with San Carlos Hospital

A Doctor's Quick Decision is Crucial in Psychiatric Treatment

Projects that apply artificial intelligence (AI) to, for example, the development of new drugs or the discovery of new treatment methods, are underway across the globe. In Spain, a field trial using AI was conducted with the objective to protect psychiatric patients from suicide or other health risks.

A failure to provide early and appropriate prevention or treatment for a mental illness may result in it turning into a chronic illness or even death. Doctors must therefore make appropriate and quick decisions while properly understanding the health risks subject to the patient. This requires consideration of patient symptoms from many sources of information, instead of only clinical examination history. Traditionally, doctors had to read materials printed in various formats. This took them hours to just select the patient records necessary to make a diagnosis.

An AI Analyzes Patient and Open Data, and Shows a Patient's Potential Health Risks in Seconds.

Advanced Clinical Research Information System at the San Carlos Hospital

To support doctors in making quick diagnostic decisions, San Carlos Hospital in Madrid, Fujitsu Laboratories of Europe, and Fujitsu Spain performed more than six months of AI field trials. These trials resulted in the development of the Advanced Clinical Research Information System.

In this system, the medical histories of over 36,000 patients, completely anonymized for privacy reasons, and a large volume of open data, including over a million medical papers, are combined in a database. An AI engine analyzes the database and shows possible health risks for a patient.

Semantic modeling technology converts patient information into a graphical data model with a semantic structure. Graph data analysis technology analyzes patient data expressed in graphical form. These technologies together present the doctor with a patient's potential health risks, such as suicide and alcohol or drug dependence within a few seconds after the data is entered.

An example of the Advanced Clinical Research Information System’s screen. Doctors can check a patient's potential health risks within five seconds after entering their information into the fields on the left.

The System Halves the Diagnosis Time! It can also Identify Patient Health Risks with Higher Accuracy

Five specialists at the San Carlos Hospital, each with nearly 20 years of experience, evaluated the system. As a result, it successfully calculated the risks of suicide, alcohol dependence, and drug dependence with an accuracy of 85% or higher.

Furthermore, the ability to check medical records and various types of information on the system instead of in written documents allowed doctors to make diagnoses 50% faster for each patient. With this system, doctors can spend more time on interviewing patients and identify their potential risks with higher accuracy.

For Creation of New Medical APIs

Co-creation with San Carlos Hospital has indicated that using AI to support medical decision-making processes has profound significance. This new technology will be the foundation of future medical application programming interfaces (APIs) in Fujitsu Spain's cloud or private cloud services.

Fujitsu will continue to contribute to society by developing technology to realize medical APIs and applying advanced AI technology to the latest solutions.

The system was introduced at the Fujitsu Forum 2016 Munich on November 16 and 17, 2016.