Medicine is changing: A journey from general treatment methods to individual therapy. Artificial intelligence in conjunction with visual computing technologies offers completely new possibilities. Researchers at Fraunhofer IGD will present various applications of artificial intelligence in medicine at the CEBIT from June 11 to 15, 2018 in Hanover – along the entire treatment chain (hall 27, booth E78).
(Darmstadt/ Rostock, Germany) People are different. Medicine of the future seeks to take this point into account and treat each patient according to an individually optimized strategy. This is based on large data volumes from different data sources. Intelligent systems are increasingly relied upon to analyze and evaluate the data mountains. But how can humans make the right use of the data generated and processed by artificial intelligence? Visual computing technologies are an essential approach. At Fraunhofer IGD, researchers have been using methods and processes of machine learning and artificial intelligence for years to analyze and evaluate vital and health data as well as disease-related patient data and to develop technologies along the entire treatment chain.
AI for diagnoses
Artificial intelligence can do a lot at the beginning of the diagnosis. For example: When the patient visits the doctor. First of all, it is important to determine and diagnose the cause of symptoms – which is about recognizing and marking the shape, position and structure of body parts, organs, tissue or cells in medical image data. If the image data is three-dimensional such as MRI or CT, this is a complex undertaking that is virtually impossible to master manually. The scientists at Fraunhofer IGD have, therefore, developed corresponding machine learning processes: These can segment anatomical structures into image data simultaneously, completely and automatically, and graphically present the results to aid in the diagnosis.
AI in analysis: Effectively learning from patient and treatment data
Once this step is complete, the physician compares the patient’s findings with those of other people. If physicians want to analyze large amounts of patient data and thus make a more robust statement on their findings, they form cohorts, that is, patient groups with relevant similarities. But does the educated cohort produce what it promises? Or could it be refined? These questions can be answered using individually adapted visual analysis tools from Fraunhofer IGD which visualize the attributes, analyze them in detail and provide the attending physician with important insights for the treatment of the patient.
Augmented reality during operation:
The knowledge gained from the previous image and data analysis simplifies the determination of suitable treatment, which also uses visual computing technologies from Fraunhofer IGD. In the operating room, physicians have to prove their skill by being able to estimate the exact location of organs, blood vessels and diseased tissue during a procedure. The integration of an augmented reality system helps to remedy this situation and supports the physician with visual markings during the operation. In this case, the position of the organ is virtually superimposed using AR glasses.
AI in follow-up care: detecting anomalies in vital data
The patient continues to receive care after the operation – based on their vital signs. What about, for example, the quality of sleep and the stress level of the patient? Do anomalies such as sleep apnea or unconsciousness occur? This can be determined by analyzing vital signs, such as heart rate, heart rate variability or respiratory rate. A solution from Fraunhofer IGD comprehensively records the data, evaluates it continuously, and detects anomalies quickly. Multiple sensors and situation-dependent algorithms increase the robustness of the detection. The data is used to support in-home care but can also be incorporated into a visual control station for hospital staff also developed by Fraunhofer IGD, thus ensuring the central observation of the patient.