Data Science and AI for better healthcare systems
Customer Visit Center
We started off with a visit to the customer Visitor Center (CVC). This is the location where Philips shows its customers and partners the range of products and services. The devices and services are arranged along the Care Cycle. This is the philosophy of taking care of the “customer” during all phases of live and all phases of the disease. Starts with healthy living, proper diagnosis, treatment (hospital) and home care, etc. as shown in Fig.1.
Figure 1 Healthcare Cycle in CVC Philips Healthcare Best.
Collecting data for health
In many cases this means collecting data to improve behavior like during toothbrushing. But also, data related to the health complaints combined with available data from comparable complaints to optimally diagnose the problem. Sometimes an operation may be needed, but we were shown the advancement of minimally invasive treatment to solve problems in a less damaging way and with faster recovery. To this end info from the analysis equipment like Ultrasound, MRI and PET etc. is combined in an easy to use way. Data plays an important role to know the ins and outs of a patient. By collecting them a model, digital twin, of the patient is defined.
Collecting data for healthcare systems
The second part of the visit was also about collecting data, but now from the healthcare systems. Mauro Barbieri, scientist at Philips Research gave a lecture on the activity to increase the availability and speed and reliability of healthcare systems by collecting data, building device models and using data science and AI. All this is enabled by the progress in parallel processing, storage technology and special data bases to be able to manage the Big Data and apply Data Science for predicting machine conditions.
MRI, CT and X-ray systems
MRI, CT and X-ray systems are important systems in our healthcare. Thousands of patients are scanned every day. Essential information on the health of the patients is obtained by each of them and combined with data from other sources a diagnosis can be deduced, and a treatment be defined. The quality and reliability of these systems needs to be optimal. At Philips the options to guarantee quality and reliability are topic of research. Techniques to monitor the systems and do preventive maintenance are being investigated.
Data logging, remote monitoring and AI
Every day, a typical MRI scanner produces an average of 800,000 log messages, which reflect how the system is working technically. Through proactive remote monitoring services at Philips these log messages are tracked and analyzed for early warning signs of impending technical issues. (To be clear: these analyses are conducted on technical data only, obtained in agreement with customers.)
To start developing these services within Philips, technical data from over 15,000 MRI, CT and interventional X-ray systems were collected and billions of data points were analyzed. After curating the data, using machine learning (AI) and other analytical methods patterns could be identified that foreshadow specific impending issues.
For proactive remote monitoring more than AI is needed
One of the elements in proactive monitoring is a physical model of the system to analyze the remote data from sensors in the system. Such a digital twin helps to test and verify the remote data and work on solutions remotely.
However, to make sense of the data, you also need intimate knowledge of that device. AI can help detect patterns in data, but it cannot tell whether these patterns are practically meaningful. You need human knowledge and understanding of the underlying technology for that. AI augments human capabilities! Therefore, Philips prefers to speak of adaptive intelligence.
Research and exploitation
The research of AI and data analytics for adaptive remote control will be discussed. Also, the impact on the application in the hospital gets attention.
Customer Visit Center
For Philips it is important to show the capabilities of the healthcare products to the customers. The Customer Visit Center serves that purpose. We will be sown around to learn about the new Philips Healthcare strategy and see the relevant healthcare systems.
Speaker:
Dr. Mauro Barbieri, Senior Scientist, Philips Research with presentation at 17:00 h in QS Auditorium:
'Towards Zero Unplanned Downtime of Medical Imaging Systems Using Big Data'
Interesting Links:
Using AI to streamline remote communication in healthcare Bit&Chips May 9, 2019
Minister Sigrid Kaag Buitenlandse Handel en Ontwikkelingssamenwerking visti Philips Healthcare in Best at January 31, 2019
Digital Twin in healthcare, blog :
Philips Radiologie / Oncologie
Name and contact for further information: Dr.Ir. Steven B. Luitjens e-mail luitjenssb@hccnet.nl
