Fujitsu is introducing a new artificial intelligence (AI)-based solution that helps process unstructured medical notes faster into a patient's medical record. The company claims that this process can be accelerated by up to 90% using the solution.
Structured information plays a crucial role in medical decision-making and healthcare in general. However, medical professionals are under increasing time pressure and have less and less time to spend per patient. At the same time, medical professionals are expected to process medical notes directly into the patient's medical record, something that quickly takes 15 minutes per patient. The solution streamlines this process by providing more flexible options for this purpose and automating this process to a large extent.
More time for patients
For instance, with the help of the solution, doctors and healthcare providers can have their medical notes analysed automatically and incorporated into the patient's medical record. As a result, they spend less time updating this record and can spend more time with the patient, e.g. collecting more useful data about the patient.
The new technology was announced by Fujitsu at the annual Fujitsu Innovation Gathering in London, UK. The technology combines semantic knowledge and Natural Language Processing (NLP) with deep learning to not only recognise medical notes, but also distil value from them.
Incidentally, solutions that allow automated processing of medical notes in medical records have been available for some time. However, according to Fujitsu, these solutions have their limitations. For example, healthcare providers have to compile their notes in a specific way to make them understandable to this software. This is not necessary with Fujitsu's AI solution; the solution can be trained to meet specific user needs. This gives healthcare providers more flexibility and, according to Fujitsu, ensures that data is recorded much more accurately in medical records.
'Measurable improvements'
Within the project, Fujitsu is working closely with healthcare partners, including San Carlos Medical Centre. "We are constantly looking for new ways to improve clinical decision-making, and our work with Fujitsu Laboratories of Europe is helping us take important steps forward in improving efficiency. Most electronic health record systems do not meet the requirements of the doctor-patient relationship. Indeed, several studies directly link the use of electronic medical records to clinical burnouts. With new technologies like Fujitsu's latest AI-based 'text mining' software, we can directly address these challenges and realise measurable improvements in medical decision-making," said Dr Julio Mayol, Chief Medical Officer, of San Carlos Medical Centre.
Dr Adel Rouz, Chief Executive Officer of Fujitsu Laboratories of Europe, says: "Our co-creation strategy with partners such as San Carlos Medical Centre has given us important insights into the challenges facing healthcare, especially in supporting clinical decision-making. We have been able to realise a number of key innovations that help us make a difference in the workflow of medical professionals. This latest development is the next step and helps us improve the accuracy of clinical data and automate the digitisation of hospitals, health insurers and government services. We expect our technology has broad applications and can be easily adapted to solve similar challenges in other domains such as insurance, legal and compliance."
IBM Watson
Fujitsu is not the only party deploying AI in healthcare. In 2014, for instance, IBM introduced Watson, which during a presentation managed to analyse a series of a patient's symptoms and provide a number of possible diagnoses based on these. Each of these diagnoses was accompanied by a certainty percentage and referred to medical sources for substantiation.
Watson has since continued to evolve, with successful and less successful results. In 2017, for instance, IBM Watson managed to analyse genome data 960 times faster than was possible until then. However, there were also lesser successes. For example, documents obtained by the website STAT show that a collaboration between the Memorial Sloan Kettering Cancer Centre and IBM Watson to diagnose and treat cancer patients was unsuccessful. According to the documents, IBM Watson allegedly gave wrong advice on a regular basis.
Several doctors and professors criticised IBM Watson to IEEE Spectrum early this year. One example is Herbert Chase, professor of Medical Sciences and Biomedical Informatics at Columbia University. Chase himself was previously involved in the initial development of Watson. "Doctors - especially those on the front line, primary care - go to work every day with the knowledge that they cannot possibly know everything they need to know to deliver the best, most efficient and most effective care," Chase said.
The AI would not deliver on its promise to transform the medical industry. Chase therefore says he is disappointed with IBM Watson's slow development in the healthcare field. "I am not aware of any spectacular home runs," said the professor.
DeepMind
DeepMind, part of Google's parent company Alphabet, is also working on AI for healthcare. In collaboration with Moorfields Eye Hospital in the UK capital London, DeepMind has developed an AI system that can diagnose 50 eye diseases as accurately as specialist ophthalmologists. The system helps determine which treatment will benefit patients most and with what priority patients should be treated.
Author: Wouter Hoeffnagel
