Artificial Intelligence (AI) in healthcare could bring better forms of therapy for millions of people and increase their quality of life. In this context, in January, ECO on SRF Swiss television dedicated a segment to the revolutionary AI-based health application called ‘Predictive healthcare with real-world evidence for neurological disorders’ (PHREND). The transmission also featured the development team around Amfin Bergmann, neurologist and MS specialist of NeuroTransData (NDT), and Christian Westermann of PwC Switzerland.
AI in healthcare: enormous potential
AI and self-learning systems are becoming increasingly popular, be it in self-driving cars, robot-assisted surgery systems, predictive error detection in production processes or digitalised customer care. All these applications are based on machine learning and intelligent algorithms. The research, development and implementation of AI is a top priority in China, France, Singapore or Japan, for instance; Switzerland, on the other hand, is yet to bridge the gap.
PwC studies reveal that on an international level, Swiss enterprises are lagging far behind when it comes to AI. A total of 15 per cent of companies worldwide are employing Artificial Intelligence, the number in Switzerland, however, is as low as 1 per cent. Unlike other countries, Switzerland has no overarching federal AI strategy, and some companies are more than hesitant to embrace the digital transformation. To make sure that Swiss enterprises will not miss the boat and will be able to exploit the potential of these key technologies, the Swiss government must create a clear regulatory framework and promote education and research.
The healthcare system in particular has a lot of potential for AI applications. Self-learning systems are continuing to evolve and are increasingly applied in many fields of medicine such as image analysis, to mention just one. AI is much faster at comparing and analysing images or detecting anomalies than any physician. Drug development as well gives a key role to AI as it may shorten research cycles quite considerably. Health insurers and hospitals can benefit from the automatisation of processes as it makes them more efficient and lowers costs. The therapy decision support tool – launched by PwC – is therefore a promising AI initiative in healthcare.
AI is the way forward to personalised MS therapy
‘Predictive healthcare with real-world evidence for neurological disorders’ is the first web application worldwide that provides physicians and patients with a data-based second opinion concerning the efficacy of personalised forms of treatment of multiple sclerosis (MS) at the mere touch of a button. The AI-based software was developed by PwC in collaboration with the NeuroTransData (NTD) network of physicians. It was introduced to the Swiss public in ECO on SRF Swiss television, alongside the development team around NTD neurologist and MS specialist Amfin Bergmann, and Christian Westermann, leader of Data & Analytics at PwC Switzerland.
SRF ECO – on 21 January 2019, Swiss television dedicated a segment to 'Healthcare: Individualised MS therapy thanks to Artificial Intelligence'
‘Predictive healthcare with real-world evidence for neurological disorders’ is the world’s first predictive and personalised software that can calculate the individual effects of any given MS treatment on a patient. It is driven by an algorithm that is based on the data collected from different forms of therapy of 25’000 patients over the course of more than ten years. It can precisely predict the efficacy of any personalised form of treatment.
In order to predict the efficacy of a suggested form of treatment, physicians simply enter a patient’s data, e.g. gender, age, time of MS diagnosis, current form of therapy and previous forms of treatment, EDSS value, time of last exacerbation and number of exacerbations over the past twelve months. Based on these factors, the software was calibrated in such a way that new patients with similar characteristics will be given a prediction of their likely progression of disease after therapy.
Amfin Bergmann explains: «By means of this algorithm based on the retrospective data, we can take a look into the future. In other words, we can predict the likelihood of a patient remaining exacerbation-free on specific medication, over a defined period of time of two to four years.»
Not only does the solution improve patients’ quality of life, it also lowers healthcare costs, as elaborate trial and error procedures will no longer be necessary. It is currently exclusively used for therapy purposes in relapsing multiple sclerosis, however, there are plans to cover the entire MS spectrum as well as other neuro-degenerative conditions. Christian Westermann affirms: «The platform's principle is applicable to a multitude of medical conditions, not just neuro-degenerative diseases such as MS, migraines or Parkinson’s.»
The right form of therapy without further ado
The platform's Artificial Intelligence acts as a physician’s right hand and facilitates the selection of the best form of treatment for the patient.
‘Predictive healthcare with real-world evidence for neurological disorders’ (PHREND)...
- reliably predicts disease progression under a variety of forms of MS treatment
- can deliver a personalised comparison between different forms of therapy
- provides physicians and patients with a second opinion and therefore increases the trust and transparency of therapy selection
- increases patients’ quality of life and lowers healthcare costs