Artificial intelligence in healthcare has unimaginable potential. Within the next two years, Big Data will revolutionize every area of our lives, including medicine. It will completely redesign healthcare, and for the better. Let's take a look at the promising solutions it offers.
There are many thought leaders who believe we are experiencing the Fourth Industrial Revolution. A “revolution” characterized by technologies that are merging the physical, digital and biological world. It is like a tsunami wave that affects all disciplines, economies and industries and even the biological boundaries of human beings. I am sure that healthcare will be the main industrial area of this revolution and the main catalysts for change will be artificial intelligence and Big Data.
And when I say Big Data, I mean very Big. As digital capabilities evolve, more and more data is produced and stored online. The amount of digital data available is growing at an astonishing rate, doubling every two years. In 2013 it comprised 4,4 zettabytes: by 2020 the digital universe will reach 44 zettabytes or 44 trillion gigabytes. The world of Big Data is so vast that we will need artificial intelligence (AI) to be able to keep track of it.
Artificial intelligence… real
We haven't reached the state of “real” AI yet, but AI is ready to sneak into our lives without much announcement or fanfare. It's already in our cars, Google searches, Amazon suggestions, and many other devices. Crab Apple, Cortana from Microsoft, Google Google and services Echo Amazon uses natural language processing to do useful things. Guy? Search for a restaurant, get directions, store a meeting or eavesdrop on our conversations (in this case it is useful to them).
But there is already more.
A 19-year-old British programmer launched a bot last September, DoNotPay, which is successfully helping people appeal their fines. It is an “AI lawyer” who can decide what to do with the parking ticket received based on a few questions. As of June, the bot has successfully appealed 160.000 of 250.000 parking tickets in both London and New York, with a success rate of 64%.
Imagine this efficiency in healthcare!
Combined with Big Data, AI in healthcare and medicine could better organize patient pathways or treatment plans, and provide doctors with all the information they need to make a good decision.
I'm not talking about a distant future.
Certainly sophisticated learning and AI algorithms will find a place in healthcare in the next few years – I don't know if it's two years or ten, but it's coming.
Andy Schuetz, senior scientist at Sutter Health.
There are already several great examples of AI in healthcare showing potential implications and possible future uses that spark optimism. Of course, it will only be a real revolution if these technologies are available to everyone, and not just the richest or most expert. Anyway, let's take a look at the future of AI in healthcare.
Deepmind Healt, medical records at supersonic speed
The most obvious application of AI in healthcare is data management. Collect them, store them, track them: it is the first step to revolutionizing existing healthcare systems. Recently, the AI search arm of the search giant, Google, launched its project Google Deepmind Health, which is used to extract medical record data and deliver better and faster healthcare services. The project is in its infancy, but promises havoc.
Design of treatment plans
IBM Watson launched its own special program for oncologists which is capable of providing doctors with extremely refined treatment options. Watson for Oncology has an advanced ability to analyze the meaning and context of structured and unstructured data in clinical notes and reports that can be critical to selecting a treatment path. By combining attributes from the patient record with clinical experience, external research and data, the program identifies potential treatment plans for a patient.
Assistance in repetitive tasks
Also IBM launched another algorithm called Medical Sieve. This is an ambitious long-term exploratory project to build the next generation of “cognitive assistants”. A range of AI with analytical and reasoning capabilities and a broad range of clinical knowledge. Medical Sieve is qualified to assist in clinical decision making in radiology and cardiology. The “health” assistant is able to analyze Big Data such as radiological images to identify and detect problems more quickly and reliably. Radiologists in the future may only have to examine the more complicated cases where human supervision is useful.
Big data and deep learning diagnostics
The medical start-up Enlitic it also aims to pair deep learning with vast archives of Big Data to improve diagnostics and patients' lives. Until recently, computer diagnostic programs were written using predefined sets of hypotheses about specific disease characteristics. A specialized program had to be designed for each part of the body and only a limited set of diseases could be identified. Programs often oversimplify reality, resulting in poor diagnostic performance. The advent of Big Data will allow enormous precision. Deep learning can handle a broad spectrum of diseases across the body and all imaging modalities (X-rays, CT scans, etc.).”
Hybrid consultation modes (live and online)
You have a headache, you feel lightheaded and you are certain that you have a fever. You should hear from a doctor. Call, talk to a secretary, ask for an appointment in two days. This is what won't happen with new medical care apps. Also in Italy they are being born, but I am talking about a reality already established as Babel. The English app offers online medical advice and subscription healthcare service. From this year it offers medical advice with AI based on the patient's personal medical history.
Users report symptoms of their illness to the app, which compares them to a big data disease database using voice recognition. After taking into consideration the patient's history and circumstances, Babylon offers an appropriate course of action. The app also reminds patients to take their medications and follows them along to find out how they feel. With such solutions, the efficiency of patient diagnosis can increase, and the waiting time in doctors' offices can be reduced.
Virtual nurses
We welcome the world's first virtual nurse. Molly was developed by the medical start-up Sensely. She has a smiling and amiable face combined with a pleasant voice and her sole focus is to help people monitor their conditions and treatment. The interface uses machine learning to support patients with chronic conditions between doctor visits. It provides proven, personalized monitoring and follow-up care, with a strong focus on chronic diseases.
Medication monitoring and treatment protocols: AiCure
There is also a specific solution to monitor whether patients are actually taking their medications. The app AiCure, supported by National Institutes of Health English, uses a smartphone's webcam and AI to autonomously confirm that patients are adhering to their prescriptions. This is very useful for people with serious medical conditions, for those who tend to go against medical advice, and for participants in clinical trials.
AI, Big Data and Precision medicine
AI and Big Data will also have a huge impact on genetics and genomics. Deep Genomics aims to identify patterns in Big Data of genetic information and medical records, looking for mutations and links to disease. A new generation of computational technologies is emerging that can tell doctors what will happen inside a cell when DNA is altered by genetic variation, both natural and therapeutic. What will we call them?
Craig Venter, meanwhile (one of the fathers of the Human Genome Project), works on algorithms that could design a patient's physical characteristics based on their DNA. With his latest feat, Human longevity, offers its (mostly wealthy) patients full genome sequencing combined with a full body scan and a very detailed medical checkup. The entire process allows you to detect cancer or vascular diseases in their early stages.
Big Data for the Creating drugs
Developing pharmaceutical products through clinical trials sometimes takes more than a decade and costs billions of euros. Speeding up this process and making it more convenient would have a huge effect on healthcare. Atomwise uses supercomputers that pull therapies from a Big Data database of molecular structures. Last year Atomwise launched a virtual search for existing, safe medicines that could be redesigned to treat Ebola. This analysis, which would normally have taken months or years, was completed in less than a day. Being able to fight off deadly viruses months or years faster could be a great shot in the arm against the next pandemic. Or this one, why not.
Another great example of using Big Data for patient management is Berg Health, a biopharmaceutical company that mines data to discover why some people survive diseases, and then improves current treatments or creates new therapies. They combine AI with biological big data from patients to map the differences between healthy and disease-friendly environments and aid in the discovery and development of drugs, diagnostics and healthcare applications.
Analysis of a health system:
97% of healthcare bills in the Netherlands are digital and contain treatment, doctor and hospital data. These invoices could be easily recovered. A local company, Zorgprisma Publiek analyze invoices and use IBM Watson in the cloud to extract data. They can tell if a doctor, clinic, or hospital repeatedly makes mistakes in treating a certain type of condition in order to help them improve and avoid unnecessary hospitalizations of patients.
Things to do for AI and Big Data to really help us: MANY
First of all, we must break down prejudices and fears about AI, and help the population understand how we can limit the risks that the use of AI entails. The greatest fear of all is that artificial intelligence will get out of hand, controlling (or fighting) its “creators”. Stephen Hawking put artificial intelligence among the threats to the survival of the human race.
I don't think the situation is that grim, but I agree with those who highlight the need to adequately prepare for the use of AI in healthcare. We need a few things to avoid problems:
- Creation of applicable and mandatory ethical standards for the entire healthcare sector.
- Gradual development of AI to have time to prevent any disadvantages.
- Medical professionals: training on how AI works in medical contexts.
- Patients: accustomed to AI and increasingly less distrust.
- For those who develop AI solutions: more communication to the general public on the potential benefits and risks.
- For health authorities: measure the success and effectiveness of the system.
If we are successful, Big Data and AI will bring us enormous medical and therapeutic breakthroughs not from time to time, but on a daily basis.
Bianca Stan – Graduated in Law, writer with several books published in Romania and journalist for the group "Anticipatia" (Bucharest). It focuses on the impact of exponential technologies, military robotics and their intersection with global trends, urbanization and long-term geopolitics. He lives in Naples.
Comments are closed.