IN FOCUS 

AI IN THE

HEALTH SECTOR

 

 

IN FOCUS 

AI IN THE

HEALTH SECTOR

A landscape rich with possibilities
A landscape rich with possibilities
red_line
There is no better sector than healthcare for intelligent technologies to be implemented, supported and developed. Healthcare has an already established culture of research and development, it also has a seemingly endless scope of applications and possibilities, but more importantly, it has the greatest potential to provide a positive impact on millions of lives of both patients and care workers.

The main opportunity of artificial intelligence (AI) and machine learning (ML) in healthcare is to capture, store, and utilise rich patient data in an automated and appropriate manner so to create an intelligent system that can provide a holistic digital record of a patients health journey and ultimately suggest, predict and inform to provide better patient outcomes.

Since John McCarthy coined the term AI in 1956 as "the science and engineering of making intelligent machines", global companies and startups alike have been exploring this field of computer science discovering a plethora of opportunities along the way for its practical application.  We are now moving with lightning pace into this age where we are seeing long formalised ideas and concepts all of a sudden becoming a very tangible and imminent reality.

 Given the sustained increase in global populations, the pressures on overworked clinicians and the tight purse strings held by those responsible for the running of many health infrastructures, the landscape could be ripe for the development of AI and all that it promises. But how does it get there? Is healthcare ready to invest in this possible future where automation, augmentation, and assistance revolutionise healthcare?

A PwC study found that 63% of healthcare executives are investing in AI today, and 74% already say they plan to do so in the near future.

PwC Report

A PwC study found that 63% of healthcare executives are investing in AI today, and 74% already say they plan to do so in the near future.

Quote details

PROJECTED GLOBAL HEALTH SPEND ON AI

PROJECTED GLOBAL HEALTH SPEND ON AI

red_line
bar_chart

OPPORTUNITIES & GOALS

red_line
cost
patient outcomes
cure finding
safeguard
predict
workload
accuracy
data capture

EMERGING TECHNOLOGY PLATFORMS

EMERGING TECHNOLOGY PLATFORMS

red_line
robot
NLP
visual
Virtual reality
Augmented
Chatbots
speech
Biometrics

The NHS could save £12.5bn a year by using AI for tasks such as communicating medical notes, booking appointments and processing prescriptions.

Former Health Minister Lord Darzi

The NHS could save £12.5bn a year by using AI for tasks such as communicating medical notes, booking appointments and processing prescriptions.

Former Health Minister Lord Darzi

Global research and development
Global research and development
red_line
With the continued spend and growth within this area of health, it’s no wonder that research has been embraced by many sectors of the computer science industry. Below is a list of just a handfull of organisations and how they are responding to this new age of tech driven opportunities.
microsoft

CURING CANCER

Scientists, engineers and programmers at Microsoft’s research labs around the world are using AI and ML within computer science to solve one of the most complex and deadly challenges humans face: Cancer. They are doing so with algorithms and computers instead of test tubes and beakers, trying to change the way research is done on a daily basis in biology.

Nuffield_logo

UNDERSTANDING AI

The Nuffield Foundation has established the Ada Lovelace Institute, with a mission to ensure data and AI work for people and society. It will guide the development and deployment of these technologies and will undertake research and long-term thinking to lay the foundations for a data-driven society with well-being at its core.

Facebook_logo

AI PREVENTING SUICIDE

In 2017 Facebook began to use machine learning to expand its ability to get timely help to people in need. This tool uses signals to identify posts from people who might be at risk, such as phrases in posts. The biggest challenges faced was that so many phrases that might indicate suicidal intent — “kill,” “die,” “goodbye” — are commonly used in other contexts.

Using ML along with human Community Operations reviewers, they created an algorithm which along with other factors, examined linguistic nuances and also comments left on the post. For instance, comments like, “Tell me where you are” or “Has anyone heard from him/her?” from an immediate family member were determined serious cases of people in imminent harm.

BUPA_logo

DIGITAL DOCTORS

Bupa, in partnership with Babylon Health now offers a range of AI-powered health services to all corporate businesses. This includes access to doctors within minutes, directly from any phone or computer.

On an international scale, HealthTap, the world’s first Health Operating System powered by the largest network of interactive doctors and Artificial Intelligence. HealthTap’s proprietary Health Operating System HOPES™ and its AI-powered apps will enable better experiences and enhance speed, convenience, and quality of care for Bupa customers.

Novartis_logo

DEEP DATA MINING

Novartis is planning to analyse huge data sets using artificial intelligence and machine learning to spot disease insights that have been undetectable to scientists thus far to identify early predictors of patient responses to treatments.

AI SALES REP

The Novartis virtual assistant helps salespeople plan better, move better and ensure that they are talking about the things that the healthcare professional is absolutely interested in and relevant. In the long run, the virtual assistant will help the company become more efficient in the emerging markets of telemedicine and improved distribution systems.

GSK_logo

AI-DRIVEN DRUG DESIGN

GSK has partnered with Cloud Pharmaceuticals who aim to use technology to accelerate the drug discovery and design process by using a proprietary AI-driven process to deliver novel molecules tailored to a drug target’s characteristics.

Don Van Dyke, COO of Cloud, said: “It is estimated that the traditional discovery process to arrive at a clinical candidate molecule takes greater than five years. Cloud has consistently been able to reduce that to a matter of a few months.”

MIT_logo

MIT CLINICAL MACHINE LEARNING GROUP

The group has two clearly defined research goals. The first ‘Clinical’ is set as ‘To truly make a difference in health care, we need to create algorithms that are useful for solving real clinical problems.’ and the second, ‘Machine learning’ is ‘We need rigorous solutions, which can pave the way for safe deployment of machine learning in high-stakes settings like healthcare.’

Deepmind_logo

DEEP MIND FOR GOOGLE

Google believes that its technology will be able to develop entirely new diagnostic procedures in the near future. The AI division of the company works with hospitals on mobile tools and research to help get patients from test to treatment as quickly and accurately as possible.

Apps like Streams, currently in use at the Royal Free London NHS Foundation Trust and Imperial College Healthcare NHS Trust, use mobile technology to review test results and receive alerts when a patient deteriorates. It is already having positive effects - nurses at the Royal Free have said that it is saving them over two hours each day, meaning they can spend more time with those in need.

It’s a profoundly hopeful moment in time.
We can reinvent things, and I think it's within our power to change the world for the better.

IBM’s Ginni Rometty

It’s a profoundly hopeful moment in time.
We can reinvent things, and I think it's within our power to change the world for the better.

IBM’s Ginni Rometty

Key areas for the application of AI
Key areas for the application of AI
red_line
It’s said that healthcare may be one of the biggest benefactors of AI and ML not only due to the automation of many manual tasks, but also the insights that better data analysis will reveal over a patients life-time.

ELECTRONIC HEALTH RECORDS (EHR)

An EHR is a collection of patient health data and information in a digital format that can be shared across different health care areas. For the first time using AI and ML algorithms, unstructured data can now be analysed, recorded and stored on the EHR. This unstructured data could be an X-ray image, PDF lab report, Email or SMS communication or clinical notes which could contain a wealth of patient data but have historically not be recorded in a standardised way.

NATURAL LANGUAGE PROCESSING (NLP)

NLP tools can identify meaning in text and voice structures, extract meaning behind the narrative, and in some cases responded with coherent responses. NLP is also used in speech recognition software to allow providers to dictate clinical notes, transcribe phone calls or find data in emails that can be turned into text documents which can then be stored in the patients Electronic Health Record

A Danish AI software company tested its NPL program by having a computer eavesdrop while human dispatchers took emergency calls. The algorithm analysed what a person says, the tone of voice and background noise and detected cardiac arrests with a 93% success rate compared to 73% for humans

robot_550px_x_550px

AI ASSISTED ROBOTIC SURGERY

No longer Science Fiction, the use of Robotics within surgery is fast becoming one of the leading fields in Healthcare. Robot-assisted surgery is considered "minimally invasive" which means less healing time for patients. In clinical tests, robots have analysed data from pre-op medical records and assisted surgeon's instruments during surgery with such efficiency it leads to a 21% reduction in a patient's hospital stay. One study even showed that the AI-assisted robotic procedure resulted in five times fewer complications compared to surgeons operating alone.

CURE DISCOVERY

Machine learning and other technologies are expected to make the hunt for new pharmaceuticals quicker, cheaper and more effective. On a long list of researchers, Pfizer is using IBM Watson, a system that uses machine learning, to power its search for immune-oncology drugs. UK start-up Exscientia’s artificial intelligence (AI) platform continues to hunt for metabolic-disease therapies, and GNS Healthcare the US is helping to drive the company’s search for cancer treatments.

IMAGING TEST RESULTS

IMAGING TEST RESULTS

Using pattern recognition techniques and machine learning, it’s possible to extract untapped information in files such as X-rays, photographs and brain scans to give better insight into a patient’s health status including emerging conditions that might escape a human-focused on diagnosing an unrelated condition.

DIGITAL THERAPEUTICS

Digital therapeutics often employ strategies rooted in cognitive behavioural therapy its treatment relies on behavioural changes usually spurred by a collection of digital impetuses. Because of the digital nature of the methodology, data can be collected and analysed as both a progress report and a preventative measure.

Treatments are being developed for the prevention and management of a wide variety of diseases and conditions, including type II diabetes, congestive heart failure, obesity, Alzheimer's disease, dementia, asthma, substance abuse, ADHD, hypertension, anxiety, depression, and several others.

PREDICTING RISK & ICU TRANSFERS

AI models can be used to find patients who are at high risk of cardiac arrest. The machine learning models use patient medical records, laboratory results, and vital signs from patients to find early warning signals of a deteriorating condition.

VIRTUAL NURSING ASSISTANTS

Since virtual nurses are available 24/7, they can answer questions, monitor patients, provide quick answers and alert clinicians when needed. From interacting with patients, providing check-ups and medication reminders to directing patients to the most effective care, virtual nursing assistants could save the healthcare sector millions of pounds in a very near future.

nurse

AI PROVIDES A HELPING HAND TO CLINICIANS

AI-based decision support and diagnosis can help clinicians make better decisions by incorporating more data into the decision-making process and by learning patterns that are outside the clinicians’ scope.

AI-based decision support helps doctors and nurses by providing a second opinion or by pointing out information they may have missed, often via mobile or tablet devices.

CLINICAL TRIALS

Better candidates for clinical trails means better results. Machine learning has several potential applications which may help direct clinical trial research. Better sampling techniques would lead to smaller, quicker, and less expensive trials.

BEHAVIOURAL HEALTH INTERVENTIONS

From an app that can recognise gestures and patterns when for when someone is smoking (or tempted to) to an algorithm that can analyse social media content to predict those at risk of suicidal tendencies. By creating mood diaries, word associations, text analysis of input data from patients, AI can continually asses in real time and clinicians can be alerted immediately if needed.

REDUCE READMISSIONS

25% of patients with serious and chronic illnesses discharged from the hospital will be readmitted within 30 days to be treated again, often with less favourable outcomes. AI models can use data about the specific patient’s recent care, their current condition, treatment, their home life and other risk factors from electronic medical records. AI can provide the reasons that will lead to readmission and also provide recommendations for the types of treatments.

CHANGING LIVES WITH
FUTURE TECHNOLOGIES


CHANGING LIVES WITH FUTURE TECHNOLOGIES


Overview of current applications
Overview of current applications
red_line
There are many examples of AI and ML in the market place already, with many more to be released imminently. Below are just a few that augment new technology and traditional care practices to improve patient outcomes.
skin_vision

SKIN CANCER DETECTION APP

SkinVision helps you check your skin for signs of skin cancer with instant results on your phone. Clinically-proven technology, combined with the knowledge of dermatologists specialised in skin cancer, helps you keep your skin healthy. After your registration, use your phone to take photos of your skin spot. Within 30 seconds you receive a risk indicator of your skin spot right on your phone. If you receive a high-risk indication senior dermatologists will review your picture and give you a Doctor’s Advice.

akili

IPAD GAME  FOR ADHD SUFFERES

Akili has created a proprietary technology platform for neurological treatments engineered to directly improve cognitive deficits. It represents an entirely new category of medicine. Digital therapeutics. Second, by second monitoring of patient progress enables continuous assessment and as well as competing with existing drugs, it may also appeal to parents who are reluctant to medicate their children for ADHD.

babylon
BABYLON

DIAGNOSIS CHAT BOT AND GP PORTAL

Babylon Healths mission is to put an accessible and affordable health service in the hands of every person on earth. The AI-driven app currently provides video consultations with doctors and access to an AI-doctor where patients can input their symptoms to the AI-doctor to check them against a database including family health history, recent illness, and predictability and risk of certain conditions.

care angel

VIRTUAL NURSE SERVICE

The World's First AI and Voice-Powered, Virtual Nurse Assistant. It aims to provide a proactive and continuous care delivery at home and extended care outside of the clinical setting through patient monitoring and data analysis. It works by sending data-driven alerts and interactions in real-time and is supported by teams of carers and doctors who are alerted to any circumstances that may need intervention.

smoke beat

SMOKING HABIT TRACKER

An ml-based application which recognises hand-to-mouth gestures’ in order to help people better understand their behaviour and make life-affirming changes, specifically in smoking cessation.

The app tracks the users key times of smoking, where they normally smoke, how many and offers actionable insights into this behaviour. It then delivers combinations of personalised CBT (Cognitive Behavior Therapy) incentives that bear real impact on negative smoking patterns, and increase smoking cessation treatment adherence and effectiveness.

apple

ECG AND FALL DETECTION

An ECG in the Apple watch can screen your heart rhythm for irregularities, such as atrial fibrillation, and alert you about any problems and enables you to share the data with your doctor.

 It also has the ability to detect a fall and deliver an alert to let you initiate an emergency call. If it senses that you’re immobile for more than a minute, the watch will start the emergency call automatically and contact the emergency contact saved in your phone.

apple_watch
bio base

WELLBEING APP BASED COURSE

WELLBEING APP BASED COURSE

Designed for companies to deliver to their employees, BioBase is an app based course that focuses on personal stress and wellbeing. It monitors stress over time by asking questions, tracking responses, recording sleep and also health data via wearables. Through this, a user can visualise when and where their stress levels are rising by comparing locations, activities and interactions. The app offers support and solutions to its users based on data recorded and aims to increase wellbeing, health, mental focus and reduce stress levels.

reset

PRESCRIPTION ONLY CRAVING TRACKER

This prescription-only app helps people with alcohol, methamphetamine, cocaine, and cannabis dependencies keep track of their moods and cravings, and provides cognitive behavioural therapy as well as offering training on topics like identifying situations that are likely to lead to a relapse, or recognising negative thinking and shifting to a more positive outlook.

Using ML to analyse the patients' data input, it can respond instantly to risks and be alerted to learned behaviour leading to negative outcomes. The patients' doctor monitors the patient’s use of the app and is updated by the system as and when deemed appropriate.

blue_ice

BEHAVIOURAL AND MOOD TRACKER

BlueIce is a prescribed evidence-based app to help young people manage their emotions and to reduce urges to self-harm. Through AI the app offers a personalised set of activities designed to reduce distress including a music library, photo library, physical activities, relaxation and mindfulness exercises. It also spots and challenges negative thoughts and has automatic routing to emergency numbers if urges to harm continue.

AI OPPORTUNITIES WITHIN A TYPICAL DIABETICS JOURNEY

AI OPPORTUNITIES WITHIN A TYPICAL DIABETICS JOURNEY

red_line

Nimbletank have created a typical diabetic patients experience map highlighting the role AI and ML could (and does) play in improving the treatment and behaviour patterns commonly associated with this disease.

Experience_map

A year spent in artificial intelligence is enough to make one believe in God.

Alan Jay Perlis, computer scientist and professor

A year spent in artificial intelligence is enough to make one believe in God.

Alan Jay Perlis, computer scientist and professor

LOOKING FORWARD TO THE FUTURE

LOOKING FORWARD TO THE FUTURE

white_line
What's next and how it affects us all.
What's next and how it affects us all.
red_line
Although healthcare professionals acknowledge that AI is not a cure-all solution to its problems, it is obviously an area rich in possibilities that will only continue to gain momentum in the coming years. The quick adoption of AI within healthcare and the increasing amount of promising case studies being published it looks like the road to recovery for the healthcare system has well and truly begun. So what will change look like?

As with any industrial or technological revolution, employment requirements will diminish in certain areas where automation can achieve better results faster, but in its place, new roles will be generated for a more augmented digital workplace. Accenture Research recently concluded a study of future workforce trends in health care and predicted that from 2018 to 2022, employment in health care would increase by 15% through growth and improved technology.

With better performance comes better patient outcomes. It is predicted that through data analysis and continued ML, the technology could significantly reduce the length of patient stay within hospitals and also reduce readmissions by allowing clinicians to predict which patients are most likely to be readmitted and how they might be able to reduce that risk.

But even before a patient becomes a patient there is an opportunity to reduce the risk of disease. Machine learning coupled with wearables and implanted data chips can provide automatic feedback to help monitor general health. This can then help identify those with undiagnosed diseases, predict the likelihood of future disease and present prevention interventions.

In the treatment phase, there is also new evidence that nanobots can now actually deliver cancer drugs to only those cells that are infected. Unlike current chemotherapy treatments which kill cancer cells as well as significantly damaging healthy cells in the body, micro nanobots in the bloodstream can intelligently detect which cells need to be treated. The idea of armies of minuscule robots patrolling our bodies is now one which is being explored for a very real future.

Wherever artificial intelligence takes us in the coming years, it’s safe to say that through research, the ethical use of data and shared patient knowledge this evolving field of computer science will have a tremendous effect not only on global health but on the role of healthcare and all those involved within it.

SUGGESTED ARTICLES

red_line

For more information or to discuss your own Intelligence Audit please contact Paul Vallois, managing director.
paul.vallois@nimbletank.com
020 3828 6440
© 2019 Nimbletank. All Rights Reserved.