European mhealth hub | Artificial Intelligence (AI)

Artificial Intelligence (AI)

Role this area should play in the mHealth space
Artificial intelligence (AI) becomes more and more prominent in the delivery of healthcare. Mobile health is benefiting immensely from this technology as AI algorithms, sensor technology, and advanced data techniques are transforming the mobile devices into full-fledged health-management platforms. AI is being used to analysze large quantities of patient data, and increase the accuracy of disease detection. AI can enhance disease surveillance and it has the potential to improve the productivity of the healthcare professionals and to augment their knowledge and abilities to make better decisions. It has also great potential to accelerate operational processes of health institutions and reduce costs. Major areas supported by AI are Clinical Decision Support and Information management. The ubiquity of smart phones and mobile devices makes Ai powered medical decision apps accessible worldwide. These apps plays an important role in patient empowerment while enabling in the same time the professionals to make more personalized and effective decisions with positive outcomes for patients. AI enables the mHealth to offer solutions for the following use cases: early prediction of medical conditions such as heart attacks from data collected 24/7 on person’s heart rate, sleep cycle, breathing rate, activity level and blood pressure measuring and monitoring the patient’s heart rate and glucose levels connecting healthcare providers with those accessing healthcare solutions. managing healthcare operations and resources, administrative procedures, appointments and reminders, dashboard and better payroll solutions more accurate diagnoses, identify at-risk populations, and better understand how patients will respond to medicines and treatment protocols. Big data, the technology to aggregate and analyse huge amounts of data, has particular relevance to medical and health data. Healthcare Analytics has the potential in disease prevention by significantly reduce the cost of treatment. Advanced data mining and analysis techniques lead to better health outcomes by enabling healthcare professionals to better analyse a patient’s condition and self-care. AI empowers healthcare professionals to make more informed decisions and to treat patients more effectively.

Role this area should play in the mHealth space

Artificial intelligence (AI) becomes more and more prominent in the delivery of healthcare. Mobile health is benefiting immensely from this technology as AI algorithms, sensor technology, and advanced data techniques are transforming the mobile devices into full-fledged health-management platforms. AI is being used to analysze large quantities of patient data, and increase the accuracy of disease detection. AI can enhance disease surveillance and it has the potential to improve the productivity of the healthcare professionals and to augment their knowledge and abilities to make better decisions. It has also great potential to accelerate operational processes of health institutions and reduce costs. Major areas supported by AI are Clinical Decision Support and Information management. The ubiquity of smart phones and mobile devices makes Ai powered medical decision apps accessible worldwide. These apps plays an important role in patient empowerment while enabling in the same time the professionals to make more personalized and effective decisions with positive outcomes for patients.

AI enables the mHealth to offer solutions for the following use cases:

early prediction of medical conditions such as heart attacks from data collected 24/7 on person’s heart rate, sleep cycle, breathing rate, activity level and blood pressure
measuring and monitoring the patient’s heart rate and glucose levels
connecting healthcare providers with those accessing healthcare solutions.
managing healthcare operations and resources, administrative procedures, appointments and reminders, dashboard and better payroll solutions
more accurate diagnoses, identify at-risk populations, and better understand how patients will respond to medicines and treatment protocols.

Big data, the technology to aggregate and analyse huge amounts of data, has particular relevance to medical and health data. Healthcare Analytics has the potential in disease prevention by significantly reduce the cost of treatment. Advanced data mining and analysis techniques lead to better health outcomes by enabling healthcare professionals to better analyse a patient’s condition and self-care. AI empowers healthcare professionals to make more informed decisions and to treat patients more effectively.

Current challenges and limitations
Identifying high-value use cases to be addressed by AI can be a challenge. Lack of standardization framework of algorithms for health. There is no common, vetted framework for assessment / benchmarking of AI-based solutions based on Open source, open data and open standards. The AI algorithms lack explainability and interpretability due to complexity of neural networks which makes it impossible to describe how they work. Hence black box nature of machine learning. The AI algorithms should be trustworthy with proven robustness to outliers and to adversarial attacks. Lack of comparative data on cost effectiveness, or safety evaluations in clinical settings. Insufficient training data, a major limiting factor for the predictive performance of models on new data previously unseen. Concerns over the privacy and protection of sensitive health data. Disparate ethnic groups or residents of different regions may have unique physiologies and environmental factors that will influence the presentation of disease. Reconfiguration of outdated business processes for adoption of AI and trusting that disruption will yield big results and will lead to better patient outcomes.

Current challenges and limitations

Identifying high-value use cases to be addressed by AI can be a challenge.
Lack of standardization framework of algorithms for health. There is no common, vetted framework for assessment / benchmarking of AI-based solutions based on Open source, open data and open standards.
The AI algorithms lack explainability and interpretability due to complexity of neural networks which makes it impossible to describe how they work. Hence black box nature of machine learning.
The AI algorithms should be trustworthy with proven robustness to outliers and to adversarial attacks.
Lack of comparative data on cost effectiveness, or safety evaluations in clinical settings.
Insufficient training data, a major limiting factor for the predictive performance of models on new data previously unseen.
Concerns over the privacy and protection of sensitive health data.
Disparate ethnic groups or residents of different regions may have unique physiologies and environmental factors that will influence the presentation of disease.
Reconfiguration of outdated business processes for adoption of AI and trusting that disruption will yield big results and will lead to better patient outcomes.

What benefit could this bring to adopters of this innovation?
It is widely accepted that the combination of mobile healthcare apps and artificial intelligence is the future of health care. mHealth can leverage the power of AI and Machine Learning to deliver health services for optimal, personalized and improved patient care. AI could improve availability of healthcare services, increase efficiency in the treatment process, reduce costs and increased opportunities for preventive care. AI-powered digital health assistants integrated in chatbots and delivered through smartphones and other mobile devices can “transform mHealth apps from a simple static source of information into smart platforms for personalized pre-primary care and assisted self-care provision”. Early adopters of AI and digital technologies will better and more successfully manage future health emergencies. Besides diagnosis and personalized medicine AI should be leveraged to optimize and streamline administrative health workers tasks and workflows. ML coupled with wearables are opening a wide array of new possibilities for the mHealth systems. These can be further improved and expanded by incorporating new emerging technologies such as personal sensing. Additional use case that can be of interest to the aging population and increase their appetite for consuming AI powered mHealth services and apps are: detecting fall incidents, seizure episodes, emotional arousal (stress level and mood swings), and nutritional intake.

What benefit could this bring to adopters of this innovation?

It is widely accepted that the combination of mobile healthcare apps and artificial intelligence is the future of health care. mHealth can leverage the power of AI and Machine Learning to deliver health services for optimal, personalized and improved patient care. AI could improve availability of healthcare services, increase efficiency in the treatment process, reduce costs and increased opportunities for preventive care.

AI-powered digital health assistants integrated in chatbots and delivered through smartphones and other mobile devices can “transform mHealth apps from a simple static source of information into smart platforms for personalized pre-primary care and assisted self-care provision”.

Early adopters of AI and digital technologies will better and more successfully manage future health emergencies. Besides diagnosis and personalized medicine AI should be leveraged to optimize and streamline administrative health workers tasks and workflows.

ML coupled with wearables are opening a wide array of new possibilities for the mHealth systems. These can be further improved and expanded by incorporating new emerging technologies such as personal sensing. Additional use case that can be of interest to the aging population and increase their appetite for consuming AI powered mHealth services and apps are: detecting fall incidents, seizure episodes, emotional arousal (stress level and mood swings), and nutritional intake.

Examples

AI for detection of COVID-19 related coughs

Artificial Intelligence (AI)

Approach or solution AI model to detect asymptomatic Covid-19 infections through cellphone-recorded coughs. An algorithm developed by a team of researchers at MIT has correctly identified people with Covid-19 only by the sound of their coughs. It is claimed that the crucial difference in the... ...

AI for early diagnosis of mental health

Artificial Intelligence (AI)

Approach or solution CompanionMx is a mental monitoring system to inform proactive mental health care. The system uses active monitoring of voice and passive monitoring of other smartphone metadata to continuously produce acoustic and behavioural biomarkers that predict core symptoms of mood and anxiety disorders. The... ...

AI powered platforms for management of hospital readmission.

Artificial Intelligence (AI)

Approach or solution Chronic Obstructive Pulmonary Disease (COPD) affects 1.2 millions in UK and is a second common reason for admission to emergency in UK. NHS Greater Glasgow and Clyde, the largest health board in the UK in partnership with KenSci is pioneering an AI... ...

Internet Of Things (IoT) – Open Artificial Pancreas System and closed-loop insulin delivery system

Artificial Intelligence (AI)

Approach or solution Bigfoot Biomedical is committed to leveraging data, people, and smart technology to create a connected ecosystem of solutions that will deliver improved outcomes valued by patients, providers, and payers. Bigfoot’s investigational automated insulin delivery system, otherwise known as an “artificial pancreas,” was... ...

Internet Of Things (IoT) – Diabetic Patient monitoring

Artificial Intelligence (AI)

Approach or solution In general the IOT broad scope is patient monitoring. Continuous Glucose Monitor (CGM) is a device that helps diabetics to continuously monitor their blood glucose levels for several days at a time, by taking readings at regular intervals. Smart CGMs like Eversense Freestyle... ...

AI and Big Data for early detection of sepsis

Artificial Intelligence (AI)

Approach or solution Dignity Health implemented a big data and predictive analytics system uses big data and advanced analytics platform to predict potential sepsis cases at the earliest stages, when intervention is most helpful. Sepsis is an extreme inflammatory response to infection that can lead... ...

AI and Big Data for rare disease

Artificial Intelligence (AI)

Approach or solution HealX is operating in an under-represented market of 7,000 rare diseases impacting over 350 million people all over the world which have no treatments. Organisation or initiative HealX, a British start-up which focused on curing rare diseases URL or reference... ...

AI-powered health assistants (Clinical trials)

Artificial Intelligence (AI)

Approach or solution AiCure uses a proprietary AI platform to directly engage and provide support to patients via smartphones to deliver meaningful, high-quality data around patient behaviour. AiCure improves predictability of study timelines, reduces costs and accelerates timelines through remote patient engagement and assessments, including... ...

AI-powered health assistants (Conversational chatbox)

Artificial Intelligence (AI)

Approach or solution AI-powered health assistants are conversational agents that have their own independent applications. Some mobile apps are using these conversational chatbots to help patients find the causes of their symptoms and integrate these chatbots into popular messenger apps for personalized pre-primary care and... ...

AI for wellbeing and detection of emotional state

Artificial Intelligence (AI)

Approach or solution The WoeBot is an AI-drive app, which provides “quick conversations to feel better,” for individuals feeling isolated or sad. It adapts to symptoms and severity, and deliver the right intervention to the right person at the right time. Organisation or initiative... ...

AI for early detection of violence at large scale

Artificial Intelligence (AI)

Approach or solution The article explores the potential for AI (and associated machine learning and big data) to prevent and address violence at a large scale. mHealth Technology combined with AI can be used for Prevention, Early Detection, and Early Response to Violence, Including Violence... ...

AI for COVID Health Emergencies

Artificial Intelligence (AI)

Approach or solution ITU and WHO established an Ad-hoc Group on “Digital Technologies for COVID Health Emergencies” (AHG-DT4HE) to review the role of AI in combatting COVID-19 throughout an epidemic’s life cycle Organisation or initiative ITU/WHO AHG-DT4HE: Ad-hoc group on digital technologies for COVID health emergency... ...

Standardization Framework for AI Algorithm for Health

Artificial Intelligence (AI)

Approach or solution ITU and WHO established the Focus Group on “Artificial Intelligence for Health” (FG-AI4H) to explore opportunities for international standardization and for the application of AI for health on a global scale. Focus Groups are open membership platforms to progress challenging themes in... ...

What is on the horizon?
Going forward, mHealth data will be a critical component of artificial intelligence tools and of the healthcare industry as a whole. mHealth, through mobile apps and devices will be an important source of big health and wellness data. Availability of these massive datasets could open new perspectives in the development of AI-based health and health care tools. More affordable sensors, rich real-time data, and powerful analysis tools are powering the next generation of mobile apps. mHealth data, extracted from wearable devices and patient monitoring tools, will play a critical role in powering artificial intelligence and analytics technology in the future, according to a recent analysis from Frost & Sullivan. “As mHealth rapidly gains traction, wearables, telehealth, social media, and patient engagement are expected to find adoption among more than half of the population in developed economies by 2025,” said Sowmya Rajagopalan, Advanced Medical Technologies Global Director. “In the future, patient monitoring data will be combined with concurrent streams from numerous other sensors, as almost every life function will be monitored and its data captured and stored,” The industry will also see a surge in wearables and biosensors, largely due to the rise of chronic disease and an industry shift from treatment to prevention. Continuous glucose monitors, pulse oximeters, and electrocardiogram monitors are some of the main tools that will dominate in the next few years. Smart prosthetics and smart implants, which help to measure key parameters to support monitoring and early intervention, will also likely be critical tools in the coming years. “The patient monitoring market is expected to be worth more than $350 billion by 2025, as the focus is likely to move beyond device sales to solutions.” The estimated increase in the global AI economy by 2022 is $3.9Tn from $1.2Tn in 2018. This increase can be attributed to machine learning tools and deep learning techniques. The spending in the healthcare industry alone is estimated to reach $36.1Bn in 2025 with a CAGR of 50.2%. It is predicted that the biggest investors in this technology would be hospitals and physicians as well as individual caregivers. According to WEF these are 3 ways AI will change healthcare by 2030 1.- AI-powered predictive care. “In 2030, the healthcare systems can anticipate when a person is at risk of developing a chronic disease, for example, and suggest preventative measures before they get worse. This development has been so successful that rates of diabetes, congestive heart failure and COPD (chronic obstructive heart disease), which are all strongly influenced by SDOH, are finally on the decline.” 2.- Networked hospitals, connected care to a single digital infrastructure . Locations connected through a supply demand balancing approach. 3.- Better patient and staff experiences, reduce admin burden. By learning from every patient, every diagnosis and every procedure, AI creates experiences that adapt to the professional and the patient. This not only improves health outcomes, but also reduces clinician shortages and burnout, while enabling the system to be financially sustainable.

What is on the horizon?

Going forward, mHealth data will be a critical component of artificial intelligence tools and of the healthcare industry as a whole. mHealth, through mobile apps and devices will be an important source of big health and wellness data. Availability of these massive datasets could open new perspectives in the development of AI-based health and health care tools. More affordable sensors, rich real-time data, and powerful analysis tools are powering the next generation of mobile apps.

mHealth data, extracted from wearable devices and patient monitoring tools, will play a critical role in powering artificial intelligence and analytics technology in the future, according to a recent analysis from Frost & Sullivan.

“As mHealth rapidly gains traction, wearables, telehealth, social media, and patient engagement are expected to find adoption among more than half of the population in developed economies by 2025,” said Sowmya Rajagopalan, Advanced Medical Technologies Global Director.

“In the future, patient monitoring data will be combined with concurrent streams from numerous other sensors, as almost every life function will be monitored and its data captured and stored,”

The industry will also see a surge in wearables and biosensors, largely due to the rise of chronic disease and an industry shift from treatment to prevention. Continuous glucose monitors, pulse oximeters, and electrocardiogram monitors are some of the main tools that will dominate in the next few years. Smart prosthetics and smart implants, which help to measure key parameters to support monitoring and early intervention, will also likely be critical tools in the coming years.

“The patient monitoring market is expected to be worth more than $350 billion by 2025, as the focus is likely to move beyond device sales to solutions.”

The estimated increase in the global AI economy by 2022 is $3.9Tn from $1.2Tn in 2018. This increase can be attributed to machine learning tools and deep learning techniques. The spending in the healthcare industry alone is estimated to reach $36.1Bn in 2025 with a CAGR of 50.2%. It is predicted that the biggest investors in this technology would be hospitals and physicians as well as individual caregivers.

According to WEF these are 3 ways AI will change healthcare by 2030

1.- AI-powered predictive care.

“In 2030, the healthcare systems can anticipate when a person is at risk of developing a chronic disease, for example, and suggest preventative measures before they get worse. This development has been so successful that rates of diabetes, congestive heart failure and COPD (chronic obstructive heart disease), which are all strongly influenced by SDOH, are finally on the decline.”

2.- Networked hospitals, connected care to a single digital infrastructure . Locations connected through a supply demand balancing approach.

3.- Better patient and staff experiences, reduce admin burden. By learning from every patient, every diagnosis and every procedure, AI creates experiences that adapt to the professional and the patient. This not only improves health outcomes, but also reduces clinician shortages and burnout, while enabling the system to be financially sustainable.

Keywords
Artificial Intelligence, Machine Learning