By Chaitanya Mukund

AI and ML, this is how the healthcare industry envisions itself to run, in the coming times through the mammoth potential and benefits by using artificial intelligence and machine learning. By its increasing use by leaps and bounds, it is believed that by 2025 AI will be ubiquitous in healthcare. It opens vast possibilities helping people focus on their work with better end results.

InkWood Research estimated the size of Al market in health care industry at about 1.21billion dollars in 2016.
Apart from India few other Asian countries like China, Japan and South Korea are using AI in healthcare.

Faster and better diagnostic capabilities

Use of algorithms and software in complex data analysis without any manual inputs and their various practical applications in diagnosis, treatment, new drug development and improving patient care requires data alignment and analysis with accurate decision making with predictive analytics.

AI tools promise faster service delivery in diagnostic issues and identifying trends, genetic predisposition towards certain diseases. Improving care requires alignment of mammoth health data and appropriate and predictive analytics can support clinical decisions and administrative tasks.

Taking healthcare to the next level

• Low cost and higher success rates for drugs research for use and safety in humans, thus saving millions of dollars that the researchers invest. USFDA approves just 1 in 10 drugs. Use of Al can help save the cost and increase success rate of researches.
• In Radiology, the identification and detection of minutest diversion from the normal image is one of the major benefits of AI.
• AI has been implemented largely in the schedule of the Radiological Society of North America.
• Pattern recognition in detection of cancer and other diseases in early stages.
• AI assisted robotic surgeries
• Tele medicine
• Clinical decision support systems help in more comprehensive approach for disease management.
• Providing Virtual nurses especially for chronic diseases. For example Sense.ly.
• These nursing assistants could save about 20 million dollars annually in the health care industry.
• Wearable health trackers to monitor activity and heart rate etc by companies like Fit Bit and Apple
• System analysis in healthcare services. For example digitalization of healthcare invoices (97%) in Netherlands.

Use of AI or robots during advanced years can help reduce dependence upon hospital visits and care homes.

What seems to be a revolutionary step in health care is the use of robots for social interaction to keep the aging minds sharp.

Handling Mundane Jobs

Tasks like maintaining medical records and keeping a track record of patient’s past history etc are very repetitive in nature in the sense that they are routine jobs which can be time consuming but doctors need to focus more on their specialization-specific areas to get the best results in the treatment of their patients.

So what is solution then???

Here is where AI and ML comes in. As stated in a 2016 report from CB Insights, about 86% of healthcare provider organizations, life science companies, and technology vendors to healthcare are using artificial intelligence technology. By 2020, these organizations will spend an average of $54 million on artificial intelligence projects.

Applications related to artificial intelligence and digital automation are used for compilation as well as the management of such data by healthcare organizations. The collection, storage, tracing and re-formatting of the data concerned, is all facilitated by the machines and bots employed for handling the same.

Talking about claim processing in healthcare, AI can help in enhancing the cost efficiency in terms of health insurance coverage options, and other medical benefits, along with minimizing the amount of money wastage and detecting fraudulent claim cases. AI also assists underwriters in the medical field by facilitating health insurance through use of software by the underwriters to evaluate the coverage of an applicant. According to Eric Horvitz, head of Microsoft Research’s Global Labs, “AI-based applications could improve health outcomes and the quality of life for millions of people in the coming years.” .

In a healthcare centre, routine responsibilities like analysis of tests, X-Rays, CT scans, doing data entry etc can be easy assigned to these bots, who can carry out such jobs much more efficiently, at a much faster speed and with utmost accuracy.

In 2016, Innovation & Digital Health Accelerator (IDHA), a team at Boston Children’s Hospital developed an app for Alexa, a virtual assistant powered by Amazon, adding to its ‘skill’ to provide basic health information and advice for parents of ailing children. It enables Alexa to answer questions regarding common health symptoms among children like fever, cold & cough, headache, rash, vomiting etc,, and helps the parents to decide if the child needs a visit by a doctor or not about medications and whether symptoms require a doctor visit. It also provides information regarding weight- or age-specific dosage guidelines for over-the-counter drugs.

With estimated value of 40 billion dollars to healthcare, robots analyze data from pre-op medical records and also guide the surgeon hence reducing about 21% of hospital stay for the patient.

Major Challenges for Al and ML

• Healthcare industry stakeholders need to be convinced about positive returns on investments. Better understanding of value proposition influences the decision making.
• Al companies need more data and case studies to be more convincing to deal with complexities the stake holders and decision makers present. Alignment of big health data is key to accurate and timely decisions.
• Predictive analytics can support clinical decision making and prioritize administrative tasks.
• Industry conservatism and regulations hinder the adoption of Al.
• Lack of confidence in replacing humans and lack of Al and ML skills.
• Educating the stakeholders regarding positive case studies and achievements can make the transition to Al easier

System development for different domains in healthcare

• Diagnostic SaaS(35%)
• Consumer apps (29%) for example: Intermedica, Your.MD, Ada Health, Babylon.
• Operational efficiencies (24%)
• Medical devices (11%)

Current trends

In 2010, Anti Dote was founded with its head office in London. Through its clinical trial matching program where millions of patients can be connected for medical research studies through more than 180 online communities.

In 2015, Deep 6 AI was founded. This California based company uses Natural Learning Processing (NLP) for matching patients for clinical trials.

Companies like IBM Microsoft, Google, Intel and startups like IDX-DR, Medicine Compassionate Lab have contributed to Al algorithms for healthcare use. Grid cell, place cell and path integration with machine learning are used for navigation of blind people. Watson for Health (IBM) and DeepMind Health (Google) are some examples.

iCarbon X provides its customers with wellness apps.

Through the app called Meum, one can monitor and record health information e.g. Diet, exercise, sleep, emotional state etc. Such platforms can be used by developers to deliver useful recommendations to the customers.

Voxel Cloud uses medical image analysis software.

Artificial Intelligence and Machine Learning technologies have a very bright future in the field healthcare industry. With the help of these innovations, doctors and other healthcare professionals can make a precise and detailed diagnosis of patient’s illness, enabling them to decide the right kind of treatment at the right time with utmost accuracy. These can also help in providing a streamlined workflow for the medical fraternity, thus ensuring effective as well as timely outcomes.