Artificial Intelligence For Healthcare

Artificial Intelligence (AI), the technology that has captivated many fields, is being hailed as a tool that will help everyone have access to quality medical care, including diagnosis, personalized medical care, prevention of diseases, and the development and improvement of new treatments. Within the next five years, the use and demand of Artificial Intelligence (AI) in medicine is expected to increase tenfold (Perry, 2016).

Artificial Intelligence (AI) can be defined as the use of computer software routines (algorithms) coded with specific instructions to perform tasks for which the human brain is normally considered necessary. Such software can help people understand and process language, recognize sounds, identify objects, and use learning patterns to solve problems. Machine learning (ML) is a way of continually refining an algorithm. The refinement process involves the use of large amounts of data and is done automatically, allowing algorithms to be changed with the aim of improving the accuracy of artificial intelligence (Zandi, 2019). Simply put, AI enables computers to create intelligent behavioural models with minimal human intervention, and has been shown to outperform humans in specific tasks. For example, in 2017, it was reported that deep neural networks (a branch of AI) were successfully used to analyse skin cancer images with greater accuracy than dermatologists and to diagnose diabetic retinopathy (DR) from retinal images (The Lancet, 2017).

However, the definition of AI is evolving. Alongside the more technical definition above, AI is also regarded as something resembling human intelligence, aspiring to surpass the capabilities of any individual technology. It is conceived as a technology interaction that gives a machine the ability to complete a task in a way that makes a human ‘feel’ like it would. The ability of a machine to perform any task that can be achieved by a human has been termed as Artificial General Intelligence (AGI). AGI systems are designed with the human brain as a reference. However, AGI has not yet been achieved; experts have recently predicted its emergence by 2060 (Joshi, 2019).

The case for examining the potential opportunities and risks of applying AI systems for healthcare purposes has taken on new importance given the outbreak of the COVID-19 virus, which has plunged the world into a public health crisis of unprecedented proportions since the beginning of 2020. AI systems can help overburdened healthcare administrations plan and rationalise resources, and predict new COVID-19 hotspots and transmission trends, as well as providing a vital tool in the search for drug treatments or vaccines. However, as governments around the world struggle to adopt technological solutions (many of which rely on ML systems) to help contain and mitigate the crisis, questions about the ethics and governance of AI are rising with equal urgency. There are already growing concerns about how the current crisis will expand government surveillance capabilities, as well as increase the power and influence of so-called ‘big tech’ companies. These concerns are particularly acute for developing countries like India. In such countries, weak public health infrastructure is increasing the appeal of AI-based solutions, while the standards and regulatory frameworks needed to drive the AI trajectory remain weak and underdeveloped.

This scholarly article focuses on some of the key opportunities and challenges in the use of artificial intelligence in healthcare and is further enhanced by a case study of implementing AI in healthcare in India by addressing the key areas of utility, challenges, and risks while using it.

What Is AI?

Artificial intelligence for healthcare includes ML, natural language processing (NLP), speech recognition (text-to-speech and speech-to-text), image recognition, and machine vision, expert systems (a computer system that simulates the decision-making ability of a human expert), robotics, and systems for planning, scheduling, and optimization.

ML is a core component of AI that provides the system with the ability to learn and improve automatically without being explicitly programmed. In fact, there cannot be AI without ML. Computer programs access data and use it for learning purposes without human intervention or assistance, and adjust actions accordingly (Expert Systems, 2017). Deep learning (DL), a type of ML, is inspired by the human brain, and uses multilayered neural networks to find complex patterns and relationships in large datasets that traditional ML may miss (Health Nucleus, undated).

NLP is a part of artificial intelligence that enables computers to read, interpret and manage the language that is part of human life. It is an interdisciplinary field that includes computer science, linguistics, information engineering and computational linguistics respectively in the interest of discussing certain issues on human communication and computer understanding. From SAS. It is the part of software that can listen to spoken words and phrases and convert them into the possible language of computers and make humans capable of making. Automatic voice recognition (AVR) and voice-to-text are its popular and synonymous names.

The Promise Of AI For Healthcare

The World Economic Forum has proposed four ways in which AI could make healthcare more efficient and affordable: enabling tailored treatment plans that will improve patient outcomes, and therefore reduce costs associated with complications arising from treatment; allowing for better and earlier diagnosis that reduces human error; enabling accelerated drug development; and empowering patients to take a more active role in managing their health (World Economic Forum, 2018). One of the main attractions of AI is the potential savings it can bring to the healthcare sector. According to a study by Accenture, when combined, key clinical AI applications could lead to annual savings of $150 billion for the US healthcare economy by 2026. AI can help reduce preventable and correctable system inefficiencies (such as over-treatment, inappropriate care delivery or, in fact, care delivery failures), ensuring a significantly more streamlined and cost-effective healthcare ecosystem (Accenture, 2017).

Another benefit of the application of AI in healthcare settings will be that healthcare workers will be relieved of hours of mundane data work. They will thus be able to focus more on patient care by leaving the task of examining and analyzing clinical data to technology. For example, this will allow healthcare practitioners to assess patients with greater accuracy, which in turn will translate into faster and more accurate diagnoses. AI can provide a diagnosis that would take a doctor (or a team of doctors) several hours to reach. It can also process large amounts of medical images and scans in a much shorter period of time than it would take a human specialist. In this regard, AI is already revolutionizing the field of radiology by improving workflow, diagnostic, and imaging support.

Similarly, the use of AI for administrative purposes would free up resources that could be used to provide care, produce new drugs and treatments, and conduct research to eliminate diseases. Doctors, nurses, and other health care professionals would be relieved of tedious jobs that contribute to burnout, reducing human errors in the practice of medicine (Ash, Petro, & Rab, 2019). For example, NLP was used to analyze unstructured clinical notes, prepare reports, and transcribe conversations with patients. Robotic process automation (actually, involving computer programs hosted on servers rather than robots) was applied to repetitive tasks such as prior authorizations, updating patient records, or billing on some health insurance plans (Davenport & Kalakota, 2019).

At least for high-income countries, one of the many AI applications that has generated significant interest is robot-assisted surgery. ‘Cognitive robotics’ can integrate information from pre-operative medical records with real-time operating metrics to physically guide and enhance the accuracy of the physician’s instrument. The technology incorporates data from actual surgical experiences to inform new, improved techniques and insights. The results of robotics include a 21 percent reduction in length of stay (Accenture, 2017). The value of robotic solutions will grow even more as their development and use increase the variety of surgeries performed.

According to Accenture, the benefits from AI continue to grow from automated operations, precision surgery, and preventive interventions (thanks to predictive diagnosis), and they are expected to radically reshape the healthcare landscape within a decade.

One can assume that this time will not only be used but also create new opportunities in low- and middle-income countries where, with a lack of adequately trained healthcare personnel and the necessary equipment, very expensive and extremely high-tech AI applications, including robot-assisted surgery, have become more accessible over time. But AI tools can also optimize available resources to overcome workforce resource constraints and improve the delivery of healthcare and the outcomes it produces in ways previously unimaginable (USAID, 2019). For some, the greatest near-term value of AI in LMICs is squeezing more value out of available data through ML. Some of the expected applications of AI for health in LMICs will be improving access to and quality of healthcare. Such programs focus on monitoring and evaluating population health and targeting public health interventions for better impact; enabling frontline healthcare workers, including community health workers, to better serve their patients through AI-powered tools such as mobile phone apps; help develop virtual ‘health assistants’ that can coach patients on how to manage their conditions or advise them when to seek care; and build tools that will help doctors diagnose and treat their patients.

Through the use of data science, a multidisciplinary field that uses scientific methods, processes, algorithms and systems to extract knowledge and insights from structured and unstructured data, AI is being used in a variety of medical fields including wellness and lifestyle management, diagnostics, wearable devices and virtual assistants; it is also being used to predict, model and slow the spread of disease in epidemic situations including resource-poor settings for disease surveillance (ibid.).

A recent example is an ML tool used to identify weather and land-use patterns associated with the transmission of dengue fever in the Philippines, a mosquito-borne disease that has spread rapidly around the world in recent years (Wahl et al., 2018). The AI machine built by AIME (US-based company Artificial Intelligence in Medical Epidemiology) is able to predict the occurrence of dengue with increasing accuracy. AIME’s technology has been deployed in Rio de Janeiro, Singapore, the Dominican Republic and two states in Malaysia. The platform provides users with three months’ advance notice of the exact geographic location and date of the next dengue outbreak. Its customized analytics platform also understands its users’ public health data and provides time charts, historical mapping of diseases and ‘rumor reports’ from social media (World Wide Web Foundation, 2017).

At the policy level, these new sciences can offer the potential for substantial time and efficiency savings in aiding health policy decision-making, better integration of healthcare with other sectors, and conducting research and driving quality improvement initiatives (Colclough et al., 2018). As healthcare systems face new challenges from an increasingly aging population suffering from multiple medical conditions, processing datasets associated with these cases using AI promises to be invaluable at this time.

AI’s ability to quickly process and analyze datasets is one of its strongest features. As more countries improve the use of health informatics and electronic medical records (EMRs), AI will become increasingly useful. In India, 30-60 percent of the population has declared that they want their health data to be shared to improve care delivery, allow research to be conducted, and inform health planning. For example, the Open EMR platform in Kenya helped to scale up HIV/AIDS treatment as well as child and maternal child healthcare in rural areas with a more comprehensive collection of data. A cloud-based EMR system was used in western Kenya in 2013. Findings from a study showed that adopting the system led to a 42.9 percent improvement in data completeness (screening for hypertension, tuberculosis, malaria, HIV/AIDS status, or ART status of HIV-positive women) (Haskew et al., 2015). The use of NLP technologies enables machines to identify key words and phrases and thus determine the meaning of text.

For example, to make clinical documentation less cumbersome and even support voice-to-text dictation, NLP algorithms are applied. These technologies are increasingly in demand by health care providers challenged by electronic health record (EHR)1 overload, as they allow them to interact with patients and produce accurate records of consultations without typing at the same time. Both Google and Amazon are exploring how to turn Google Home and Alexa, their popular ambient home computing devices, into innovative healthcare ‘assistants’, respectively. For example, in May 2018, it was reported that Amazon is set to use Alexa to manage chronic diseases and home care (Health IT Analytics, undated). NLP is also being used to drive cancer treatment in low-resource areas, including Thailand, China and India, where AI mines medical literature and patient records – doctors’ notes and lab results among them – to provide treatment advice (Wahl et al., 2018).

With the increasing sophistication of healthcare-focused IT tools such as NLP, the ability to use such tools to help improve the continuum of care should improve over time.

Along with AI and other complementary technologies, it could also help overcome some of the barriers within the healthcare system. High penetration of mobile phones, rapid growth in cloud computing, heavy investments in digitizing health information, and advancements in mHealth applications are offering an expanding platform for AI to enhance individual and population health.

Read Also: