AI shouldn’t demean worth of welfare & humanity, but must turn challenges into opportunities! (A brief analysis)

Artificial intelligence is rapidly advancing and swiftly exploring areas of science, health and other sectors. But, to the best of our knowledge and understanding, it is very important to know that every technology comes up with new threats and challenges- ‘challenge to adhere with ethics, transparency, and to work with integrity and compliance’.

In order to bring a chain of transparency and to minimize the indiscernible risk within the healthcare system over the use and misuse of advanced technologies alike AI, AR (Augmented Reality), VR (Virtual Reality), XR (Extended Reality), Machine Learning (ML), Deep Learning (DL) and so on, the World Health Organization released its first global report, a living document on Artificial Intelligence with six guiding principles that are appropriately measured for ethical usage of AI in health care sector and clinical trials to treat mild to severe diseases and to come up with faster and better solutions during any global public health crisis.

The report ‘Ethics and Governance on Artificial Intelligence’ is the result of two years of rigorous consultation held by a panel of 20 experts appointed by the World Health Organization, who came up with solutions for resolving challenges over its use and misuse in health sector. The body of the report is purely based on theoretical and analytical aspects and need to have a wider consultation with policy makers, business chiefs and industrialists.

While providing a valuable insight over the usage of AI in health care system, the Chief Scientist at WHO, Dr. Soumya Swaminathan said that Artificial Intelligence is an evolving field and has to act and work on a more improved and improvised set of regulations and protocols. At present, this technology is catering only to the highly developed nations where the issue of accessibility of internet, accessibility of technology is not a problem. The world has to find a more evolved model that is easily accessible and cost-effective in terms of its use for low and middle-income nations like Africa. One size doesn’t fit all!

She believed that it was paramount to organize the big data available with the private companies as they held datasets in maximum proportion over other emerging digital health start-ups/companies and thus the private companies should be held responsible over traversing the possible challenges of extracting, organizing, categorizing and sharing those data with global agencies. Big data helps in providing impeccable information to streamline customer service processes that supports in creating the best practices to work with consumers and patients, as they get access to a more comprehensive and personalized experience.

As per market research, big data in the healthcare domain is expected to surge by $ 34.27 billion worldwide by the year 2022 with a compound annual growth rate (CAGR) of 22.06 percent, whereas it is estimated to see an upsurge of 14 percent by the end of 2025, propelled largely through western companies in electronic health records, applying management tools and workforce management solutions. Moreover, keeping in mind the easy access to data these days, the urge for the use of advance technologies is becoming more and more imperative. Medical and healthcare professionals can easily generate more personalized patient treatments with the use of analysed critical data.

Healthcare is a crucial sector in the growth of economy considering its rising significance over the period of time. Recently, this sector drew a lot of attention because of COVID-19 pandemic, particularly the pharmaceutical companies across the world that strove for rat race to develop the best vaccination to curb the impact of the disease under testing protocols.

The year 2020 witnessed a huge investment of tech skills in the field of Healthcare & Technology-based start-ups. Artificial Intelligence & Machine Learning corroborated a new proposition in the overhauling of the entire healthcare system to sustain its durability while providing best of the efforts and in minimizing the challenges faced by the sector at the same time.

AI-based technology has enormous potential in strengthening the delivery of healthcare and medicine and assisting countries across the world to achieve ‘Universal Health Coverage’. This sophisticated tech-tool could be used to improve the public health surveillance and enable healthcare providers to better understand the complexity of the disease and engage in providing better services to their patients.

These AI-augmented expert systems boost the abilities in healthcare sector by providing improved, quick and accurate diagnoses of a disease, optimize treatment plans, invigorate preparation for public health emergencies and response, stimulate in creating informed decisions by the ministers and policy-makers and also extend its support in the allocation of meaningful resources within health system.

Just a couple of decades back, AI was just a dream and was a part of the scripts of Hollywood directors, producers and script writers. Most of the famous science-fiction based movies articulated their story-line on AI based tools or services- Star Wars, Mission Impossible 1, 2 & 3, chapters of Resident Evil and so on. But now, the fiction has turned into reality. Now, AI is a reality.

Artificial Intelligence is nothing but a technique that enables computers to learn and impersonate human intelligence with the use of logical interpretation. According to a report, published by the Accenture in the year 2017, the market value of Artificial Intelligence in health sector is estimated to reach $6.6 billion by the end of the year 2021.

Crisis like COVID-19 has forced us to look at the other side of the picture, which could be sudden, critical, adverse, odd, risky, and could often bring in other severe challenges related to socio-economic conditions. And thus, predictive models of AI could gradually become one of the essential tools to get hands-on experience in the medical and healthcare domains to confront those challenges.

With the rapid proliferation and evolving practices of AI in the field of healthcare and in response to global health crisis like COVID-19 pandemic situations, local and central administration and agencies, academic institutions, foundations, non-governmental organisations and National Ethics committee are trying to build consensus-based regulations to come out with solutions and norms over the use and misuse of such technologies in healthcare sector.

Currently, AI is being carefully observed for its use in radiological diagnostic purposes to check early detection of cancer-like diseases. Most of AI-based guiding tools like thoracic imaging, abdominal and pelvic imaging, colonoscopy, mammography and brain imaging are being applied for radiological treatment. Additionally, its being used for autonomous detection of vision-threatening diabetic retinopathy, ophthalmology-based issues, to identify RNA and DNA based sequencing for guiding immune-based therapies etc.

One of the quantified AI-based technology, Prognostic Scoring Systems such as Sequential Organ Failure Assessment (SOFA) to analyse the severity of ailments and to predict mortality. SOFA is a version of AI and has widely been since many years. It has been majorly used during pandemic in assisting guidance over needed interventions in specified situations, distribution of ration supplies, in identifying criticality of the condition of patients, and so on.

AI-based techniques and applications, for instance- robotics, chatbots, speech recognition, face recognition, artificial creativity, computer vision, virtual reality, augmented reality, extended reality, image processing, natural language generation, smart assistants, disease mapping, social media monitoring, neural networks, self-correction, predicting consumption patterns, machine learning, deep learning are gradually picking up momentum in every defined sector across the world.

AI based application, NLP (Natural Language Processing) helps in decoding the cognitive aspect of human thought-process, making early detections and future predictions related to mental health issues with the use of coding algorithms.

Use of AI-guided technologies and applications- mobile health apps and wearable units like activity trackers, smart watches and smart glasses are expected to increase to $1.5 billion, creating more opportunities to monitor an individual’s health and further capturing data to predict any health risks, as per WHO’s report.

A newly healthcare digital start-up, NIRAMAI (Non-Invasive Risk Assessment with Machine Intelligence) based in Bengaluru, India, has developed a novel breast cancer screening solution (Thermo-imaging) with the use of Artificial Intelligence and Machine Learning. The application provides more accurate information, detects early cancer development at an early stage, no harmful radiation, non-invasive, no pain, works for women of all ages, portable, light and based on machine learning expert systems.

Meanwhile, there are concerns over the use and misuse of algorithms that could lead to situations where decisions could directly be controlled by the machine and human anatomy might get compromised.

As per the World Health Organisation, predictive algorithms based on inadequate or inappropriate data can result in significant racial or ethnic bias, and so, use of high quality and comprehensive datasets is essential. Moreover, digital divide could exacerbate inequitable access to healthcare technologies, if countries do not take appropriate measures. Biased inferences, misleading data analyses and poorly designed health application and tools could be harmful. Ethically optimized tools and applications could sustain widespread use of Artificial Intelligence to improve human health and the quality of life, while mitigating or eliminating many risks and bad practices.

Most of the healthcare-based businesses are trying to incorporate Artificial Intelligence based services into their analytics practice to get a clearer understanding of all the critical issues defined in their unprecedented list, but strangely, most of these efforts are becoming more of concern than success. For instance, predictive model for accessible health services can only work in the presence of expert guidance and this limits the effectiveness over the use of AI in healthcare field. Most of the AI applications are standalone practices that needs expertise to understand and adopt programming and language for the existing work culture and business operations.    

To overcome and address aforementioned challenges and the existing biases related to age, gender, ethnicity race encoded in data to train algorithms in health care system and services, WHO has identified six core ethical principles with the help of its experts -a) Protect human autonomy b) promote human well-being, human safety and the public interest c) ensuring transparency, explainability and intelligibility d) foster responsibility and accountability e) ensure inclusiveness and equity, and f) promote AI that is responsible and sustainable.

Implementing the above-mentioned principles and human rights obligation into practice, requires substantial attention from health and IT-based ministries, policy makers, government institutions, stakeholders, designers, programmers, healthcare providers, and public to incorporate ethical norms at every step while designing, developing and deploying any novel AI-based technology in the field of health sector.

A more life-changing, transformational and improved set of expert services could enhance the efficacy of Artificial Intelligence in Healthcare institutions. Ready-to-choose best predictive models and self-service analytical model-building mechanisms will ease a lot of problems for health across the globe.

#EthicalGuidance #AIpractices #AITools #WHO #Healthcare #GlobalReportWHO #PublicHealthCrisis #ArtificialIntelligence #MachineLearning

Note- a few quotes from WHO's report on Ethical guidance were taken as such

Image courtesy: Creative Commons