TECH – X – Neurons to Wires: The Impact of artificial intelligence in healthcare dynamics

Rupali Sachdev, Medical Intern

Grant Medical College and Sir JJ Group of Hospitals, Mumbai

Before we begin a discussion about artificial intelligence let’s have a conversation with the most popular artificial intelligence technology available at our fingertips – our voice assistants.

“Hey Siri, explain artificial intelligence”

“Artificial intelligence means the theory and development of computer systems able to perform tasks normally requiring human intelligence, such as visual perception, speech recognition, decision-making, and translation between languages.”

Well, that’s a rather technical definition, as expected from a machine. Let’s simplify it. Artificial intelligence or AI is the ability of a machine to behave like a human. It signifies the ability of a machine to use the comprehension abilities of an individual to perform simple or complex activities. Each decision that a human being makes is a series of choices that occur at multiple dimensions starting from the synapses running across the neurons in our brain. Artificial intelligence is a form of similarly imitating these human neuronal circuits in order to give machines the ability to independently make decisions.

When we expand this concept to healthcare, it opens a whole new playing field thereby simultaneously leading to more opportunities and dilemmas. The most common application of artificial intelligence is seen in the area of imaging and diagnostics. (1) AI software has the ability to quickly analyze X-ray, CT, and MRI imaging to identify the pathology. It works on the principle of machine learning. Convolutional neural networks (CNNs) are a class of artificial neural networks that is most commonly used to interpret medical imaging. For example, a machine learning software after being fed enough images of kidney stones can start identifying kidney stones in scans itself without any human aid. Despite the overwhelmingly positive response to the technology, warnings have been published about the potential dangers of AI. Concerns have been expressed stating that future medicine based on AI will render radiologists irrelevant. (2)

Coming to diagnostics, Artificial intelligence technology diagnoses the result at a micro level, which can help in avoiding errors in pathological diagnosis. In oncology, for instance, a group at Stanford University has done an experiment with AI algorithms pitched against 21 dermatologists on the algorithm’s ability to identify skin cancers. In some cases, AI has been able to outperform clinicians because AI systems can learn more from successive cases and can be exposed to multiple cases within minutes, which far outnumber the cases a clinician could evaluate in one lifetime. AI-based decision-making approaches bring used in situations where experts often disagree, such as identifying pulmonary tuberculosis on chest radiographs. (1)

Artificial intelligence technology is used as a powerful data mining tool in various fields of drug design, such as virtual screening, activity scoring, quantitative structure-activity relationship (QSAR) analysis, de novo drug design, and in silico evaluation of absorption, distribution, metabolism, excretion, and toxicity (ADME/T) properties. One real-world example of this technology is Cyclica, a Canadian startup, which uses computational algorithms to evaluate and predict how drugs might interact with the human body. Such kind of testing helps pharmaceutical companies identify side effects before clinical trials and ease the process.

Predictive models for personalized treatments with engineered stem cells, immune cells, and regenerated tissues in adults and children are being developed, based on AI technology. These intelligent machines could dissect the whole genome and isolate the immune particularities of an individual patient’s disease in a matter of minutes and create the treatment that is customized to the patient’s genetic specificity and immune system capability.

Machine learning technology can also help in efficiently managing healthcare resources. The ability to identify, analyze and interpret large data sets is essential when public health planning needs to be taken into consideration. The Netherlands uses AI for their healthcare system analysis – detecting mistakes in treatment, workflow inefficiencies to avoid unnecessary hospitalizations. (4)

The advances seen in the field of AI aren’t just limited to clinicians. DXplain, a decision support system developed at the Laboratory of Computer Science at the Massachusetts General Hospital, has the characteristics of both an electronic medical textbook and a medical reference system. The current DXplain knowledge base (KB) includes over 2600 diseases and over 5700 clinical findings (symptoms, signs, epidemiologic data, and laboratory, endoscopic and radiologic findings). DXplain suggests possible diagnoses and provides brief descriptions of every disease in the database. Technological advances in computational power, graphics, display systems, tracking, interface technology, haptic devices, authoring software, and artificial intelligence (AI) have supported the creation of low-cost, user-friendly virtual reality (VR) technology and virtual patients (VPs). (5)

On the other hand, this development in the field of artificial intelligence and machine learning has raised relevant psychological, ethical, and medico-legal issues which need to be addressed for AI to be considered fully capable of patient management. There are concerns that AI-mediated decisions may be hard to validate and challenge, or may result in rogue decisions. Other issues which are brought to light include informed consent, lack of emotional intelligence, bias, patient privacy, and liability have been raised. The simple question remains – if anything were to go wrong, who is to be blamed? The doctor? The machine? The algorithm behind the machine? The engineer who developed the software? Or the administration which brought the software into the hospital? While it is the responsibility of AI researchers to ensure that the future impact is more positive than negative, ethicists and philosophers need to be deeply involved in the development of such technologies from the beginning. (6)

In conclusion, If AI is to augment and complement rather than replace human judgment and expertise in biomedicine it is important to train the new generation of medical trainees regarding the concepts and applicability of AI and how to function efficiently in a workspace alongside machines for better productivity along with cultivating soft skills like empathy in them. Our goal should be to strike a delicate mutually beneficial balance between effective use of automation and AI and the human strengths and judgment of trained clinicians. (3)

References:

1. (N.d.). Einfochips.Com. Retrieved November 9, 2021, from https://www.einfochips.com/blog/how-artificial-intelligence-is-transforming-the-healthcare-sector/
2. Pesapane, F., Tantrige, P., Patella, F., Biondetti, P., Nicosia, L., Ianniello, A., Rossi, U. G., Carrafiello, G., & Ierardi, A. M. (2020). Myths and facts about artificial intelligence: why machine- and deep-learning will not replace interventional radiologists. Medical Oncology (Northwood, London, England), 37(5), 40.
3. Amisha, Malik, P., Pathania, M., & Rathaur, V. K. (2019). Overview of artificial intelligence in medicine. Journal of Family Medicine and Primary Care, 8(7), 2328–2331.
4. Sniecinski, I., & Seghatchian, J. (2018). Artificial intelligence: A joint narrative on potential use in pediatric stem and immune cell therapies and regenerative medicine. Transfusion and Apheresis Science: Official Journal of the World Apheresis Association: Official Journal of the European Society for Haemapheresis, 57(3), 422–424.
5. The Laboratory of Computer Science. (n.d.). Mghlcs.Org. Retrieved November 9, 2021, from http://www.mghlcs.org/projects/dxplain
6. Keskinbora, K. H. (2019). Medical ethics considerations on artificial intelligence. Journal of Clinical Neuroscience: Official Journal of the Neurosurgical Society of Australasia, 64, 277–282.