Life with Artificial Intelligence: Basic Sciences, Medical Education, and Medical Treatment

Ritu Sharma

Department of Physiology, Army College of Medical Sciences, New Delhi, India

Corresponding Author: Ritu Sharma, Department of Physiology, Army College of Medical Sciences, New Delhi, India, Phone: +91 9717438306, e-mail: sharmaritu990@gmail.com

How to cite this article: Sharma R. Life with Artificial Intelligence: Basic Sciences, Medical Education, and Medical Treatment. J Med Acad 2024;7(1):1-2.

Source of support: Nil

Conflict of interest: None

Artificial intelligence (AI), in its broadest sense, is intelligence exhibited by machines, particularly computer systems. In today’s world, where the virtual world has taken over reality and people are more concerned about their digital image, AI is involved in our day-to-day life from social media recommendations to news feed; everything is customized for a person. It is so powerful that we end up buying unnecessary things online, start investing time on social media, and get diverted from the actual task. We have lesser attention span and maybe we are less patient in any given situation.

So, AI has captured and made us dependent in every single aspect of life from finance to social, from physical to psychological, from academics to research. Every field is using this machine intelligence so is the medical field.

The application of AI in the medical field has grown significantly, evidenced by increased research and development in recent years. People are discussing that AI doctors will eventually replace human physicians in the future. We believe that human intelligence will not be replaced by machine doctors in the near future, but AI can and is definitely playing a key role in assisting healthcare personnel and doctors in making correct clinical decision for the best of the patient. Since AI works on collecting large clinical data, analyzing it and correcting it through continuous feedback, the increasing availability of healthcare data and big data analytic methods has made possible the recent successful applications of AI in healthcare. An AI system can assist healthcare professionals by keeping them up-to-date with recent medical information from journals, textbooks, and clinical practices for proper patient care. In addition, an AI system can help to reduce diagnostic and therapeutic errors that are inevitable in human clinical practice. Using multiple relevant clinical questions, powerful AI techniques can analyze clinically relevant information through large stored data and can help or even guide the physician in making correct clinical decision. Not only this, but AI is keeping and analyzing useful information from a large patient population to assist in making real-time inferences for health risk alert and health outcome prediction.¹

Use of AI is not just limited to analyzing symptoms and assisting in correct clinical diagnosis or research; it is beyond that. With advancements of AI, the medical education has changed so much and has become more learner centric. It is now easier and more effective to showcase an AI-based clinical scenario to make medical students understand. It is really important to incorporate AI-based learning in curriculum for the present digital generation of budding doctors and future patients. The students who are mostly engaged in handling gadgets are being trained to serve human patients. Here, the use of machine-based learning by use of simulators, gaming devices, online peer interactions and then analysis of data may help in better learning and understanding.

Artificial intelligence devices in healthcare mainly fall into two major categories:

Machine learning (ML) techniques: Analyze structured data such as imaging, genetic, and electrophysiological (EP) data. It helps in assessing the disease outcome in a given patient.²
Natural language processing (NLP) methods: AI technologies help in extracting information from unstructured data such as clinical history, examination notes, and medical journals. It supports the structured medical data. NLP procedures extract the unstructured data and try to convert it into machine-readable structured data for better analysis.³

Thus, AI is now playing an integral part in keeping electronic medical records (EMRs). This is convenient, easily accessible, and authentic.

Further, depending on whether to incorporate the outcomes, ML algorithms can be divided into two major categories: unsupervised learning and supervised learning. Unsupervised learning is useful for feature extraction. Clustering and principal component analyses (PCA) are two major unsupervised learning methods. Clustering is grouping subjects having similar features without using outcome information. It is used to sort patients having maximal and minimal similarities within the same cluster and between the clusters. PCA is mainly for reduction of size, especially when the trait is recorded in a large number of dimensions, such as in genome studies. PCA projects the data onto a few principal component (PC) directions, without losing too much information about the subjects. One can use PCA first to reduce the data dimension and then use clustering.

Supervised learning considers the subjects’ outcomes together with their traits. It is seen that supervised learning is more commonly used in healthcare. Various supervised learning techniques are used for medical applications such as support vector machine (SVM), neural network, linear regression, logistic regression, naïve Bayes, decision tree, nearest neighbor, random forest, discriminant analysis, and many others. Of these, SVM and neural network are most popular in medical field. SVM is being used in early diagnosis of many neurological and psychiatric disorders, for example, Alzheimer’s disease, certain cancers.

Natural language processing: The unstructured data which is unreadable for ML such as medical history and clinical examination notes. NLP has two main components:

Text processing: The NLP identifies a series of disease-relevant keywords in the clinical notes based on the historical databases.⁴
Classification: Then classified into normal and abnormal cases based on these keywords.

Now these keywords are entered with structured data and analyzed for proper diagnosis and treatment plan.

The cloud-based CC-Cruiser⁵ is one prototype to connect an AI system with the front-end data input and the back-end clinical actions (Flowchart 1).

Flowchart 1: Functioning of prototype AI system in healthcare showing inputs, outputs, and feedback

Thus, we can see that AI can be used for early diagnosis, planning treatment, better outcome assessment, and keeping medical records.

With this changing scenario in healthcare system, it is the need of the hour to keep medical students, practicing physicians, and medical teachers up to date with use of AI. AI can be used in medical education in the following ways: virtual inquiry system, medical distance learning and management, and recording teaching videos in medical schools. Many AI-based systems are being used these days both for teaching and for actual diagnosis. Human Diagnosis Project, or “Human Dx,” commonly known as is being used for better, accurate, inexpensive, and accessible care for everyone by merging the collective intellect of physicians with ML. To eliminate drug delivery error, systems such as MedEye (confirms the accuracy of medication by comparing it to the hospital information or prescription and using image recognition and ML to verify) MedPass™ (it is personalized medicine dispensing technology that verifies every pill and prevents error) are being used, which help in correct drug delivery and faster picking up of errors if they happen.

DxR Clinician (a Virtual Inquiry System) is a virtual patient system for teaching hospitals, medical institutions, and residents. The software compiles hundreds of genuine patient data into individual cases, which are then studied with the help of AI. These data deal with variety of clinical problems and help students to analyze, examine, and diagnose the case in a simulated environment for better understanding. AI robots are successfully doing surgeries and are being developed to perform complex surgeries without the use of humans someday.⁶

Despite being very useful, the use of AI has got some issues such as ethics, infrastructure for learning and development of proper feedback and assessment system for quality check. There should be proper regulatory body, and policies should be made to use AI in healthcare in an ethical, unbiased, and more transparent manner. Physicians and medical teachers should be trained first before they start using it to prevent any unwanted incidents. Because one wrong decision can cost a life.

Artificial Intelligence should be used cautiously and intelligently in healthcare system.

REFERENCES

1. Neill DB. Using artificial intelligence to improve hospital inpatient care. IEEE Intell Syst 2013;28:92–95. DOI: 10.1109/MIS.2013.51

2. Darcy AM, Louie AK, Roberts LW. Machine learning and the profession of medicine. JAMA 2016;315:551–552. DOI: 10.1001/jama.2015.18421

3. Murff HJ, FitzHenry F, Matheny ME, et al. Automated identification of postoperative complications within an electronic medical record using natural language processing. JAMA 2011;306:848–855. DOI: 10.1001/jama.2011.1204

4. Afzal N, Sohn S, Abram S, et al. Mining peripheral arterial disease cases from narrative clinical notes using natural language processing. J Vasc Surg 2017;65:1753–1761. DOI: 10.1016/j.jvs.2016.11.031

5. Long E, Lin H, Liu Z, et al. An artificial intelligence platform for the multihospital collaborative management of congenital cataracts. Nat Biomed Eng 2017;1:0024. DOI: 10.1038/s41551-016-0024

6. Mir MM, Mir GM, Raina NT, et al. Application of artificial intelligence in medical education: current scenario and future perspectives. J Adv Med Educ Prof 2023;11(3):133–140. DOI: 10.30476/JAMP.2023.98655.1803

________________________
© The Author(s). 2024 Open Access. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by-nc/4.0/), which permits unrestricted use, distribution, and non-commercial reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.