Editorial Feature

Accelerating Medical Diagnostics with AI

Artificial intelligence (AI) has become a tool used in many industries globally to increase productivity and efficiency. Historically known for being the driver and early adopter of new technology, the healthcare industry has significantly benefitted from the advancements made in the field of machine learning and AI.

Accelerating Medical Diagnostics with AI

Image Credit: Gorodenkoff/Shutterstock.com

One of the most significant benefits is disease detection, diagnosis and progression identification. Traditionally, these tasks would need to be completed by a trained medical professional, using their years of experience; however, given the ability of machine learning to find correlations in large sets of data, it is now possible to complete these tasks just as accurately as their experienced human counterpart. As a result, machine learning has provided a method to allow disease diagnoses to be accelerated, allowing disease to be identified before typical symptoms show.

Many terminal illnesses, such as cancers, dementia and autoimmune diseases, are usually diagnosed after severe symptoms are shown. This, therefore, means that the diagnoses come late within the disease's progression, delaying treatment and reducing the likelihood of a successful recovery. Research has shown a  positive correlation between an early diagnosis and the rate of survival of most diseases, meaning that it has always been the goal of the medical industry to improve disease identification.

Advancements in machine learning over the last decade have provided a solution to this problem by employing statistical analysis to collect data to find correlations associated with the disease. With the wealth of machine learning models available, it is now possible to diagnose conditions using a variety of patient data.

Image-Based Diagnosis

The medical industry has seen great advancements in the imagining techniques used to visualize the internal conditions of the body. Common imaging tools include PET scans, X-ray images, Ultrasound and MRI images.

Each of these provides a different type of image, requiring years of training to identify various disease types and how they present themselves.

With the advent of computer vision technology, clinicians are now able to use algorithms designed to identify and diagnose specific conditions within a patient. This would be done by designing a convolutional neural network (CNN) to complete a classification task, where the aim is to determine with some probability how likely the image shown is one of a set number of diseases.

Given the large number of scans that have been taken over the years, training networks in this way can be very simple to do, while having the effect of being able to detect patterns that would not have been overtly visible to the patient's doctor. As a result of this type of technology, AI is now used in several hospitals as an experimental technology to assist clinicians in the early diagnosis of disease. AI has helped detect various cancers while also helping characterize their genetic origin.

Sound Based Diagnosis

While many diseases present themselves with internal discomfort or behavioral changes, researchers have begun to investigate how the sound of a person's voice can be used to detect an underlying condition.

Academics across the globe, most notably those at MIT and Cambridge, have developed techniques that can listen to the sound of a person, voice or cough to determine whether the person suffers from a viral condition. Most recently, this type of technology was used as a diagnostic method for the SARS-CoV-2 (COVID-19) virus, allowing individuals to be diagnosed without having to do a test or leave their homes.

Another example of this would be in the prediction of Alzheimer's disease. It is now possible for machine learning models to listen to a patient's speech and determine whether a person is likely to suffer from Alzheimer's disease in the future. From this, we can see that the use of AI in this sector provides novel ways to diagnose disease, which are both non-invasive and accurate and time-efficient.

Wearable Technology Integration

Wearable technology has become a much larger part of modern life, with items like smart watches and earphones a popular choice. There has been an increase in the number of body monitoring devices available on the market, which allow us to monitor our internal state at any given moment.

Emotiv's new MN8 earphones claim to provide insights into the functioning of our brain, monitoring our focus and attention, whilst the NOWATCH wearable watch functions as a cortisol monitoring device, alerting the user of their stress levels even before the peak of their stressful activity.

Both of these devices function due to machine learning and are but a few of the many available devices on the market. This new era of body monitoring wearable technology hints at the future of medical diagnosis;  visits to the doctor's office may be replaced by a device and algorithm capable of diagnosing its wearer. At the very least, these devices will provide a wealth of timestamped and personalized data that a medical professional can use to assist with diagnoses in the future.

Interview: Advancing Cardiovascular Care Through the Use of AI

References and Further Readings

Shaheen, M.Y., (2021) AI in Healthcare: medical and socio-economic benefits and challenges. ScienceOpen Preprints. Available at: https://www.scienceopen.com/hosted-document?doi=10.14293/S2199-1006.1.SOR-.PPRQNI1.v1

Mansour, R.F., El Amraoui, A., Nouaouri, I., Díaz, V.G., Gupta, D. and Kumar, S., (2021) Artificial Intelligence and Internet of Things Enabled Disease Diagnosis Model for Smart Healthcare Systems. IEEE Access9, pp.45137-45146. Available at: https://ieeexplore.ieee.org/document/9380633

Shaheen, M.Y., (2021) Applications of Artificial Intelligence (AI) in healthcare: A review. ScienceOpen Preprints. Available at: https://www.scienceopen.com/hosted-document?doi=10.14293/S2199-1006.1.SOR-.PPVRY8K.v1

Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.

Written by

Robert Clarke

Robert is a PhD student studying Artificial Intelligence at the University of Bath. He really enjoys learning and reading about current technology advancements, particularly in areas such as robotics, machine learning and neuroscience. In his spare time, Robert uses artificial intelligence to create artwork and model robots on computers.

Citations

Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    Clarke, Robert. (2022, January 04). Accelerating Medical Diagnostics with AI. AZoRobotics. Retrieved on January 17, 2022 from https://www.azorobotics.com/Article.aspx?ArticleID=449.

  • MLA

    Clarke, Robert. "Accelerating Medical Diagnostics with AI". AZoRobotics. 17 January 2022. <https://www.azorobotics.com/Article.aspx?ArticleID=449>.

  • Chicago

    Clarke, Robert. "Accelerating Medical Diagnostics with AI". AZoRobotics. https://www.azorobotics.com/Article.aspx?ArticleID=449. (accessed January 17, 2022).

  • Harvard

    Clarke, Robert. 2022. Accelerating Medical Diagnostics with AI. AZoRobotics, viewed 17 January 2022, https://www.azorobotics.com/Article.aspx?ArticleID=449.

Tell Us What You Think

Do you have a review, update or anything you would like to add to this article?

Leave your feedback
Submit