Examples of AI in healthcare

ELANA SÜGIS: AI applications have already become a common part of our everyday lives. Lots of fields, including healthcare are experiencing transformation due to the application of AI-based solutions. In this video, you will learn about some of the solutions that are already in practice. Although there are many of them, we will focus on the solutions that enhance diagnostics, support clinical decision making, and improve patient care. Let’s start with the systems that help to enhance diagnostics. Computer-aided detection and computer-aided diagnosis systems are built to help clinicians in making diagnoses. For example, in the field of breast imaging, AI applications are designed to improve radiologists ability to find and assess breast abnormalities using magnetic resonance imaging.

The software, called QuantX detects abnormalities in these MRI images, highlights areas of interest, and provides a metric that is correlated with the likelihood of malignancy. Similar to the previous case. ProFound AI is able to analyse mammograms for lesions suspicious of cancer from digital tumour synthesis scans. The algorithm detects malignant soft tissue densities and classifications in the scans. Scores are assigned to each detection and represent how confident the algorithm is that the detection is malignant. Clinical study indicates an 8% improvement in the detection rate when the software is used. In April 2018, the US Food and Drug Administration approved the first AI-based diagnostic tool called IDx-DR This tool detects diabetic retinopathy in people with diabetes by analysing retinal images.

It is the first device authorised that provides a screening decision without the need for a clinician to also interpret the image or results. The software for 4D Medical rapidly and automatically analyses x-rays and applies algorithms to identify and quantify any functional lung impairment. It generates a ventilation report and sends it to the hospital to enable clinicians to determine the course of actions for the treatment. Such AI based systems can be applied for a wide range of pulmonary diseases, such as COVID 19, asthma, chronic obstructive pulmonary disease, cystic fibrosis, and lung cancer. Next to diagnosing, AI can also help healthcare professionals to improve clinical decision making.

Doctors often have to work with massive amounts of clinical textual data, such as epicrises patient medical history, clinical notes, and many more. These are so-called free form or unstructured texts that contain acronyms and abbreviations and may also involve spelling and typing errors. AI-based systems can be used to work with this data more efficiently and save time for healthcare professionals. For example, AI solutions can help to categorise clinical nodes, summarise the text, find patient cohorts for clinical trials, visualise the data for the review, and support administrative tasks. Finally, AI has major applications in the improvement of patient care. Nowadays, most patient health data, such as their medical history, prescribed medications, laboratory test results and procedures are stored in digital form.

These digital records are known as electronic health records. AI systems trained on large amount of such electronic records, can be used for accurate predictions of medical events, such as in-hospital mortality, 30-day unplanned readmission, length of stay in the hospital, and discharge diagnosis. Patient experience can also be improved with the use of AI-robot assisted surgery. For example, MicroSure’s robot helps surgeons to overcome physical limitations in lymphatic system. procedures. The motion stabiliser system improves performance and precision during surgical procedures. Another example is Mazor Robotics 3D tools. These tools help to visualise surgeons surgical plans, read images with the eye that recognises anatomical features, and perform a more stable and precise spinal operation.

Finally, the availability of massive data sets and advanced algorithms, has driven more interest in major improvements in the use of artificial intelligence in the field of drug discovery and development. These are only a few examples of how AI systems are currently used in healthcare Do you think these AI based systems are useful here?