How is Artificial Intelligence (AI) used in Medicine currently?

The use of Artificial Intelligence in medicine has steadily increased in the past decade. With the US Food and Drug Administration (FDA) and the UK’s Medicines and Healthcare products regulatory commission (MHRA) both allowing for the use of AI in the healthcare sector, we have seen many innovations that use machine learning models to analyse, diagnose and treat patients. 

Contemporary medicine only uses AI in very specific clinical settings, specifically those that diagnose diseases based on medical imaging. As a result, wide-spread AI use in medicine occurs most frequently in the Radiology field, where detection of cancers are often based on interpretations of medical images. Many other fields, albeit some slower than others, are starting to adopt AI programmes to further aid in diagnosis. 

A compilation of specific uses of AI in medicine has been listed below, with sources for additional reading. 

“The early detection of atrial fibrillation was one of the first application [sic] of AI in medicine. AliveCor received FDA approval in 2014 for their mobile application Kardia allowing for a smartphone-based ECG monitoring and detection of atrial fibrillation. The recent REHEARSE-AF study showed that remote ECG monitoring with Kardia in ambulatory patients is more likely to identify atrial fibrillation than routine care. Apple also obtained FDA approval for their Apple Watch 4 that allows for easy acquirement of ECG and detection of atrial fibrillation that can be shared with the practitioner of choice through a smartphone.” – [1].

“In the fall of 2018, researchers at Seoul National University Hospital and College of Medicine developed an AI algorithm called DLAD (Deep Learning based Automatic Detection) to analyze chest radiographs and detect abnormal cell growth, such as potential cancers (Figure 2). The algorithm’s performance was compared to multiple physician’s detection abilities on the same images and outperformed 17 of 18 doctors.” – [2]

“The second of these algorithms comes from researchers at Google AI Healthcare, also in the fall of 2018, who created a learning algorithm, LYNA (Lymph Node Assistant), that analyzed histology slides stained tissue samples) to identify metastatic breast cancer tumors from lymph node biopsies. This isn’t the first application of AI to attempt histology analysis, but interestingly this algorithm could identify suspicious regions undistinguishable to the human eye in the biopsy samples given. LYNA was tested on two datasets and was shown to accurately classify a sample as cancerous or noncancerous correctly 99% of the time. Furthermore, when given to doctors to use in conjunction with their typical analysis of stained tissue samples, LYNA halved the average slide review time.” – [2]

“Deep learning has also made progress in gastroenterology, especially in terms of improving colonoscopy, a key procedure used to detect colorectal cancer. Deep learning has been used to automatically predict whether colonic lesions are malignant, with performance comparable to skilled endoscopists. Additionally, because polyps and other possible signs of disease are frequently missed during the exam, AI systems have been developed to assist endoscopists. Such systems have been shown to improve endoscopists’ ability to detect irregularities, potentially improving sensitivity and making colonoscopy a more reliable tool for diagnosis.” – [6]

A 2019 paper from the Radboud University Medical Centre found that their AI model for breast cancer detection was able to significantly outperform radiologists in identifying breast cancer with a 60% higher characteristic in diagnosis. This paper has been cited over 60 times.  – [7]

In the field of Dermatology, AI was shown to outperform dermatologists in identification of melanoma and various carcinomas. These studies were not done in a clinical setting and are not fully indicative of total effectiveness, but show promising results for further use of AI in dermatology – [8]

Glossary of Artificial Intelligence Terms

Algorithm: An algorithm is a specific procedure for solving a specific problem. 

Artificial Intelligence: Artificial Intelligence (AI) is the science and engineering of creating intelligent machines able to perform tasks successfully. AI is often measured against a metric of human intelligence. 

Training: Training is the use of data to improve an AI algorithm

Types of AI: Artificial Narrow Intelligence (ANI), Artificial General Intelligence (AGI) and  Artificial Super Intelligence (ASI) are 3 categories of AI. ANI is information-processing systems built for a specific task that is meant to excel in one goal. AGI seeks to make machines learn, understand and act in a way indistinguishable from humans in multiple goals. ASI is a hypothetical future concept where AI far exceeds the capacity of human intelligence. ANI is considered weak AI, while AGI and ASI are considered strong AI.  

Machine Learning: Machine Learning (ML) is a subset of AI where algorithms are able to predict an outcome after undergoing training on a sample set of data. ML is often mistakenly used when referring to AI and vice versa.

Supervised, Unsupervised and Reinforcement: There are 3 primary ML approaches. Supervised Learning is when the training model contains both the inputs and desired outputs, allowing for the algorithm to find a relationship and test its validity. There is human effort required in this process to label the data. Unsupervised Learning only provides the inputs, and the algorithm attempts to find commonalities in the dataset. This cuts down on human labour needed at this stage. Reinforcement Learning uses a trial and error system where the ML algorithm operates in an environment. Undesirable products are discarded and desirable outcomes are ‘reinforced’, so the algorithm eventually finds the most high-reward process to solving a problem. 

Classification & Regression: Classification involves categorising the data-point into a class. Regression involves assigning a numerical value to the data-point. 

Neural Network: Artificial Neural Networks are computer systems that are inspired by the way the brain processes information. They are composed of layers of points, called nodes, that resemble neurons in the method of passing information. Layers are activated when data is observed.

Deep-Learning: Deep-Learning (DL) is a subset of ML, which uses multiple neural networks to make predictions. It is able to use relatively unstructured data to generate an outcome unlike in general ML.  

Computer Vision: Computer Vision is an AI field that addresses how systems respond to visual stimuli, whether active or through processing digital images. It is meant to mimic how humans see and process that visual information. 

Receiver Operating Characteristic: A Receiver Operating Characteristic (ROC) is a plot of false-positives and true-positives. The area under this graph (AUC) is used as a measure of accuracy of diagnosis by physicians.


  1. Artificial Intelligence in Medicine: Today and Tomorrow –,in%20different%20areas%20of%20medicine.
  2. Artificial Intelligence in Medicine: Applications, implications, and limitations –
  3. AI revolution in medicine, risks and benefits, –
  4. IBM What is artificial intelligence in medicine? –, conflict of interest, they’re marketing their own stuff
  5. The medic portal AI In Medicine –
  6. AI in health and medicine –
  7. Stand-Alone Artificial Intelligence for Breast Cancer Detection in Mammography: Comparison With 101 Radiologists –
  8. High-performance medicine: the convergence of human and artificial intelligence – *(This article is very good!)
  9. Artificial intelligence in biomedical engineering and informatics: An introduction and review –
  10. Deep learning-enabled medical computer vision –