The idea of allowing machines to make decisions traditionally reserved for humans might seem frightening, but the truth is that technologies such as artificial intelligence (AI) have the potential to improve our lives significantly. AI has already made an impact on multiple industries, and the field of medicine is no different. Advances in AI are continuing to drive the adoption of this technology in healthcare, which makes the future even more exciting. For instance, it’s estimated that the global value of AI in healthcare will reach $22.79 billion by 2023, compared to $2.16 billion in 2017.
An Algorithm That Can Detect Skin Cancer
One of AI’s biggest potential benefits is early detection and prevention of deadly diseases. Imagine having an app that can tell you when something’s wrong with your body, even before your patients visit a doctor or feel unwell. That’s exactly what researchers at Stanford University are trying to accomplish, and they’ve already made a huge breakthrough. In early 2017, a group of scientists announced the development of an AI algorithm that detects skin cancer. They first created a system containing 130,000 images of skin abnormalities and diseases. Based on this large data-set, they trained an algorithm to diagnose skin cancer. The results were then compared to diagnoses made by board-certified dermatologists, showing that the algorithm was 91 percent accurate.
This lays a solid basis for further developments in the field. According to the Stanford researchers, their breakthrough could be used to develop an app that would be able to detect skin cancer based on a photo taken via smartphone. Just imagine how many lives it could save! Every year, around 5.4 million people in the US, are diagnosed with skin cancer. Although the 10-year survival rate for skin cancer is 95 percent if caught early enough, not all patients are lucky enough to get an early diagnosis. With Stanford’s innovation, anyone who notices suspicious changes on their skin could use the app and get diagnosed instantly.
An AI System Called Corti Identifies Heart Attacks by Listening
While Stanford’s solution detects abnormalities based on photos, an AI system called Corti accomplishes the same by simply listening. Corti is designed to detect a heart attack during an emergency call and share the information with the dispatcher. By harnessing the power of machine learning, Corti can analyze the caller’s voice and recognize background noise. This can help emergency dispatchers identify heart attacks more efficiently.
A Medical Diagnosis Made by a Chatbot
Today’s healthcare system isn’t cheap. In fact, rising healthcare costs are a primary reason why most people avoid going to the doctor. To make healthcare services affordable and accessible to everyone, a startup called Babylon Health created an AI system capable of identifying different medical conditions. Babylon, which operates as a chat-bot, communicates with users via text. Once they’ve entered their symptoms, Babylon analyzes the data and compares it to known conditions to find the right match and provides relevant medical information. Patients can also use the service to book a video appointment with a doctor, receive drug prescriptions, and book exams at nearby medical facilities. Babylon will also provide you with a health report based on the information you enter.
Previous testing relied on entering various symptoms into the system and comparing the results to those made by human doctors. Babylon showed 98 percent accuracy in identifying common conditions in primary care, while human doctors provided accuracy ranging between 52 percent and 99 percent. Of course, Babylon isn’t a replacement for a human doctor. As the company’s website warns, this solution shouldn’t be used in emergency medical situations, and does not diagnose the user’s health condition or make treatment recommendations.
AI trends are affecting various industries, and healthcare is not exempt from these changes. The wave of innovation driven by AI could lead to the development of cutting-edge solutions capable of saving more patients’ lives. Despite the fact that most of these solutions are still in early development, they take us one step closer toward a brighter future where healthcare is efficient, more affordable and easily accessible to everyone, regardless of where they live.