Artificial Intelligence can help diagnose, prevent, and treat disease. Radiologists used to rely on a trained eye to diagnose, but often misdiagnosed patients as well. Recently AI has introduced computer algorithms that diagnose patients more accurately.
My research focused on CheXNet, an algorithm developed by Stanford researchers to evaluate chest X-rays for signs of pneumonia. CheXNet takes a chest X-Ray image and computes the percent chance of pneumonia. The algorithm can also diagnose up to 14 other types of diseases.
Detecting pneumonia in patients using X-ray images is very difficult. Two statistics that stress the significance of getting an accurate answer for patients are that one million adults are hospitalized, and fifty thousand die in the United States from pneumonia each year.
The objective of my research into CheXNet is to inform students of the evolving role AI is playing in diagnosing disease. Students will learn:
How AI can assist humans and how humans can assist AI.
Facilitation of accelerated diagnosis & medication nonadherence.
Career information for students seeking entrance into the healthcare industry.
Researchers collected data from four radiologists observing 420 pneumonia images from ChestX-ray14.
Researchers made modifications to CheXNet to detect all 14 types of medical conditions from ChestX-ray14.
The radiologists did not have prior access to patient information
The researchers found that CheXNet exceeded radiologist performance on pneumonia diagnosis. The results were based on a measure of accuracy called the F1 Score. The CheXNet score of 0.435 compared to 0.387 for the radiologists, was significant.
The researchers also found CheXNet achieves better results on other pathologies. The algorithm outperformed published results on all 14 pathologies in the ChestX–ray14 dataset.
With technology like CheXNet, artificial intelligence can be sent to locations that have a shortage of skilled radiologists. In conclusion, the algorithm helps reduce the number of missed cases of pneumonia, and by showing radiologists where to look first, AI can provide better medical care to patients.
According to an article titled “3 ways artificial intelligence is changing the healthcare industry”, between 1988 and 1994, roughly 38 percent of adults living in the United States were taking at least one prescription drug. Later the number would increase to 49 percent with 3.2 billion of those not being taken as directed or not taken at all. The algorithmic tools of AI will better identify which patients experience medication nonadherence and as a result cut out wasted prescriptions.
In the above graph from Accenture, AI could fill the gap in disparity between the supply and demand of clinicians. Since growth in the AI health market is expected to reach $6.6 billion by 2021, a background in computer science could help students break into the medical field since must curriculums have not yet shifted from information to the age of artificial intelligence. Mckinsey estimated in 2017 that 50% of activities carried out by workers have the potential to be automated. Something to keep an eye on for professionals seeking a career in healthcare.