San Diego, CA- June 10th, 2021
Artificial intelligence has infiltrated various aspects of our everyday lives through social media algorithms to targeted digital ads to speech recognition in our mobile devices. Naturally, AI has already navigated its way into healthcare, looking to further influence the way physicians provide patient care.
How It All Began
Interest in artificial intelligence began in the 1950’s when Alan Turing questioned a computer’s ability to “think” with reason. After the publication of Turing’s paper, “Computing Machinery and Intelligence”, AI enthusiasm exploded across the nation as scientists explored if and how it would be possible for machines to embody human characteristics during computation. It continued to grow with monumental milestones achieved over the next 70 years with big dreams for its potential.
By the beginning of the 21st century, scientists were equipped with better, faster, and more advanced technology; the capabilities of machine computations seemed limitless. AI integration and research branched into all aspects of life, business, and healthcare.
Present Day Integration
Speech recognition software, introduced as a tool to help physicians report more efficiently, was one of the first widely adopted applications of AI in radiology. By eliminating the monotonous task of writing or typing out findings, AI helped enable radiologists to do what they do best- interpret medical images and provide diagnoses for patients. As speech recognition technology improved, so did its acceptance in radiology. Today, it is viewed as an essential tool in medical image interpretation and directly influential on the productivity of radiologists across the nation.
Future Role of AI
While it has been envisioned by sci-fi books and movies that robots would eventually replace humans, it is commonly accepted in the healthcare industry that AI will be used as a tool to improve our own human capabilities. With medical data doubling every 2 months, physicians can improve efficiency, accuracy and overall productivity with AI assistance and keep up with overwhelming demand. The future of AI has the potential to impact radiology far beyond speech recognition and reporting.
In the coming years, artificial intelligence will grow to play a role in advanced image visualization. Detecting areas of concern, measuring volumes of legions or tumors, or even acting as a second set of eyes to catch possible errors, AI can help improve overall patient care and physician performance. Artificial intelligence is a strong tool that has the capability of truly changing the way radiologists perform medical image interpretations and ensure patients receive excellent care, regardless of which hospital they go to.
So Why Not Now?
AI is advancing every day, but it needs to be able to check off some key criteria before it is completely ready to be adopted by radiologists.
- Accuracy- Quite obvious, yet still an important feature to point out. Software needs to be precise and reproduceable. As a tool, it will have difficulty attracting use from large hospitals if it is unable to provide accurate and similar results when used in various environments. More development is necessary to bring consistency in performance.
- Productivity- AI needs to be easily incorporated into a radiologist and hospital’s workflow. Does it have the ability to adapt to each hospital’s needs? Can it easily work with existing technology, or does it require a workaround? It should be saving radiologists time- providing shortcuts without losing quality.
- Cost- AI is expensive to develop and expensive to use. It needs high-quality, diversified data (patient demographics, locations, techniques, etc.) that can be used to train the model, which comes at a premium price. Factoring all aspects of development, the client faces steep costs to gain access to the product. Until the overall return is worth the investment, AI will continue to struggle to be completely adopted by radiologists and hospitals due to the high price tag.
Conclusion
Regardless of the challenges it faces during research and developmental phases, AI will not disappear from headlines anytime soon. Patients, physicians, and developers hold excitement about the positive impact AI can have on standardizing and improving overall patient care. The question is no longer if AI will integrate into radiology, but rather when and how. To find out about that, we will just have to wait and see.
About StatRad:
With over 25 years of experience providing quality reads, quick TATs, and customized workflows, StatRad is a true partner you can count on to provide excellent patient care and teleradiology services. Contact us today to learn why StatRad is the best teleradiology provider for you!