Embedded Files
Background
Will artificial intelligence one day take over the world? Possibly, but AI has many other uses that are not commonly considered. Artificial intelligence has the ability to constantly improve its knowledge and decision capabilities, and this technology can be used in many areas of radiology to improve current techniques. However, utilizing this technology comes with drawbacks, such as doctors' willingness to use it. The rapid advances in artificial intelligence should be utilized in multiple areas of radiology to improve human observations and reduce radiation exposure to patients, despite minimal drawbacks. It is important that people first understand how artificial intelligence functions.
When thinking about artificial intelligence, most people think about robots that think for themselves, but this is not the whole picture. Ohio State University students commented, “Artificial intelligence (AI) is the development of computer algorithms to process and interpret data as well as perform tasks with partial or complete autonomy, while continuously refining its logic and decision making” (Von Ende et al. 1). The field of artificial intelligence has only become relevant to radiology with the most recent developments and capabilities including its decision making ability and the ability to continuously refine its logic.Radiologicial fields have different uses for artificial intelligence, specifically in interventional radiology. AI can improve image processing and read x-rays as well as determining a patient’s future care or procedures (Von Ende et al. 1). Throughout the field of radiology AI can be used in diagnostic radiology, interventional radiology, and even an area as overlooked as stock.
Uses
Augmented reality allows doctors to be in a simulated procedure which creates an environment where radiologists can still manipulate the procedure and carry it out without risk to patients. By fixing problems in the procedure before it starts, rather than while in the actual process, doctors speed up the process and decrease the exposure of radiation to both the patient and the surgeon. With advances in virtual reality simulations, a patient can potentially go through the procedure using VR to gain an idea of what will happen (Von Ende et al. 6). Artificial intelligence has many more uses in the radiology field such as reading x-rays.
Artificial intelligence is useful in the field of diagnostic radiology, as it is a specialty full of data rapidly progressing in modern times. In fact, radiologists have already begun implementing artificial intelligence into diagnostic radiology. These successes have led to improved efficiency and patient outcomes (Von Ende et al. 2). Incorporating artificial intelligence into diagnostic radiology can accomplish many tasks. For example, this new technology has been used to assess different brain related injuries, including strokes and tumors. Even though these types of injuries are under the field of diagnostic radiology, much overlap occurs between these examples and interventional radiology (Von Ende et al. 2). The programs are designed to scan the given image, process the information and then can read the image if designed for use in diagnostic radiology. If used in interventional radiology, the program can decide which treatments need to be followed and provide a step by step plan on how to accomplish this (Von Ende et al. 2).
Using recently created algorithms, artificial intelligence can produce summarized reports of information for specific patients. Radiologists can use this to be more efficient in daily practices and also reduce human errors. This technology would help in making accurate and thorough decisions for the given patients (Von Ende et al. 7). Error from humans can also be seen when it comes to reading the images. When trying to read medical images, humans can not take in the whole medical picture. When using artificial intelligence, DL (deep learning) can automatically distinguish almost all features from the given data allowing for growth in new patterns and deciphering more complex relationships. When integrated with interventional oncology and analyzing tumoral variables, which can be hard to see with the human eye, artificial intelligence can improve using an algorithm designed to work with numbers (Posa et al. 2). Artificial intelligence can also provide information regarding supply stock availability.
Benefits
Currently tools and information needed for a procedure are either gathered beforehand or by the other members on stand. Depending on the situation, this can be time-consuming. The introduction of automated devices such as voice-driven appliances in the field of interventional radiology could solve many of these issues (Von Ende et al. 7). Additionally, this technology can help with a serious issue-radiation. During procedures, radiation is often present to guide the surgeon through x-rays. This exposure level can be greatly reduced if AI is utilized. A study completed in an article titled Artificial Intelligence, Augmented Reality, and Virtual Reality Advances and Applications in Interventional Radiology states, “...it has already been evaluated in endoscopy with AI equipped fluoroscopy that reduces radiation exposure by 38% via ultrafast collimation” (Von Ende et. al 7). This shows that radiation exposure is significantly decreased when utilizing artificial intelligence in radiology. However, artificial intelligence can miss details that need a human eye to detect them.
One difficulty with integrating AI systems into radiology is that interventional radiology is constantly evolving making it difficult for the AI systems to benefit as they are intended. AI systems must be able to work with both modern and old imaging machines (Von Ende et al. 10). Another problem occurs because of human anatomy. Normal human anatomy varies across the globe depending on age, gender, and ethnicity. As of right now, no artificial intelligence program can be set up in a way to overlap the normal anatomy with diseases present. This is because the organs inside the human body are set up differently depending on the previous factors which makes it hard to set up a system that can adjust to all that information. This can lead to many complications such as misdiagnosis and mistakes in the procedures because of poor differentiation (Von Ende et al. 11). Another challenge arises with a willingness to use artificial intelligence.
Challenges
Some people accept integrating this technology in their day to day work, however, staff within hospitals and clinics might challenge integrating AI into practice. Some staff might either be unwilling to change the current process of how their clinic does everything, or they might be restricted by their own knowledge and ability to employ the technology that is new to them. This occurs because technology is advancing quickly, and staff are getting older. Many high level physicians believe AI technology will only disrupt their current way of practice and create new errors that can harm patients (Von Ende et al. 10). One difficulty observed with artificial intelligence is performance. Since this technology is so new, there could be many errors with artificial intelligence because of how complex the system is in general. AI consists of many unclear limits that could silently and detrimentally affect patient care (Von Ende et al. 10).
Despite the disadvantages of integrating artificial intelligence into the field of radiology, the technology should be utilized. AI is a relatively new technology that has the ability to keep progressing on its own. It can be used in many different ways, from an area as complicated as diagnostic radiology, to keeping items stocked in the medical room. With this technology, patients can experience reduced radiation exposure when taking much needed medical images. The radiologists themselves can also be trained more efficiently with the use of AI. Artificial intelligence can be utilized in the field of radiology to improve human observations and reduce radiation exposure to patients, despite few drawbacks. AI might take over the world some day, but until then, we should take advantage of this opportunity and utilize it in the medical field.
Page updated
Report abuse