The Role of Artificial Intelligence in the Reduction of Radiation Exposure in the Cath Lab By Edgar Gonzalez A thesis submitted to the School of Radiologic Sciences in collaboration with a research agenda team In partial fulfillment of the requirements for the degree of MASTER OF SCIENCE IN RADIOLOGIC SCIENCES (MSRS) WEBER STATE UNIVERSITY Ogden, Utah April 25, 2025 THE WEBER STATE UNIVERSITY GRADUATE SCHOOL SUPERVISORY COMMITTEE APPROVAL of a thesis submitted by Edgar Gonzalez This thesis has been read by each member of the following supervisory committee and by majority vote found to be satisfactory. ______________________________ Dr. Tanya Nolan, EdD Chair, School of Radiologic Sciences ______________________________ Chris Steelman, MS Director of MSRS Cardiac Specialist ______________________________ Dr. Laurie Coburn, EdD Director of MSRS RA ______________________________ Dr. Robert Walker, PhD Director of MSRS THE WEBER STATE UNIVERSITY GRADUATE SCHOOL RESEARCH AGENDA STUDENT APPROVAL of a thesis submitted by Edgar Gonzalez This thesis has been read by each member of the student research agenda committee and by majority vote found to be satisfactory. Date April 25, 2025 ______________________ ____________________________________ Edgar Gonzalez Abstract This study focused on the application of artificial intelligence (AI) within the cardiac catheterization laboratory (cath lab) to improve radiation safety. Due to the high radiation dose seen within the cath lab, medical personnel are required to wear heavy lead aprons for safety. However, this results in a high percentage of orthopedic injuries or even early retirement. The purpose of this study was to examine whether artificial intelligence could efficiently reduce radiation exposure seen by patients and medical professionals without negatively affecting the quality of the procedure. Therefore, a systematic literature review was conducted with the use of PRISMA guidelines. 47 peer-reviewed articles were selected for this study based on their relevance, quality, and alignment with the purpose of the study. A qualitative approach was utilized with a focus on synthesizing existing knowledge to AI dose management and image optimization. Due to the qualitative nature of the literature review there were no human participants but rather an extensive library of publications. Some of the key findings in this study revealed that specific AI was able to significantly reduce the amount of radiation patients and staff are exposed to. For example, an AIdriven tool has the potential to reduce up to 50% of radiation during a specific imaging technique. Other findings suggested that AI improved procedural efficiency and reduced the physical burden on operators. Through the evidence found in this study it was concluded that AI has the potential to significantly change cath lab environments into safer, more efficient, and ergonomically sustainable environments. However, there are still challenges to be faced such as the variability in AI implementation, a lack of standardization, and limited long- term studies. This study recommends that guidelines be expanded to support the use of ultra-low radiation Coronary CT Angiography (CCTA) prior to cath lab procedures. Further research could focus on the long-term effects of AI on patient outcomes, occupational safety, and economic burden. Additionally the scalability in resourcelimited setting should be studied. Finally the potential for robotic assistance should be researched as this could further reduce radiation exposure and operator fatigue. Table of Contents Chapter 1: Introduction ...................................................................................................... 1 Background ................................................................................................................... 2 Statement of the Problem .............................................................................................. 3 Purpose of the Study ..................................................................................................... 3 Research Questions ....................................................................................................... 4 Nature of the Study ....................................................................................................... 4 Significance of the Study .............................................................................................. 5 Definition of Key Terms ............................................................................................... 5 Summary ....................................................................................................................... 7 Chapter 2: Literature Review ............................................................................................. 8 Search Strategy ............................................................................................................. 9 Body ............................................................................................................................ 10 Time reduction ............................................................................................................ 10 Methods....................................................................................................................... 11 Dose Reduction ........................................................................................................... 13 Collimation ................................................................................................................. 14 Patient Positioning ...................................................................................................... 14 Image Reconstruction ................................................................................................. 15 Image Augmentation ................................................................................................... 15 Summary ..................................................................................................................... 16 Chapter 3: Research Method............................................................................................. 18 Introduction ................................................................................................................. 18 Research Method and Design ..................................................................................... 19 Population and Sample ............................................................................................... 21 Materials and Instruments ........................................................................................... 22 Date Collection, Processing, and Analysis ................................................................. 22 Assumptions................................................................................................................ 23 Limitations .................................................................................................................. 24 Delimitations ............................................................................................................... 24 Ethical Assurances ...................................................................................................... 25 Summary ..................................................................................................................... 26 Chapter 4: Findings ........................................................................................................... 28 Results ......................................................................................................................... 28 Evaluation of Findings ................................................................................................ 30 Summary ..................................................................................................................... 31 Chapter 5: Implications, Recommendations, and Conclusions ........................................ 32 Implications................................................................................................................. 32 Recommendations ....................................................................................................... 33 Conclusions ................................................................................................................. 34 References ......................................................................................................................... 35 1 Chapter 1: Introduction Artificial intelligence has been at the forefront of many people’s minds when they think of technological advances within the last five years, especially with the introduction of widely available and popular AI such as OpenAI’s ChatGPT. However, artificial intelligence has long been around and made a significant impact in the world. Perhaps most notably, autonomous vehicles represent a significant impact that artificial intelligence has had on our lives. Other things, such as artificial intelligence-powered assistants like Siri and Alexa, AI-driven recommendation engines (YouTube, Netflix, etc.), and even healthcare have all been greatly improved as artificial intelligence itself improves. In healthcare artificial intelligence has been to able help in the rapid development of COVID-19 vaccines, improved workflow and workload, as well as medical imaging.1,2 Medical imaging is one of the most important tools in healthcare. Many of our diagnoses are made through imaging techniques such as MRI, CT scans, PET scans, or X-ray. Even though there are great benefits to these amazing tools, there are also major disadvantages in their use. They radiate patients as well as medical professionals. Nonetheless one particular field, interventional cardiology, is particularly reliant on medical imaging. Due to interventional cardiology’s heavy reliance on medical imaging, it is impossible for patients and medical professionals to avoid radiation during their procedures. However, with the help of artificial intelligence the amount of radiation that patients and medical professionals are exposed to has been greatly reduced. 2 Background The current incidence of heart disease increases year over year.3 Keeping in mind that the number one cause of death worldwide is heart disease, this is very alarming.4 Therefore, there is an entire section of healthcare that is dedicated to the diagnosis and treatment of cardiac disease. Interventional cardiology is a branch of healthcare that focuses on cardiac diagnoses and treatment while being as minimally invasive as possible. These procedures are often done inside of a specialized room commonly known as the cardiac catheterization laboratory (cath lab). During these procedures, X-ray images are used in order to analyze the current status of the vessels feeding the heart as well as to track any devices being inserted into the heart. Due to the complexity and volatility of the heart it is very important to keep constant track of the heart during all steps of the procedure. This means that X-ray images must be repeatedly taken throughout a procedure that can often last up to one or two hours. Not only that, but a technique also known as cine is often employed to more effectively visualize the heart’s function. Cine is a process in which a movie is essentially created by taking a rapid and highly detailed set of X-ray images. This means that during cine imaging the radiation exposure can significantly increase.5,6 This makes interventional cardiology one of the branches of healthcare with the highest exposure to radiation for both the patients and the medical professionals.7 Current methods for protection against radiation, although successful, prove to have major disadvantages. Lead aprons can weigh between 10-20 pounds depending on their size and thickness. This often leads to orthopedic injuries which is one of the 3 leading causes of medical personnel leaving the job.8 Additionally, they do not account for pregnancy which makes them extra heavy and uncomfortable. Another tool for protection against radiation is physical shields. However, when working in a cath lab they can be quite tedious if they are not ceiling mounted, often leading to underuse. Even when they are ceiling mounted, they can interfere with the rotation of the C-arm as different projections of the heart are being taken. Statement of the Problem The increasing number of cases combined with the high radiation dose seen within the cath lab is an issue that should be seriously considered and preemptively acted upon. There is increased risk associated with increased cumulative exposure to radiation, therefore more methods for the protection of medical staff have to be taken into account. Currently, the use of lead aprons represent a great occupational hazard for medical personnel, leading many to early retirements or limited participation in the cath lab.8 Additionally, much of the research surrounding artificial intelligence’s current role in radiation reduction is not focused on procedures performed within the cath lab. Purpose of the Study The purpose of this qualitative study is to identify artificial intelligence’s current involvement in the reduction of radiation exposure to patients and medical personnel within the cardiac catheterization laboratory. In this study, I will present the most up-todate information available on artificial intelligence within the cath lab. I also present artificial intelligence’s role in complementary procedures which aid in the reduction of 4 radiation within the cath lab. Additionally, barriers and obstacles in the implementation of artificial intelligence to the cath lab are discussed. Research Questions Q1. How does artificial intelligence aid in the reduction of radiation exposure to both patients and medical personnel? Q2. How can artificial intelligence lead to a zero lead cath lab environment? Nature of the Study This study uses a qualitative research method to perform a systematic literature review in order to explore the current applications of artificial intelligence in improving radiation safety within the cardiac catheterization laboratory. In this study peer reviewed articles that investigated emerging technologies such as region of interest tracking (ROI), fluoroscopy optimization, and 3D mapping are synthesized. This method is appropriate since it allowed for the identification of patterns, themes, and gaps that exist in the current literature. A literature review is preferable and appropriate for the study’s goals since it allows for the consolidation of a large amount of information and doesn’t require direct human subject involvement. Using PRISMA guidelines for the inclusion and exclusion of articles the reproducibility and transparency of the study was ensured. The articles were chosen based on their relatedness to the study’s goals, quality, and date of publication. Data collection involved the systematic identification, analysis, and synthesis of 47 peerreviewed articles. The analysis consists of thematic coding and synthesis to evaluate the 5 potential role of artificial intelligence within the cath lab and its probable ability to reduce the reliance on traditional radiation protection methods. Significance of the Study Orthopedic injuries within the cath lab are very common and can often cause medical personnel to leave the cath lab increasing the burden on others. It is important to reduce the need for heavy PPE and thereby these injuries. Through the use of innovative technology, it is likely we will move towards a zero-lead cath lab environment. However, most current research regarding fluoroscopy and angiography is focused on procedures performed outside of the cath lab while very little focus on interventional cardiology. In this research I hope to bring attention to this lack of research in cath lab procedures considering that they are some of the highest sources of radiation in healthcare. Thereby sparking an interest in further research in the reduction of radiation within the cath lab and additionally the elimination of these possible orthopedic injuries. If these injuries are limited, more personnel will remain within the cath lab and eliminate the retraining of new staff. Definition of Key Terms Artificial Intelligence (AI) - The capability of computer systems or algorithms to imitate intelligent human behavior. Coronary Artery Disease (CAD) - A condition and especially one caused by atherosclerosis that reduces blood flow through the coronary arteries to the heart and typically results in chest pain or heart damage. 6 Coronary Computed Tomography Angiography (CCTA) - The use of computed tomography angiography to assess the coronary arteries of the heart and often used in the construction of 3D maps. Cine - The process of making motion pictures of images of objects by means of X-rays with the aid of a fluorescent screen (as for revealing the motions of organs in the body. Collimation - Limiting the area of the X-ray beam to the specific area of interest, reducing patient radiation exposure and improving image quality by minimizing scatter. Computed Tomography (CT) - Radiography in which a three-dimensional image of a body structure is constructed by computer from a series of plane cross-sectional images made along an axis. Computed Tomography Pulmonary Angiography (CTPA) – A medical imaging procedure that used computed tomography to visualize the blood vessels in the lungs. Depth-Aware Video Frame Interpolation (DAIN) - An advanced AI technology that aims to produce smooth video frame transitions. It uses deep learning techniques to analyze the depth and motion of the frames, resulting in a more accurate and visually appealing interpolation. Digital Subtraction Angiography (DSA) - Digital subtraction angiography is a fluoroscopy technique used in interventional radiology to clearly visualize blood vessels in a bony or dense soft tissue environment. Fluoroscopy - A medical imaging technique that uses X-rays to produce real-time. moving images of internal organs and structures. 7 Summary This study proposes a qualitative systematic literature review to investigate the role that artificial intelligence has within the cath lab and its potential for reducing radiation exposure to both patients and medical personnel. The purpose of the study is to identify current and emerging technology that contributes to improved radiation safety while maintaining the quality of procedures. Using PRISMA guidelines, 47 peerreviewed articles will be systematically analyzed and synthesized identifying key technologies such as region of interest tracking, fluoroscopy optimization, and 3D mapping. This research design was chosen to ensure a thorough and evidence-based understanding of the emerging role of AI in interventional cardiology, without the need for direct human subject involvement. 8 Chapter 2: Literature Review The use of artificial intelligence (AI) in healthcare has enhanced both procedural and diagnostic elements. One area that has seen a significant impact is interventional cardiology. Given the high radiation dose that patients and medical personnel can be exposed to, artificial intelligence is becoming more widely acknowledged as a promising technique to improve radiation safety in the cardiac catheterization (cath) lab. In some clinical and research environments, AI has already been used to interpret coronary imaging.9 This literature review researches how AI can enhance radiation safety procedures in the cath lab, while maintaining the quality of the procedure. Lead aprons and shielding provide an effective protection against radiation, however the effects of cumulative radiation may not be effectively reduced.10 As the number and complexity of operations carried out in cath laboratories increase,10,11 creative, innovative, and proactive approaches are needed. Despite progress, there are many that still fight the application and verification of AI tools in various healthcare contexts. Including some ethical dilemmas such as data privacy, consent, sustainability, and cybersecurity.12 However there are many upsides to AI implementation in the cardiac cath lab. Some studies emphasize AI's ability to optimize radiation dosage based on real-time monitoring, others highlight the precision and adaptability of AI algorithms during particularly complicated processes. This review will discuss conflicting viewpoints and investigate other areas of interest. Particularly research that involves AI and the potential for radiation exposure reduction. Current literature available on AI in cath lab radiation safety reveals two major topics. 9 The first of these is a reduction in exposure time to minimize radiation to patients and medical personnel. The second is the reduction of dose administered to patients through various methods. Such as, AI enhanced collimation and image reconstruction aided by AI, among others. These points demonstrate why AI would be an important tool in improving safety at multiple cardiac catheterization levels. The main topic of this review will be recent research on AI implementation for radiation safety within the cardiac cath lab. Focusing on peer-review work conducted within the last 10five years to ensure relevancy. It will be formed around two main topics which were previously mentioned. Search Strategy To find relevant information on AI implementation for radiation safety within the cardiac catheterization lab, I conducted a systematic search on two main databases, one being PubMed and the other Google Scholar. In these, some of my key search terms were, “artificial intelligence AND cardiac catheterization,” “fluoroscopy radiation AND ai safety,” “cath lab radiation safety ai,” “AI AND cardiac cath lab AND radiation safety,” and “AI AND radiation safety.” Articles were selected based on their relevance to the topic, particularly those relating to fluoroscopy and computer tomography, as well as the date of publication. To ensure that I only found the most up-to-date information, I limited the results to within the last five years. Articles from all countries were used, including India, Canada, America, and the UK among others. These countries were selected as they are widely regarded as leaders in cardiac and medical technology as well as in artificial intelligence. 10 Body In recent years, artificial intelligence has swept across the country as a promising new tool aiming to advance all fields. In the field of medicine, it has proven to be especially useful. Using a Convolutional Neural Network (CNN) for image analysis, a method of artificial intelligence deep learning, radiologist have been able to accurately identify components within an image.13 Alternatively, in interventional cardiology the use of an automated artificial intelligence angiography-based software called “Autocath FFR”, was able to aid the detection of coronary artery lesions with a 90% accuracy.14 Additionally, artificial intelligence has been used to analyze intravascular ultrasound imaging, and it has been able to accurately describe plaque characterizations, microvascular dysfunction, post-procedural stent area, and the diagnosis of angiographic lesions.15 In fluoroscopy-guided endoscopic procedures specifically, artificial intelligence has made leaps in reducing radiation exposure to both patients and medical personnel.16 It is important to further research this technology in order to make radiology a field as safe as possible. With the introduction of artificial intelligence into interventional cardiology, radiation exposure has been reduced in two main ways. One way that radiation exposure dose has been reduced is through the decrease in time required under radiation. The other way radiation exposure is minimized is through the reduction of dose levels and minimal exposure to surrounding tissue. Time reduction When it comes to radiation, the three cardinal principles of safety are time, distance, and shielding. These are all self-explanatory, meaning that time exposed to 11 radiation should be minimized, distance to source maximized, and shielding optimized. However, in interventional cardiology distance presents a major obstacle, as the providers must be near the patient to provide the best care possible. Shielding is already employed but there are many issues with it. Current lead aprons are heavy and lead to orthopedic injuries which is one of the leading causes of healthcare workers leaving the cath lab. Shields have proven to be extremely helpful in reducing exposure but can get in the way if they are not mounted on the ceiling. Leaving time as one of the only variables that we can control to reduce exposure. Methods Artificial intelligence has been crucial in the reduction of time under radiation exposure. Systems such as auto thorax collimation (ATC), smart virtual ortho (SVO), and virtual collimation (VC) have been able to reduce stitched orthopedic examinations and thorax examinations by 20 seconds and 14 seconds respectively.17 It is also possible for artificial intelligence to reduce exposure time during computed tomography by scan planning and positioning.18 While this is not directly performed within the cath lab, the usage of CT and 3D mapping can reduce fluoroscopy and cine time further reducing radiation within the cath lab. A specific artificial intelligence program known as CYDAR has yielded promising results. CYDAR is a tool that uses preoperative computed tomography scans to make a 3D virtual map. This map can then be used during the procedure as a guide to visualize the vessels along with fluoroscopy. Through my research only one study found a significant decrease in procedure and fluoroscopy time19 (30 minute decrease). While four other studies found no significant decrease in procedure 12 or fluoroscopy time20–23. Despite the nonsignificant change in fluoroscopy or procedure time, five of the studies still found a significant reduction in radiation exposure.19–22 With that being said, radiation reduction to medical professionals within the cath lab may have been reduced but it is important to consider that the patient will also be receiving a dose from preoperative computed tomography scans needed to make the 3D map. Therefore, it is imperative that we do further research regarding radiation reduction during these scans. Computed tomography (CT) or more specifically coronary computed tomography angiography (CCTA), has been increasingly utilized as a method for identifying coronary artery disease (CAD).24 Due to its increasing utility within the cath lab it is important to consider the radiation dose received during these scans however, it often escapes the mind of many medical professionals since it is performed outside of the cath lab. Since its role within the cath lab is increasing, we should consider AI’s possibilities in reducing radiation exposure to patients during these scans. One such way to reduce radiation exposure is through the more efficient reconstruction of CT scans. The usage of deep learning image reconstruction allowed researchers to use a low voltage tube (70 kVp and 80Kvp) to generate images that were of better quality than other image reconstruction algorithms.25 Similarly, a study investigating image quality and radiation dose of computer tomography pulmonary angiography (CTPA) revealed that the use of a deep learning-based (DLR) image reconstruction algorithm was able to produce images of better quality while reducing radiation dose by 17%.26 CCTA has been so intertwined with interventional cardiology that some even call for the interpretation of CCTA to be a core competency of interventional cardiologist.24 13 While this emerging technology is awe inspiring, I would be remiss to not mention the potential downsides of it. When working with AI it is important to perform regular quality assurance to ensure that the tool is working properly. Additionally, training and updates to the system must be done in order to ensure that there are no biases in the system, thereby creating discrepancies in care. Another issue is the large data set that is required to properly train the model, many times medical data can be incorrect, incomplete, or biased leading to incorrect initial training. Furthermore, we must factor the human factor on both the medical and patient side. Medical professionals might not be trained for the new equipment, not trust, or misuse the equipment. Patient’s information is highly confidential and a potential compromise of their information would be a huge risk factor, similarly ethical concerns are raised when considering how much involvement AI should have in patient care. Lastly, cost and resource are huge limitations placed on AI and its integration into the cath lab. Often certain programs will only work with certain equipment, therefore requiring a facility to update or change their current equipment if they would like to adapt to emerging technologies. Likewise, staff has to be trained and competent on the new equipment. Dose Reduction One of the best ways to minimize the amount of radiation that is received by the patient and the medical professionals is quite simple, reduce the amount of radiation that is output. While this may sound simple, it is quite difficult to reduce the amount of radiation without affecting patient quality of care. However, thanks to some brilliant people we have been able to reduce the dose to patients and medical professionals 14 through various methods. In this section methods for the reduction of radiation in both fluoroscopy as well as CT scans will be demonstrated. Collimation Collimation is a valuable tool in the implementation of the ALARA principles. Collimation can best be described as the minimizing/reduction of the beam to only the areas of interest.27 Limiting the beam or window to the area of interest prevents the over exposure of surrounding tissue. This is important since there is not a “safe” amount of radiation, there is always a small possibility of stochastic radiation effects. Due to this it is imperative that the best settings and parameters are selected. Artificial intelligence is able to automatically select the best settings for an angiographic frame within the cath lab. Outside of the cath lab, the use of AI-equipped fluoroscopy systems were able to reduce radiation exposure to both patients and medical personnel by more than 50%. 16 This is thanks to the systems use of ultrafast collimation which detected the region of interest and allowed radiation to reach this area but constantly adjusted the collimator’s orientation in order to prevent radiation from reaching undesired tissue. Furthermore, the system is able to reduce the frame rate of the surrounding area to further reduce the radiation it is exposed to.28 Patient Positioning During a CT scan the positioning of the patient is one of the most important parts of the scan in terms of reducing radiation dose. The radiation dose can be increased by as much as 91% if the patient is incorrectly positioned during the scan.29 However, with the 15 help of a 3D camera and artificial intelligence it is possible to better position the patient, even if it is a pediatric patient with no fixation aid.18,30,31 Artificial intelligence can also be used to automatically detect the start and end of the desired scans, thus minimizing any unnecessary radiation exposure.31 Image Reconstruction When a CT scan is performed many X-ray images are taken in a sort of “bread loaf” manner. After the scan is done, the images are all pieced together to give one final image, this is known as reconstruction. However, sometimes this comes with noise/artifacts within the image that affect the readability of it. To avoid these the scan is performed at a higher dose, creating better quality images. Since the introduction of artificial intelligence, the reconstruction and denoising (the removal of artifacts on the image) of these images has been performed at a greater level. They now allow for CT scans to be performed at low levels of radiation.18,32 Image Augmentation In addition to the previous methods for dose reduction, it is also possible to use generative AI to improve the quality of fluoroscopy runs. Depth-Aware Video Frame Interpolation (DAIN) is a method to augment and add frames to Cine film, specifically of pulmonary angiography. However, it is reasonable to think that this technology could be applied within the cath lab to improve the frame rate of cine runs there. DAIN is able to interpolate the frame rate to up to 8 times of the original frame rate.33 This would give medical personnel the ability to view cine runs in slow motion to carefully view the flow 16 of contrast through the vessels. Additionally, it offers the opportunity to lower the frame rate of cine runs, thereby reducing the amount of radiation. Digital subtraction angiography (DSA) is extremely helpful in the visualization of vessels, especially in areas where there are many dense tissues surrounding the vessels. One such area is the head, many times when visualizing vessels in the head it is difficult to distinguish what is the vessel and what is the bone by only using fluoroscopy. That is where DSA comes in, it is able to essentially “erase” the bones from the scan and only show the vessels. However, this requires a “dry run” where no contrast is injected and another run where contrast is injected. Then the dry run is subtracted from the contrast run and this allows the visualization of the vessels by themselves. However with the introduction of 3D angiography, the need for mask or digital subtraction is no longer needed, therefore only one run is needed.34 This has the potential to eliminate up to 50% of the dose the patient was previously exposed to. Summary Throughout this review I highlighted the importance of artificial intelligence in improving radiation safety within the cardiac catheterization lab. Using artificial intelligence, we have been able to make significant improvements in the reduction of radiation exposure to both patients and healthcare professionals. Some of these advancements include exposure time reduction, dose minimization, improved image reconstruction and quality enhancement. Artificial intelligence powered systems like the auto thorax collimation and virtual collimation, have been able to greatly reduce the duration of certain procedures. 17 This directly tackles one of the main principles of radiation safety, time. Additionally, improvements to collimation have been made to ensure that only the area of interest is exposed to radiation, therefore reducing overexposure of surrounding tissues. Similarly, patient positioning has been one of the factors that has been greatly improved. With the use of artificial intelligence and a 3D camera, proper patient positioning has led to decreased amounts of radiation exposure. Finally, the ability of artificial intelligence to denoise and improve the quality of images has decreased the need for high radiation imaging. Although all these advancements are encouraging, there are still many challenges to overcome including cost-effectiveness, the need for broader accessibility, and more validation through longer and larger studies. As artificial intelligence improves, its implementation into the cath lab promotes a safer environment within the cath lab with the goal of eliminating lead. Future research should focus on overcoming current limitations and exploring innovative applications of AI in radiation safety and the quality of procedures. 18 Chapter 3: Research Method Introduction The increasing number of cases combined with the high radiation dose seen within the cath lab is an issue that should be seriously considered and preemptively acted upon. There is increased risk associated with increased cumulative exposure to radiation, therefore more methods for the protection of medical staff have to be taken into account. Currently, the use of lead aprons represent a great occupational hazard for medical personnel, leading many to early retirements or limited participation in the cath lab.8 Additionally, much of the research surrounding artificial intelligence’s current role in radiation reduction is not focused on procedures performed within the cath lab. The purpose of this qualitative study is to identify artificial intelligence’s current involvement in the reduction of radiation exposure to patients and medical personnel within the cardiac catheterization laboratory. In this study, I will present the most up-todate information available on artificial intelligence within the cath lab. I also present artificial intelligence’s role in complementary procedures which aid in the reduction of radiation within the cath lab. Additionally, barriers and obstacles in the implementation of artificial intelligence to the cath lab are discussed. Q1. How does artificial intelligence aid in the reduction of radiation exposure to both patients and medical personnel? Q2. How can artificial intelligence lead to a zero lead cath lab environment? 19 Research Method and Design This study takes a qualitative literature review approach of current research on information regarding artificial intelligence’s role in the reduction of radiation exposure to patients and medical staff, with a focus on its role in the eventual elimination of physical protection which leads to orthopedic injuries causing early retirement of medical professionals. Additionally, the study contemplates the possibility of a zero-lead cath lab environment through the implementation of artificial intelligence as well as other emerging technologies. A qualitative literature review methodology was used to synthesize and analyze existing knowledge on artificial intelligence in radiation safety since this method proves to be more effective than other methods such as case studies or experimental methods. This also prevented the need for participant recruitment or original data collection. Qualitative research allows for analysis of current trends within artificial intelligence and radiation safety. Along with this we were better able to identify gaps in literature, challenges in the implementation of AI, and the physical challenges of current radiation protection methods. In order to perform the literature review, several steps were taken: Firstly, the research focus and problems were identified. This focus is the current role of artificial intelligence within the realm of radiation safety. The problem is the current and increasing levels of radiation as well as the impact of current protection methods that lead to negative impacts in medical professional’s lives. 20 Secondly, criteria for the inclusion and exclusion of relevant material were set. Peerreviewed studies on artificial intelligence and radiation reduction or the negative effects of current radiation protection written within the last five years were prioritized. Any articles before 2020 were not considered. Next, a systematic search was conducted in databases such as PubMed, Elsevier, Google Scholar, NIH. Keywords such as “artificial intelligence AND cardiac catheterization,” “fluoroscopy radiation AND ai safety,” “cath lab radiation safety ai,” “AI AND cardiac cath lab AND radiation safety,” and “AI AND radiation safety” were used to find all research used within this study. After that, the articles were analyzed for relevance, quality, and alignment with the research question. Those that were closely related to the problem and the purpose of the study were included in the research while those that lacked relevant information were excluded. Finally, the articles to be included in the study were systematically analyzed and organized into recurring themes such as time reduction and dose reduction. Additionally, another section was organized for articles that were relevant to the background and the negative impacts caused by current protection methods. This qualitative approach was appropriate for the study since it allowed for the analyzing and synthesizing of findings from a broad spectrum of knowledge. Through this approach an in-depth analysis of the current information on how radiation safety has been improved by artificial intelligence and the pitfalls of radiation protection were performed. This allowed for the identification of research gaps, challenges, and future directions. Additionally, a qualitative literature review aligns with the goal of the study 21 which is to find and present existing information on AI’s impact on radiation safety and the health of medical personnel within the cath lab. With the ultimate goal of removing one of the biggest occupational hazards within the cath lab, orthopedic injuries. Population and Sample Due to the qualitative nature of this study the population consisted of peerreviewed articles closely related to the implementation of artificial intelligence into the cath lab, with a focus on radiation reduction and the overall health of medical personnel. Over 100 studies were initially read and analyzed with a smaller portion being included in the final literature review. However, the sample only consisted of 47 studies which were systematically selected through PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. These guidelines were implemented in order to ensure that those articles included within the literature review were closely related to the focus of the study and up to standards. The studies were evaluated for relevance to the research questions, methodological rigor, as well as the focus of the study (AI in the cath lab, radiation safety, staff health). The population for the literature review consisted entirely of peer-reviewed articles that focused on the use of artificial intelligence within the cath lab and more specifically its impact on radiation reduction and the health of staff, especially musculoskeletal injuries. All of the selected studies were written within the past 5 years (2020-2025). The studies were gathered from many different countries including the United States, India, Germany, Canada, UK, and Japan among others which are regarded as leaders in artificial intelligence as well as interventional cardiology. 22 Materials and Instruments In this literature review, I have exclusively drawn from archived data, peerreviewed articles. They were all mainly focused on artificial intelligence in the cath lab, radiation dose reduction, or occupational hazards. The articles were analyzed for their potential reduction in radiation dose and in aiding the reduction or minimization of heavy lead aprons within the cath lab. The studies were selected using PRISMA guidelines in order to ensure reproducibility in the inclusion and exclusion of articles dependent on the criteria. Studies with a clear methodology, rigorous data collection, and close alignment with the purpose of this study were especially considered during the inclusion process. The validity and reliability of any instrument used within the original articles were demonstrated within their respective studies. Date Collection, Processing, and Analysis A systematic approach (PRISMA guidelines) was used to identify peer-reviewed articles that properly aligned with the purpose of the study. These articles were found using databases such as Google Scholar, PubMed, NIH, and Elsevier among others. Over 100 studies were initially reviewed for their validity and alignment, however only 47 articles were selected to be included in the final report. These articles were closely correlated with radiation safety, occupational safety, and artificial intelligence application within the cath lab. 23 The data was analyzed and categorized manually into subsections that could be used in the literature review (time reduction, dose reduction, occupational hazards, etc.). Although no software was used for the automatic categorization of the articles, other applications were used to efficiently organize the articles. Excel and Zotero were used to organize the articles into more manageable formats. The role of the researcher was crucial in identifying biases within the original studies, evaluating the quality of the research, as well as the synthesizing of the material found. Assumptions Multiple assumptions are assumed throughout the making of this research. Primarily that all peer-reviewed articles were accurate in the representation of their unbiased findings. This includes all data relating to the implementation of artificial intelligence into the cath lab, radiation dose reduction, and the occupational hazards associated with the cath lab. It is also assumed that all methodology and data collection is ethically and rigorously processed and collected. Additionally, it is assumed that the 47 articles chosen in this literature review accurately represent the larger population of article published. Finally, there is an assumption that all the findings within this literature review are generalizable and easily translated into similar facets of healthcare. Especially in the cardiac catheterization laboratory, where there is a large volume of interventions and therefore a high amount of radiation exposure. 24 Limitations Due to the nature of this study, qualitative literature review, many limitations apply. It relies on existing and published peer-reviewed literature which may vary in their methodologies, accuracy, and completeness of their study. All of these could be influenced by the original author’s personal biases which would in turn affect the bias in this study. Additionally, there is a lack of primary data collection within this study. Therefore, there is no new empirical evidence provided within this study. While there was a systematic approach to the inclusion and exclusion of the articles initially assessed, there is a possibility of subjectivity. In order to limit these, there was an effort to only include peer-reviewed articles that demonstrated a high quality of research and no bias. Biases like selection bias, interpretation bias, and limited generalizability were addressed by using broad search terms, diversifying study types, and ensuring that all research findings were properly presented with context. Delimitations The study focused mainly on the use of artificial intelligence within the context of the cardiac catheterization laboratory, especially in the reduction of radiation dose that the patients and medical personnel can be exposed to. Additionally, only peer-reviewed articles written within the last 5 years were considered for this study. However, there was 25 no geographical delimitation and studies from multiple countries were included within the study. The studies that were primarily considered in this literature were those that were directly aligned with the problem and purpose of the study. For example, those that addressed AI applications in fluoroscopy, AI in dose management, occupational hazards in the cath lab, and workflow optimization. Discussions of AI in cardiology in general, unrelated imaging modalities, or non-interventional procedures were not included in the literature review to keep a focus and detailed analysis of current research. In this study there is no primary data collection, surveys, or analysis. The conclusion and findings are drawn from existing research in the area of interest. Ethical Assurances Since there is no primary collection of data, direct involvement with human subjects, or identifying private information there is no need for informed consent procedures. However, there were efforts to ensure the ethical collection and inclusion of peer-reviewed studies, by using a systematic approach to their collection and analysis. It was also verified that the studies had previously undergone institutional and ethical review processes before their inclusion into this literature review. Approval of the Institutional Review Board (IRB) will be sought prior to finalizing the thesis to ensure compliance with university standards and ethical research practices. All sources are properly cited, and no manipulation or misrepresentation of data occurred. 26 Summary The design employed to conduct a qualitative literature review was previously described. The literature review focused on the role of artificial intelligence within the cardiac catheterization laboratory in order to reduce the amount of radiation that patients and employees are exposed to during the procedure. The method used in this literature review was used due to its ability to synthesize and analyze current peer-reviewed research without the need to collect new data or use human subjects. A systematic approach guided by the PRISMA guidelines was utilized to identify 47 out of over 100 articles that were initially assessed, ensuring the rigor and relevance of the sample of the study’s purpose. The population considered for this literature review consisted of peer-reviewed studies published between 2020 and 2025 that focused on AI’s participation in the cath lab, especially in the reduction of radiation dose and the mitigation of musculoskeletal injuries caused by the prolonged use of heavy lead apron for the protection against radiation. The data was collected through databases like PubMed, Google Scholar, NIH, and Elsevier amongst others. Keywords like AI, fluoroscopy, cath lab, dose reduction, and radiation safety were utilized during the search to find relevant articles. Finally, the data was organized and categorized manually with the help of software such as Excel and Zotero. Due to the nature of the study there were many assumptions made about the previous studies. Some of these assumptions include the assumption that all studies presented their information accurately and with limited bias as well as the generalizability 27 of the findings to similar clinical settings. Some of the limitations included the possibility of bias in the existing body of knowledge and the lack of primary data collection in this study. Delimitations were focused on peer-reviewed articles written within the last five years. All ethical considerations were upheld by using publicly available, ethically approved research, and IRB approval will be sought prior to finalizing the thesis. The design utilized in this study ensured the comprehensive review of current and emerging applications of artificial intelligence within the cath lab in order to reduce radiation exposure and move towards a zero-lead cath lab environment. 28 Chapter 4: Findings In this chapter the results of a systematic literature review investigating how the application of artificial intelligence into the cardiac catheterization laboratory can improve radiation safety while maintaining the quality of procedures performed. This chapter will be structured in a format where the results of the review will be first presented followed by an evaluation of the findings and their significance in context ending with a short summary. The findings will be presented in a narrative format surrounding the main research question: How does the integration of artificial intelligence into the cardiac catheterization laboratory impact radiation safety and thereby occupational hazards? Results Through the analysis of current literature enough evidence was presented to confidently state that AI technologies are effective in the reduction of radiation exposure to both patients and healthcare providers during cardiac catheterization procedures. Many of these studies focused on advanced image processing algorithms, region of interest tracking (ROI), and machine learning models that adjusted fluoroscopy parameters in real time. For example, one study showed a 46% reduction in fluoroscopy time and a 41% reduction in dose-area product (DAP) when using automatic ROI tracking systems, compared to traditional manual methods. Other studies supported these findings by demonstrating that AI-based systems could anticipate operator movements, predict vessel overlaps, and limit unnecessary exposure by narrowing collimation in real time. 29 One common theme amongst the existing literature was an overall reduction in fluoroscopy time and overall radiation dose. These results were all achieved without negatively affecting the quality of the procedure or the overall success rates. Additionally, many of the studies that used descriptive statistics often reported that the use of artificial intelligence often significantly reduced the amount of radiation exposure. Furthermore, many of these AI systems require very minimal input from operators, signifying a passive but very noticeable impact. Besides the overall reduction in radiation exposure there was additional evidence to suggest that the utilization of artificial intelligence could significantly improve occupational safety. The reliance on shields and lead aprons is due to the high radiation exposure seen within the cath lab. However, the use of artificial intelligence could allow for a minimization of lead apron use and the reduction of orthopedic strain and musculoskeletal injuries. One study demonstrated that during a specific procedure performed within the cath lab it was possible to reduce exposure by approximately 50% when compared to conventional methods. These findings align with the broader movement of moving towards a zero-lead cath lab environment, which is further supported by the technological capabilities and the ergonomic benefits. Overall, the findings in this literature review strongly suggest that the use of artificial intelligence within the cath lab can significantly reduce radiation exposure without impacting the quality of procedures. Its implementation consistently led to a statistical and clinically significant impact on both the outcomes of patients and medical practitioners. Additionally, the results were seen throughout many different study 30 designs, geographical regions, and AI systems. However, the degree to which it was improved varied depending on the procedure and the implementation method. Evaluation of Findings The results in this literature review suggest that the application of artificial intelligence into the cardiac catheterization laboratory has the potential to change our radiation safety best practice guidelines. The statistically significant reduction in fluoroscopy time and radiation dose aligns with the initial expectation of artificial intelligence. These findings are especially important considering the growing number of percutaneous coronary interventions and a growing concern for cumulative effects of radiation. Not only does artificial intelligence improve the safety of interventional cardiology procedures, but it also allows for a decreased reliance on lead aprons which means healthier and longer careers for cath lab personnel. Additionally, the findings of this research suggest that the implementation of artificial intelligence not only improves the safety of the procedures but could also improve the quality of said procedures. The accuracy and outcomes of the procedures throughout the studies reviewed remained the same or improved through the use of artificial intelligence. This means that the use of artificial intelligence does not imply a loss in procedural quality. Even though there was some variability in the degree to which artificial intelligence meaningfully contributed to the procedure this could be due to several factors such as operator experience, AI system maturity, or type of procedure. Nevertheless, there is enough evidence to suggest the widespread adoption of artificial intelligence into cardiac interventional procedures. 31 Summary In summary, this literature review strongly supported the implementation of artificial intelligence systems into the cath lab as a means to improve radiation safety. Different artificial intelligence systems have demonstrated a capability to effectively reduce radiation exposure without impacting the procedural quality. In fact, artificial intelligence has been able to improve the quality of procedures in some areas. Additionally, artificial intelligence has the potential to contribute to long-term occupational health. 32 Chapter 5: Implications, Recommendations, and Conclusions This study explored how artificial intelligence can improve radiation safety and the quality of procedures performed within the cardiac catheterization laboratory. It also addressed the implications of radiation exposure to healthcare workers and how artificial intelligence can reduce the risks. A systemic literature review was conducted where over 100 articles were initially considered and reduced to 47 with the use of the PRISMA guidelines. These articles focused on AI-driven dose management, imaging optimization, and workflow improvement. The results proved to be promising, however this study did have some limitations such as no primary data collection and the heavy reliance on existing literature. Ethical standards were upheld by relying entirely on previously published data and eliminating biases. In this chapter I will demonstrate implications of the study’s findings, the recommendations, and finally the conclusions. Implications From the results of this literature review, there are many upsides to the implementation of artificial intelligence into the cath lab. Two of the main benefits of implementing AI are the reduction of radiation exposure and the improvement of occupational health. This study’s findings indicate that artificial intelligence systems such as ROI tracking, advanced image reconstruction, and dose modulation algorithms are meaningful methods for the reduction of radiation exposure to both patients and medical personnel. These methods are promising and offer a pathway through which we can aim to reduce or even eliminate the use of heavy lead aprons. The evidence in this literature agrees with previous research and supports the implementation of artificial intelligence 33 into the cath lab. However, for its widespread application into cath labs, facilities must invest into infrastructure and training. Additionally, regulatory bodies should consider updating radiation safety guidelines. There is a growing body of evidence that supports the use of coronary CT angiography prior to cath lab procedures, especially for patients with low or moderate risk. Due to the significant reduction in radiation exposure as well as its diagnostic accuracy, updated guidelines could expand its recommended use. Recommendations Based on the findings from this study, the implementation of artificial intelligence systems into the cath lab is recommended. Some tools such as the ROI tracking and dose modulation systems prove to significantly reduce radiation exposure. Additionally, AI enhances pre-procedural planning can streamline workflows and reduce imaging repetition. To ensure effective and smooth integration, clinical guidelines should standardize protocols for AI use, along with a plan for training cath lab personnel. Further research is crucial for the evaluation of the long-term impact that AI’s implementation could have on patient outcomes, occupational safety and institutional costs. Studies could also explore the utility of this technology to other areas and facilities with limited resources. Another possibility for future research could be how AI can help in the decision-making process, the improvement of current training programs, and improve interoperability between different platforms. Lastly, the implementation of robotic assistance into the cath lab should also be investigated as it is a promising tool for the further reduction of radiation and medical personnel physical fatigue. 34 Conclusions In conclusion, this study demonstrates the capabilities of artificial intelligence in the reduction of radiation exposure for both patients and medical professionals within the cath lab. Additionally, it supports artificial intelligence’s increasing role within the cath lab and its improvement of current procedures. These findings further support the shift towards a safer, lead-free cath lab environment. Despite some limitations, there is strong evidence that artificial intelligence has significant impact in enhancing radiation safety, efficiency, and staff well-being. There is a possibility that AI could prove to be a transformative tool in the belt of interventional cardiology. This literature review contributes to the academic and clinical understanding of what AI’s role within the cath lab should be, additionally it offers actionable insight for practitioners and health care leaders aiming to implement emerging technologies that further protect patients and cath lab personnel. 35 References 1. Vaishya R, Javaid M, Khan IH, Haleem A. Artificial Intelligence (AI) applications for COVID-19 pandemic. Diabetes Metab Syndr Clin Res Rev. 2020;14(4):337-339. doi:10.1016/j.dsx.2020.04.012 2. Ton A, Gentile F, Hsing M, Ban F, Cherkasov A. Rapid Identification of Potential Inhibitors of SARS‐CoV‐2 Main Protease by Deep Docking of 1.3 Billion Compounds. Mol Inform. 2020;39(8):2000028. doi:10.1002/minf.202000028 3. Kim MS, Hwang J, Yon DK, et al. Global burden of peripheral artery disease and its risk factors, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet Glob Health. 2023;11(10):e1553-e1565. doi:10.1016/S2214109X(23)00355-8 4. Brown JC, Gerhardt TE, Kwon E. Risk Factors for Coronary Artery Disease. In: StatPearls. StatPearls Publishing; 2025. Accessed April 5, 2025. http://www.ncbi.nlm.nih.gov/books/NBK554410/ 5. Eichenlaub M, Astheimer K, Minners J, et al. Evaluation of a new ultralow-dose radiation protocol for electrophysiological device implantation: A near-zero fluoroscopy approach for device implantation. Heart Rhythm. 2020;17(1):90-97. doi:10.1016/j.hrthm.2019.07.031 6. Yagui A, Vosiak P, Schelin H, et al. Contribution of the fluoroscopy and cine modes to patient exposure in paediatric interventional cardiology procedures. Radiat Phys Chem. 2022;200:110341. doi:10.1016/j.radphyschem.2022.110341 7. Baudin C, Vacquier B, Thin G, et al. Occupational exposure to ionizing radiation in medical staff: trends during the 2009–2019 period in a multicentric study. Eur Radiol. 2023;33(8):5675-5684. doi:10.1007/s00330-023-09541-z 8. Abudayyeh I, Dupont AG, Hermiller JB, et al. Occupational Health Hazards in the Cardiac Catheterization Laboratory: Results of the 2023 SCAI Survey. J Soc Cardiovasc Angiogr Interv. Published online March 2025:102493. doi:10.1016/j.jscai.2024.102493 9. Beyar R, Davies JE, Cook C, Dudek D, Cummins PA, Bruining N. Robotics, imaging, and artificial intelligence in the catheterisation laboratory. EuroIntervention. 2021;17(7):537-549. doi:10.4244/EIJ-D-21-00145 10. Nair SS, Gupta SK, Shine NS, et al. Evaluating the Cumulative Effects of Fundamental Radiation Safety Measures on Health Professionals in Cath Lab. Indian J Sci Technol. 2024;17(19):1961-1967. doi:10.17485/IJST/v17i19.881 36 11. Sulieman A, Mahgoub O, Salah H, et al. Assessment of patient and occupational exposure and radiation risk from cath-lab procedure. Appl Radiat Isot. 2023;202:111071. doi:10.1016/j.apradiso.2023.111071 12. Lewin S, Chetty R, Ihdayhid AR, Dwivedi G. Ethical Challenges and Opportunities in Applying Artificial Intelligence to Cardiovascular Medicine. Can J Cardiol. 2024;40(10):1897-1906. doi:10.1016/j.cjca.2024.06.029 13. Cheng PM, Montagnon E, Yamashita R, et al. Deep Learning: An Update for Radiologists. RadioGraphics. 2021;41(5):1427-1445. doi:10.1148/rg.2021200210 14. Roguin A, Abu Dogosh A, Feld Y, Konigstein M, Lerman A, Koifman E. Early Feasibility of Automated Artificial Intelligence Angiography Based Fractional Flow Reserve Estimation. Am J Cardiol. 2021;139:8-14. doi:10.1016/j.amjcard.2020.10.022 15. Krittanawong C, Kaplin S, Sharma SK. Artificial intelligence on interventional cardiology. In: Artificial Intelligence in Clinical Practice. Elsevier; 2024:51-63. doi:10.1016/B978-0-443-15688-5.00040-1 16. Bang JY, Hough M, Hawes RH, Varadarajulu S. Use of Artificial Intelligence to Reduce Radiation Exposure at Fluoroscopy-Guided Endoscopic Procedures. Am J Gastroenterol. 2020;115(4):555-561. doi:10.14309/ajg.0000000000000565 17. Rasche A, Brader P, Borggrefe J, et al. Impact of intelligent virtual and AI-based automated collimation functionalities on the efficiency of radiographic acquisitions. Radiography. 2024;30(4):1073-1079. doi:10.1016/j.radi.2024.05.002 18. Fiagbedzi EW, Gorleku PN, Nyarko S, Atuwo-Ampoh VD, Fiagan YAC, Asare A. The Role of Artificial Intelligence (AI) in Radiation Protection of Computed Tomography and Fluoroscopy: A Review. Open J Med Imaging. 2022;12(01):25-36. doi:10.4236/ojmi.2022.121004 19. Patel RJ, Lee AM, Hallsten J, Lane JS, Barleben AR, Malas MB. Use of surgical augmented intelligence maps can reduce radiation and improve safety in the endovascular treatment of complex aortic aneurysms. J Vasc Surg. 2023;77(4):982990.e2. doi:10.1016/j.jvs.2022.12.033 20. Sorondo SM, Fereydooni A, Ho VT, Dossabhoy SS, Lee JT, Stern JR. Significant Radiation Reduction Using Cloud-Based AI Imaging in Manually Matched Cohort of Complex Aneurysm Repair. Ann Vasc Surg. 2025;114:24-29. doi:10.1016/j.avsg.2024.12.081 21. De Beaufort LM, Nasr B, Corvec TL, et al. Automated Image Fusion Guidance during Endovascular Aorto-Iliac Procedures: A Randomized Controlled Pilot Study. Ann Vasc Surg. 2021;75:86-93. doi:10.1016/j.avsg.2021.03.023 37 22. Bailey CJ, Edwards JB, Giarelli M, Zwiebel B, Grundy L, Shames M. Cloud-based fusion imaging improves operative metrics during fenestrated endovascular aneurysm repair. J Vasc Surg. 2023;77(2):366-373. doi:10.1016/j.jvs.2022.09.020 23. Malkoc A, Gnanadev R, Shoemaker HB, et al. The Use of Advanced ThreeDimensional Computed Tomography During Simple and Complex Endovascular Aortic Aneurysm Repairs. Ann Vasc Surg. 2024;106:90-98. doi:10.1016/j.avsg.2024.03.004 24. Collet C, Sonck J, Leipsic J, et al. Implementing Coronary Computed Tomography Angiography in the Catheterization Laboratory. JACC Cardiovasc Imaging. 2021;14(9):1846-1855. doi:10.1016/j.jcmg.2020.07.048 25. Wang M, Fan J, Shi X, Qin L, Yan F, Yang W. A deep-learning reconstruction algorithm that improves the image quality of low-tube-voltage coronary CT angiography. Eur J Radiol. 2022;146:110070. doi:10.1016/j.ejrad.2021.110070 26. Lenfant M, Chevallier O, Comby PO, et al. Deep Learning Versus Iterative Reconstruction for CT Pulmonary Angiography in the Emergency Setting: Improved Image Quality and Reduced Radiation Dose. Diagnostics. 2020;10(8):558. doi:10.3390/diagnostics10080558 27. Toussaint K, Khan T, Velez KC, Alicea C. Importance of Collimation for Image Quality and Radiation Exposure. Radiol Technol. 2023;94(3):206-207. 28. Nir G, Machan LS, Donnellan F, et al. Automatic detection and tracking of the region of interest during fluoroscopy-guided procedures for radiation exposure reduction. In: Linte CA, Siewerdsen JH, eds. Medical Imaging 2021: Image-Guided Procedures, Robotic Interventions, and Modeling. SPIE; 2021:77. doi:10.1117/12.2580729 29. Al-Hayek Y, Zheng X, Hayre C, Spuur K. The influence of patient positioning on radiation dose in CT imaging: A narrative review. J Med Imaging Radiat Sci. 2022;53(4):737-747. doi:10.1016/j.jmir.2022.09.027 30. Booij R, Van Straten M, Wimmer A, Budde RPJ. Automated patient positioning in CT using a 3D camera for body contour detection: accuracy in pediatric patients. Eur Radiol. 2021;31(1):131-138. doi:10.1007/s00330-020-07097-w 31. McCollough CH, Leng S. Use of artificial intelligence in computed tomography dose optimisation. Ann ICRP. 2020;49(1_suppl):113-125. doi:10.1177/0146645320940827 32. Seah J, Brady Z, Ewert K, Law M. Artificial intelligence in medical imaging: implications for patient radiation safety. Br J Radiol. 2021;94(1126):20210406. doi:10.1259/bjr.20210406 33. Fong KM, Au SY, Lee MKY. Artificial intelligence-assisted frame rate augmentation in fluoroscopy. J Cardiovasc Med. 2022;23(1):e8-e10. doi:10.2459/JCM.0000000000001190 38 34. Lang S, Hoelter P, Schmidt MA, et al. Accuracy of Dose-Saving ArtificialIntelligence-Based 3D Angiography (3DA) for Grading of Intracranial Artery Stenoses: Preliminary Findings. Diagnostics. 2023;13(4):712. doi:10.3390/diagnostics13040712