AI-powered Hyperrealism: Next Step in Cinematic Rendering?

Background Recent advancements in artificial intelligence (AI)-powered image generation present opportunities to enhance three-dimensional medical images. Diffusion, an iterative denoising process, represents the standard of many of the current tools used for this purpose. Purpose To demonstrate the current capabilities of diffusion technology by using Midjourney, version 5.2, a text-to-image generative AI tool, and present a practical guide for its use. Materials and Methods This exploratory study investigates the principles, parameters, and prompt engineering techniques for generating images focusing on Midjourney from July 27 to August 3, 2023. Step-by-step instructions show the innate capability of this technology in creating realistic medical images. Results Thirty images were selected, including eye, skin, and vascular aneurysm images. Varying prompt phrasing and weighting techniques allowed for the customization of output image characteristics. Although the details of Midjourney's model training are confidential, it is estimated that it was trained on at least hundreds of millions of images from the web. Anatomic fidelity was not always maintained because the training data set is not necessarily based on accurate medical images. There are shortcomings in this nascent technology regarding its ability to create entities such as digits of the hand or precise text. Conclusion AI image generation has the potential to improve three-dimensional medical images for certain applications through added visual detail and appeal but ongoing collaboration is needed between radiologists and AI developers due to the overreliance on art and photography in the training data, which may result in inaccurate anatomic results. Moreover, the evolving landscape of ethical discussions and copyright stipulations warrants close attention. © RSNA, 2024 Supplemental material is available for this article.

The Intersection of Radiology With Blockchain and Smart Contracts: A Perspective.

INTRODUCTION: Although blockchain technology and smart contracts are garnering attention in various sectors, their applications and familiarity within the realm of radiology remain largely unexplored. Blockchain, a decentralized digital ledger technology, offers secure, transparent, and resilient data management by distributing the verification process across a network of independent entities. This decentralized technology presents a possible solution for a range of healthcare challenges, from secure data transfer to automated verification processes. To address such challenges in the context of medical imaging, blockchain could provide different approaches, including smart contracts, machine learning algorithms, and the secure dissemination of large files among key stakeholders such as patients, healthcare providers, and institutions. This manuscript aims to explore the current attitudes and perspectives of trainees and radiologists to the utilization of blockchain technology and smart contracts in clinical radiology. Additionally, the study provides an in-depth analysis of the potential applications for incorporating blockchain into radiology. METHODS: After obtaining The George Washington University Committee on Human Research Institutional Review Board (IRB) approval, we conducted a 10-question survey among radiologists and trainees at several institutions and private practices. Surveys were created via the Google Forms application and were emailed to potential participants. Participants were asked about their current academic level (medical student, resident/fellow, academic radiologist, private practice radiologist, others), their knowledge level about the field of imaging informatics and blockchain and smart contract technologies, their level of interest in learning more about blockchain and smart contracts, and their opinion about possible applications of blockchain and smart contract in the future of medical imaging. RESULTS: A total of 118 survey requests were distributed; 83 were returned, reflecting a 70.3% overall response rate. Of these, 19 were sent to private practices with a 15.8% response rate (3/19), and 99 to academic centers, yielding an 80.8% response rate (80/99). The survey respondents demonstrated a strong interest and need to further understand these technologies among radiologists and trainees. This study focuses on key components of this technology as it relates to healthcare and the practice of radiology, including data storage, patient care, secure communication, and automation, as well as strengths, weaknesses, opportunities, and threats (SWOT) analysis. DISCUSSION: To our knowledge, this is the first study to investigate and establish a baseline for the current perspectives on the application of blockchain technology and smart contracts in clinical radiology amongst trainees and radiologists across academic and private settings. Incorporating blockchain and smart contracts technologies into the field of radiology has the potential to achieve greater efficiency, security, and patient empowerment. However, the adoption of this technology comes with challenges, such as infrastructure, interoperability, scalability, and regulatory compliance. Collaboration between radiologists, hospital administration, policymakers, technology developers, and patient advocacy organizations will help guide and advance our understanding of the potential applications of blockchain and smart contracts in radiology and healthcare.

AI-Powered Chatbots in Medical Education: Potential Applications and Implications.

Artificial intelligence (AI) is anticipated to have a considerable impact on the routine practice of medicine, spanning from medical education to clinical practice across specialties and, ultimately, patient care. With the imminent widespread adoption of AI in medical practice, it is imperative that medical schools adapt to the use of these advanced technologies in their curriculum to produce future healthcare professionals who can seamlessly integrate these tools into practice. Chatbots, AI systems programmed to process and generate human language, are currently being evaluated for various tasks in medical education. This paper explores the potential applications and implications of chatbots in medical education, specifically in learning and research. With their capability to summarize, simplify complex concepts, automate the creation of memory aids, and serve as an interactive tutor and point-of-care medical reference, chatbots have the potential to enhance students' comprehension, retention, and application of medical knowledge in real-time. While the integration of AI-powered chatbots in medical education presents numerous advantages, it is crucial for students to use these tools as assistive tools rather than relying on them entirely. Chatbots should be programmed to reference evidence-based medical resources and produce precise and trustworthy content that adheres to medical science standards, scientific writing guidelines, and ethical considerations.

Eyes on AI: ChatGPT's Transformative Potential Impact on Ophthalmology.

ChatGPT, a large language model by OpenAI, has been adopted in various domains since its release in November 2022, but its application in ophthalmology remains less explored. This editorial assesses ChatGPT's potential applications and limitations in ophthalmology across clinical, educational, and research settings. In clinical settings, ChatGPT can serve as an assistant, offering diagnostic and therapeutic suggestions based on patient data and assisting in patient triage. However, its tendencies to generate inaccurate results and its inability to keep up with recent medical guidelines render it unsuitable for standalone clinical decision-making. Data security and compliance with the Health Insurance Portability and Accountability Act (HIPAA) also pose concerns, given ChatGPT's potential to inadvertently expose sensitive patient information. In education, ChatGPT can generate practice questions, provide explanations, and create patient education materials. However, its performance in answering domain-specific questions is suboptimal. In research, ChatGPT can facilitate literature reviews, data analysis, manuscript development, and peer review, but issues of accuracy, bias, and ethics need careful consideration. Ultimately, ensuring accuracy, ethical integrity, and data privacy is essential when integrating ChatGPT into ophthalmology.

Pearls and pitfalls in emergency CT neuroangiography through the lens of bias and error.

Computed tomography angiography (CTA) of the head and neck is central in emergency department (ED) evaluation of clinically suspected acute stroke and intracranial hemorrhage. Timely and accurate detection of acute findings is crucial for best clinical outcomes; missed or delayed diagnosis can be devastating. Our pictorial essay presents twelve CTA cases that provided significant diagnostic dilemmas to on-call trainees while reviewing current bias and error classifications in radiology. Among others, we discuss anchoring, automation, framing, satisfaction of search, scout neglect and zebra-retreat bias. Each imaging vignette depicts a potential diagnostic "pitfall" while introducing types of cognitive bias/error before concluding with a concrete "pearl" for CTA interpretation. We believe that familiarity with bias and error is particularly important in the ED setting where high case volume, high acuity and radiologist fatigue intersect. Particular attention to personal cognitive biases and these potential CTA pitfalls may help emergency radiologists transition from habit-driven pattern recognition to analytical thinking, ultimately improving diagnostic decision making.

An Unusual Case of Neurenteric Cyst in a Patient with Split Cord Malformation.

Neurenteric cyst in a split cord malformation is a rare finding. We report an adult female becoming acutely symptomatic secondary to an expanding neurenteric cyst, though previous imaging had demonstrated stability. We discuss our workup and management with surgical resection and possible etiologies of her acute decline.

Generative Artificial Intelligence Terminology: A Primer for Clinicians and Medical Researchers.

Generative artificial intelligence (AI) is rapidly transforming the medical field, as advanced tools powered by large language models (LLMs) make their way into clinical practice, research, and education. Chatbots, which can generate human-like responses, have gained attention for their potential applications. Therefore, familiarity with LLMs and other promising generative AI tools is crucial to harness their potential safely and effectively. As these AI-based technologies continue to evolve, medical professionals must develop a strong understanding of AI terminologies and concepts, particularly generative AI, to effectively tackle real-world challenges and create solutions. This knowledge will enable healthcare professionals to utilize AI-driven innovations for improved patient care and increased productivity in the future. In this brief technical report, we explore 20 of the most relevant terminology associated with the underlying technology behind LLMs and generative AI as they relate to the medical field and provide some examples of how these topics relate to healthcare applications to help in their understanding.

Organs in Color: Utilizing Free Software and Emerging Multi Jet Fusion Technology to Color and Surface Label 3D-Printed Anatomical Models.

3D printing (3DP) is a rapidly evolving innovative technology that has already been utilized for the development of educational anatomic models. Until recently, it was difficult and tedious to create multi-colored models and especially labels due to technological constraints. In this technical note, a comprehensive guide for creating labeled and color-coded anatomic models was created using free software, Blender. We have composed a step-by-step process for taking an existing 3D model and adding labeling and color that is compatible with modern high-quality 3D printing technologies (Multi Jet Fusion). We provided colored and labeled 3D renderings of the surface anatomy of the brain, ventricular system of the brain, the segments of the liver, and coronary arteries as examples of the diverse potential of this technology. Additionally, we 3D printed actual models of the surface anatomy of the brain and ventricles of the brain using HP Multi Jet Fusion to demonstrate the potential of this technology in the creation of anatomic models.

Congress of Neurological Surgeons systematic review and evidence-based guidelines update on the role of imaging in the management of progressive glioblastoma in adults.

TARGET POPULATION: These recommendations apply to adults with glioblastoma who have been previously treated with first-line radiation or chemoradiotherapy and who are suspected of experiencing tumor progression. QUESTION: In patients with previously treated glioblastoma, is standard contrast-enhanced magnetic resonance imaging including diffusion weighted imaging useful for diagnosing tumor progression and differentiating progression from treatment-related changes? LEVEL II: Magnetic resonance imaging with and without gadolinium enhancement including diffusion weighted imaging is recommended as the imaging surveillance method to detect the progression of previously diagnosed glioblastoma. QUESTION: In patients with previously treated glioblastoma, does magnetic resonance spectroscopy add useful information for diagnosing tumor progression and differentiating progression from treatment-related changes beyond that derived from standard magnetic resonance imaging with and without gadolinium enhancement? LEVEL II: Magnetic resonance spectroscopy is recommended as a diagnostic method to differentiate true tumor progression from treatment-related imaging changes or pseudo-progression in patients with suspected progressive glioblastoma. QUESTION: In patients with previously treated glioblastoma, does magnetic resonance perfusion add useful information for diagnosing tumor progression and differentiating progression from treatment-related changes beyond that derived from standard magnetic resonance imaging with and without gadolinium enhancement? LEVEL III: Magnetic resonance perfusion is suggested as a diagnostic method to differentiate true tumor progression from treatment-related imaging changes or pseudo-progression in patients with suspected progressive glioblastoma. QUESTION: In patients with previously treated glioblastoma, does the addition of single-photon emission computed tomography (SPECT) provide additional useful information for diagnosing tumor progression and differentiating progression from treatment-related changes beyond that derived from standard magnetic resonance imaging with and without gadolinium enhancement? LEVEL III: Single-photon emission computed tomography imaging is suggested as a diagnostic method to differentiate true tumor progression from treatment-related imaging changes or pseudo-progression in patients with suspected progressive glioblastoma. QUESTION: In patients with previously treated glioblastoma, does 18F-fluorodeoxyglucose positron emission tomography add useful information for diagnosing tumor progression and differentiating progression from treatment-related changes beyond that derived from standard magnetic resonance imaging with and without gadolinium enhancement? LEVEL III: The routine use of 18F-fluorodeoxyglucose positron emission tomography to identify progression of glioblastoma is not recommended. QUESTION: In patients with previously treated glioblastoma, does positron emission tomography with amino acid agents add useful information for diagnosing tumor progression and differentiating progression from treatment-related changes beyond that derived from standard magnetic resonance imaging with and without gadolinium enhancement? LEVEL III: It is suggested that amino acid positron emission tomography be considered to assist in the differentiation of progressive glioblastoma from treatment related changes.

Temporal lobe epilepsy with a contralateral parietal seizure-onset zone.

Achieving sustained seizure freedom following epilepsy surgery remains a challenge in some patients. Lesional temporal lobe epilepsy (TLE), for example, in patients with mesial temporal sclerosis or other MRI abnormalities, carries a good prognosis for seizure freedom compared to significantly lower chances of seizure freedom in patients with non-lesional epilepsy. However, even in some lesional TLE cases, persistent post-operative seizures suggest seizure onset from a brain region that is clinically and electrographically silent but manifests only after propagation to the temporal lobe. A notable example of such a brain region is the parietal lobe, which has extensive connectivity to various brain regions. While certain seizure semiologies, for example, sensory seizures, suggest parietal lobe onset, some medial parietal seizures may be semiologically indistinguishable from temporal lobe seizures. Here, we report a patient with focal impaired awareness seizures that manifested semiologically and electrographically as left TLE but proved to originate from the contralateral medial parietal lobe. We discuss putative seizure propagation pathways.

The Time Has Come: a Paradigm Shift in Diagnostic Radiology Education via Simulation Training.

Current radiology training for medical students and residents predominantly consists of reviewing teaching files, attending lectures, reading textbooks and online sources, as well as one-on-one teaching at the workstation. In the case of medical schools, radiology training is quite passive. In addition, the variety of important and high-yield cases that trainees are exposed to may be limited in scope. We utilized an open-source dcm4chee-based Picture Archiving and Communication System (PACS) named "Weasis" in order to simulate a radiologist's practice in the real world, using anonymized report-free complete cases that could easily be uploaded live during read-outs for training purposes. MySQL was used for database management and JBOSS as application server. In addition, we integrated Weasis into a web-based reporting system through Java programming language using the MyEclipse development environment. A freeware, platform-independent, image database was established to simulate a real-world PACS. The sever was implemented on a dedicated non-workstation PC connected to the hospital secure network. As the client access is through a webpage, the cases can be viewed from any computer connected to the hospital network. The reporting system allows for evaluation purposes and providing feedback to the trainees. Brief survey results are available. Implementation of such a low-cost, versatile, and customizable tool provides a new opportunity for training programs in offering medical students with an active and more realistic radiology experience, junior radiology residents with potentially better preparation for independent call, and senior resident and fellows with the ability to fine-tune high-level specialty-level knowledge.

Effectiveness of Three-Dimensionally Printed Models in Anatomy Education for Medical Students and Resident Physicians: Systematic Review and Meta-Analysis.

INTRODUCTION: Despite a surge in the use of three-dimensional printing (3DP) in medical education, a comprehensive evaluation of randomized trials in its effectiveness is lacking. Radiologic studies play an integral role in affording educators the ability to create customized realistic anatomic models. This systematic review and meta-analysis sought to assess the effect of 3DP versus traditional 2-D methods for anatomy education. METHODS: PubMed, Scopus, Cochrane Library, ERIC, and IEEE Xplore were queried to identify randomized controlled trials that quantitatively investigated anatomy education via postintervention assessments of medical students or resident physicians who were exposed to 3DP versus traditional methods. Criteria for the meta-analysis required that studies additionally included a pre-intervention assessment. RESULTS: A total of 804 articles were reviewed, identifying 8 and 7 studies for systematic reviews of medical students and resident physicians, respectively, of which 4 and 7 were included in the meta-analyses. 3DP models were associated with higher anatomy examination scores for medical students (P < .0001), but for resident physicians were statistically not significant (P = .53). DISCUSSION: The 3DP models are shown to positively impact medical students especially given their limited fund of knowledge in anatomy. It is postulated that the lack of a statistically significant result for the resident physicians was multifactorial, in part because of the small test group sizes introducing noise and nonrepresentative samples, as well as relative simplicity of the 3DP models used with resident physicians, which were below their level of training. More trials are required to evaluate the usefulness of highly customized 3DP models.

Use of augmented reality for image-guided spine procedures.

PURPOSE: Because of its ability to superimpose imaging data on a patient, while anchoring the user's view to the immediate surroundings, augmented reality (AR) has the potential to dramatically improve the accuracy and reduce the time required for preoperative planning and performance of minimally invasive spine surgeries and procedures. Described and reported herein is the direct clinical application of AR navigation on a series of common percutaneous image-guided spine procedures. MATERIALS AND METHODS: AR, including a "virtual needle" (VN) asset, was used to guide and navigate a total of 18 procedures performed on 10 patients. Comparative control data were generated using a phantom model (n = 32). These data are used to determine the accuracy of AR for federal drug administration submissions. Optical codes were implemented to allow automatic and real-time registration. A manual process was used when the use of optical codes was not available. Target error, distance to the target and target size were measured for both phantom and clinical groups. Mean errors between the two groups were compared. RESULTS: Target error between the control and clinical data sets showed no significant difference. Moreover, the distance to the target site and the target size had no effect on target acquisition. CONCLUSIONS: This data set suggests that AR navigation, utilizing a VN, is an emerging, accurate, valuable additive method for surgical and procedural planning for percutaneous image-guided spinal procedures and has potential to be applied to a broad range of clinical and surgical applications.

Unusual presentation of submandibular lingual nerve sheath tumour as sublingual stone.

A 60-year-old woman was referred to the otolaryngologist for 18 months of left-sided tongue pain and taste changes. Surgeon-performed ultrasound of the submandibular region revealed a hyperechoic mass. Wharton's duct was dilated proximally and the submandibular gland demonstrated normal vascularity. While these findings were highly suspicious for submandibular gland sialolith, an in-office attempt at sialolithotomy suggested an alternate process or mass. After imaging failed to further elucidate an aetiology, surgical exploration revealed a well-circumscribed submandibular mass associated with the lingual nerve. The mass was removed en-bloc and pathology revealed a schwannoma of the lingual nerve.

From CT to 3D Printed Models, Serious Gaming, and Virtual Reality: Framework for Educational 3D Visualization of Complex Anatomical Spaces From Within-the Pterygopalatine Fossa.

We describe the framework for capturing the internal view of complex anatomical spaces via multiple media and haptic platforms, exemplified by realistic and conceptual representations of the pterygopalatine fossa (PPF). A realistic three-dimensional (3D) mesh of the PPF was developed by segmenting the osseous anatomy on computed tomography (CT) using Materialize InPrint. Subsequently in Autodesk 3D Studio Max, the realistic mesh was enhanced with graphically designed neurovascular anatomy and additionally a conceptual representation of the PPF with its connections and contents was created. An interactive web-compatible Adobe Flash tutorial using ActionScript was developed, allowing users to advance through a series of educational slides that contained interactive rotatable interior camera views and scrollable CT cross-sectional content, incorporating both the realistic and conceptual models. Both models were also 3D printed using polyamide material. In the realistic model, the neurovasculature was colored with water-based acrylic paint. A 3-piece modular design with embedded magnets allows for internal visualization and seamless assembly. A serious gaming environment of the conceptual PPF was also developed using Truevision3D application programming interface, where users can freely move around rooms and hallways that represent various spaces. Lastly, the realistic model was incorporated into a headset-based virtual reality environment, Surgical Theater, allowing visualization and fly-through inside and outside the model. Multiple 3D techniques for visualization of complex 3D anatomical spaces from within were described, with the necessary software and skills detailed. A rough estimate of the time and cost needed to develop these tools as well as multiple supplementary source and end result files are also made available. Educators could utilize multiple advanced delivery methods to incorporate custom digital 3D models of complex anatomical spaces understood from inside.