Geoffrey Currie, Johnathan Hewis, Elizabeth Hawk, Eric Rohren
{"title":"Gender and Ethnicity Bias of Text-to-Image Generative Artificial Intelligence in Medical Imaging, Part 2: Analysis of DALL-E 3.","authors":"Geoffrey Currie, Johnathan Hewis, Elizabeth Hawk, Eric Rohren","doi":"10.2967/jnmt.124.268359","DOIUrl":"https://doi.org/10.2967/jnmt.124.268359","url":null,"abstract":"<p><p>Disparity among gender and ethnicity remains an issue across medicine and health science. Only 26%-35% of trainee radiologists are female, despite more than 50% of medical students' being female. Similar gender disparities are evident across the medical imaging professions. Generative artificial intelligence text-to-image production could reinforce or amplify gender biases. <b>Methods:</b> In March 2024, DALL-E 3 was utilized via GPT-4 to generate a series of individual and group images of medical imaging professionals: radiologist, nuclear medicine physician, radiographer, nuclear medicine technologist, medical physicist, radiopharmacist, and medical imaging nurse. Multiple iterations of images were generated using a variety of prompts. Collectively, 120 images were produced for evaluation of 524 characters. All images were independently analyzed by 3 expert reviewers from medical imaging professions for apparent gender and skin tone. <b>Results:</b> Collectively (individual and group images), 57.4% (<i>n</i> = 301) of medical imaging professionals were depicted as male, 42.4% (<i>n</i> = 222) as female, and 91.2% (<i>n</i> = 478) as having a light skin tone. The male gender representation was 65% for radiologists, 62% for nuclear medicine physicians, 52% for radiographers, 56% for nuclear medicine technologists, 62% for medical physicists, 53% for radiopharmacists, and 26% for medical imaging nurses. For all professions, this overrepresents men compared with women. There was no representation of persons with a disability. <b>Conclusion:</b> This evaluation reveals a significant overrepresentation of the male gender associated with generative artificial intelligence text-to-image production using DALL-E 3 across the medical imaging professions. Generated images have a disproportionately high representation of white men, which is not representative of the diversity of the medical imaging professions.</p>","PeriodicalId":16548,"journal":{"name":"Journal of nuclear medicine technology","volume":null,"pages":null},"PeriodicalIF":1.0,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142502306","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A Rare Case of Orbital Metastasis from Invasive Lobular Carcinoma: Challenges of <sup>18</sup>F-FDG PET/CT and the Search for Consensus on Imaging.","authors":"Marjorie Lam, Johnny Yang, Vani Vijayakumar","doi":"10.2967/jnmt.124.267732","DOIUrl":"https://doi.org/10.2967/jnmt.124.267732","url":null,"abstract":"<p><p>Invasive lobular carcinoma (ILC) frequently underlies orbital metastasis. A 74-y-old woman, who was current with mammograms and had no cancer history, presented to her ophthalmologist with visual complaints and was found to have metastatic ILC. MRI was contraindicated, and an <sup>18</sup>F-FDG PET/CT scan revealed a mildly hypermetabolic right orbital mass and low uptake in the left subareolar breast, suggestive of metastatic ILC. Small studies have found that in ILC, 16α-<sup>18</sup>F-fluoroestradiol and fibroblast activation protein inhibitors are more avid than <sup>18</sup>F-FDG. There is currently no consensus regarding imaging for ILC. Many people have contraindications to MRI, and the higher rate of false-negative findings on mammography for ILC than for other breast cancers makes this patient population more vulnerable to inaccurate staging, incorrect assessment of tumor burden, and, consequently, insufficient treatment. We provide this interesting case to highlight the potential of <sup>18</sup>F-fluoroestradiol PET/CT and fibroblast activation protein inhibitors over <sup>18</sup>F-FDG in this breast cancer subtype.</p>","PeriodicalId":16548,"journal":{"name":"Journal of nuclear medicine technology","volume":null,"pages":null},"PeriodicalIF":1.0,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142502303","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Geoffrey Currie, Johnathan Hewis, Elizabeth Hawk, Eric Rohren
{"title":"Gender and Ethnicity Bias of Text-to-Image Generative Artificial Intelligence in Medical Imaging, Part 1: Preliminary Evaluation.","authors":"Geoffrey Currie, Johnathan Hewis, Elizabeth Hawk, Eric Rohren","doi":"10.2967/jnmt.124.268332","DOIUrl":"https://doi.org/10.2967/jnmt.124.268332","url":null,"abstract":"<p><p>Generative artificial intelligence (AI) text-to-image production could reinforce or amplify gender and ethnicity biases. Several text-to-image generative AI tools are used for producing images that represent the medical imaging professions. White male stereotyping and masculine cultures can dissuade women and ethnically divergent people from being drawn into a profession. <b>Methods:</b> In March 2024, DALL-E 3, Firefly 2, Stable Diffusion 2.1, and Midjourney 5.2 were utilized to generate a series of individual and group images of medical imaging professionals: radiologist, nuclear medicine physician, radiographer, and nuclear medicine technologist. Multiple iterations of images were generated using a variety of prompts. Collectively, 184 images were produced for evaluation of 391 characters. All images were independently analyzed by 3 reviewers for apparent gender and skin tone. <b>Results:</b> Collectively (individual and group characters) (<i>n</i> = 391), 60.6% were male and 87.7% were of a light skin tone. DALL-E 3 (65.6%), Midjourney 5.2 (76.7%), and Stable Diffusion 2.1 (56.2%) had a statistically higher representation of men than Firefly 2 (42.9%) (<i>P</i> < 0.0001). With Firefly 2, 70.3% of characters had light skin tones, which was statistically lower (<i>P</i> < 0.0001) than for Stable Diffusion 2.1 (84.8%), Midjourney 5.2 (100%), and DALL-E 3 (94.8%). Overall, image quality metrics were average or better in 87.2% for DALL-E 3 and 86.2% for Midjourney 5.2, whereas 50.9% were inadequate or poor for Firefly 2 and 86.0% for Stable Diffusion 2.1. <b>Conclusion:</b> Generative AI text-to-image generation using DALL-E 3 via GPT-4 has the best overall quality compared with Firefly 2, Midjourney 5.2, and Stable Diffusion 2.1. Nonetheless, DALL-E 3 includes inherent biases associated with gender and ethnicity that demand more critical evaluation.</p>","PeriodicalId":16548,"journal":{"name":"Journal of nuclear medicine technology","volume":null,"pages":null},"PeriodicalIF":1.0,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142502305","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Anja Strok, Maja Dolenc Novak, Barbara Guzic Salobir, Monika Stalc, Katja Zaletel
{"title":"The Effect of Monitored Walking on Extracardiac Intestinal Activity in Myocardial Perfusion Imaging.","authors":"Anja Strok, Maja Dolenc Novak, Barbara Guzic Salobir, Monika Stalc, Katja Zaletel","doi":"10.2967/jnmt.124.267917","DOIUrl":"https://doi.org/10.2967/jnmt.124.267917","url":null,"abstract":"<p><p>Various techniques have been used in attempts to reduce interfering gastrointestinal activity in myocardial perfusion imaging (MPI); however, these approaches have yielded inconsistent results. The goal of this study was to investigate the efficacy of monitored walking, a previously unexplored technique, in reducing subdiaphragmatic activity-related artifacts during pharmacologic stress <sup>99m</sup>Tc-tetrofosmin MPI with SPECT to improve the overall image quality. <b>Methods:</b> The study included patients who underwent MPI with pharmacologic stress. They were given a step counter immediately after the radiotracer injection and were randomized into a group A, with a request to walk at least 1,000 steps before imaging, and a group B, with no specific instructions about walking. The reconstructed SPECT images were assessed visually. Moderate and severe levels of subdiaphragmatic tracer activity were considered relevant for the interpretation of the scans. Additionally, myocardial and abdominal activity was semiquantitatively assessed on raw planar images, and the mean myocardium-to-abdomen count ratios were calculated. <b>Results:</b> We enrolled 199 patients (95 patients in group A and 104 patients in group B). Clinical characteristics did not differ significantly between the 2 groups. Patients in group A walked more steps than patients in group B (<i>P</i> < 0.001), but there were no differences in the proportion of accepted scans between the 2 groups (<i>P</i> = 0.41). Additionally, there were no differences in the proportion of relevant subdiaphragmatic activity between the groups (<i>P</i> = 0.91). The number of steps did not impact the acceptance rate (<i>P</i> = 0.29). <b>Conclusion:</b> A higher number of steps walked during the waiting period between pharmacologic stress and acquisition does not affect subdiaphragmatic activity-related artifacts or the proportion of accepted scans after pharmacologic stress. However, pedometer use and clear instructions motivate patients to walk while awaiting imaging. Larger studies are required to compare a higher-step-count group with a sedentary control group to assess the influence of walking on gastrointestinal artifacts in MPI.</p>","PeriodicalId":16548,"journal":{"name":"Journal of nuclear medicine technology","volume":null,"pages":null},"PeriodicalIF":1.0,"publicationDate":"2024-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142289486","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Comment Regarding \"Vapocoolant Analgesia for Breast Lymphoscintigraphy: A Prospective Clinical Trial\".","authors":"Andrew Ditto","doi":"10.2967/jnmt.124.267498","DOIUrl":"https://doi.org/10.2967/jnmt.124.267498","url":null,"abstract":"","PeriodicalId":16548,"journal":{"name":"Journal of nuclear medicine technology","volume":null,"pages":null},"PeriodicalIF":1.0,"publicationDate":"2024-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142289483","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Navigating a Transatlantic Career Shift: Guidance for U.S. Nuclear Medicine Technologists Looking to Relocate to the U.K.","authors":"Jessica Settle","doi":"10.2967/jnmt.124.268651","DOIUrl":"https://doi.org/10.2967/jnmt.124.268651","url":null,"abstract":"","PeriodicalId":16548,"journal":{"name":"Journal of nuclear medicine technology","volume":null,"pages":null},"PeriodicalIF":1.0,"publicationDate":"2024-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142289484","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"So You Are a Clinical Instructor-Now What?","authors":"Jennifer L Prekeges","doi":"10.2967/jnmt.124.268054","DOIUrl":"https://doi.org/10.2967/jnmt.124.268054","url":null,"abstract":"<p><p>Many nuclear medicine technologists find themselves in the role of clinical instructor, often without much in the way of educational background. This article provides a few recommendations on how to get started in this role. After distinguishing between the roles of affiliate education supervisor and clinical instructor, the article discusses 2 basic tools: the clinical course learning outcomes and the student handbook. Expectations for students are reviewed. An important aspect of clinical instruction is the attitude of the instructor. Clinical instructors can motivate students or demotivate them, with this choice having a significant impact on the student's development. Overall, the desire and determination to be pleasant and helpful to students make the greatest difference in their development into nuclear medicine technologists.</p>","PeriodicalId":16548,"journal":{"name":"Journal of nuclear medicine technology","volume":null,"pages":null},"PeriodicalIF":1.0,"publicationDate":"2024-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142289485","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yung Hsiang Kao, Nadia Falzone, Michael Pearson, Dinesh Sivaratnam
{"title":"First-Strike Rapid Predictive Dosimetry and Dose Response for <sup>177</sup>Lu-PSMA Therapy in Metastatic Castration-Resistant Prostate Cancer.","authors":"Yung Hsiang Kao, Nadia Falzone, Michael Pearson, Dinesh Sivaratnam","doi":"10.2967/jnmt.123.267067","DOIUrl":"10.2967/jnmt.123.267067","url":null,"abstract":"<p><p>We devised and clinically validated a schema of rapid personalized predictive dosimetry for <sup>177</sup>Lu-PSMA-I&T in metastatic castration-resistant prostate cancer. It supersedes traditional empiric prescription by providing clinically meaningful predicted absorbed doses for first-strike optimization. <b>Methods:</b> Prostate-specific membrane antigen PET was conceptualized as a simulation study that captures the complex dosimetric interplay between tumor, marrow, and kidneys at a single time point. Radiation principles of fractionation, heterogeneity, normal-organ constraints (marrow, kidney), absorbed dose, and dose rate were introduced. We created a predictive calculator in the form of a free, open-source, and user-friendly spreadsheet that can be completed within minutes. Our schema achieves speed and accuracy by sampling tissue radioconcentrations (kBq/cm<sup>3</sup>) to be analyzed in conjunction with clinical input from the user that reflect dosimetric preconditions. The marrow-absorbed dose constraint was 0.217 Gy (dose rate, ≤0.0147 Gy/h) per fraction with an interfraction interval of at least 6 wk. <b>Results:</b> Our first 10 patients were analyzed. The first-strike mean tumor-absorbed dose threshold for any prostate-specific antigen (PSA) response was more than 10 Gy (dose rate, >0.1 Gy/h). The metastasis with the lowest first-strike tumor-absorbed dose correlated the best with the percentage decrease of PSA; its threshold to achieve hypothetical zero PSA was 20 Gy or more. Each patient's PSA doubling time can be used to personalize their unique absorbed dose-response threshold. The predicted mean first-strike prescription constrained by marrow-absorbed dose rate per fraction was 11.0 ± 4.0 GBq. Highly favorable conditions (tumor sink effect) were dosimetrically expressed as the combination of tumor-to-normal-organ ratios of more than 150 for marrow and more than 4 for kidney. Our schema obviates the traditional role of the SUV as a predictive parameter. <b>Conclusion:</b> Our rapid schema is feasible to implement in any busy real-world theranostics unit and exceeds today's best practice standards. Our dosimetric thresholds and predictive parameters can radiobiologically rationalize each patient's first-strike prescription down to a single becquerel. Favorable tumor-to-normal-organ ratios can be prospectively exploited by predictive dosimetry to optimize the first-strike prescription. The scientific framework of our schema may be applied to other systemic radionuclide therapies.</p>","PeriodicalId":16548,"journal":{"name":"Journal of nuclear medicine technology","volume":null,"pages":null},"PeriodicalIF":1.0,"publicationDate":"2024-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141432148","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Else A Aalbersberg, Tammie T Cao, Chelvi Mylvaganan-Young, Desiree Verwoerd, Kirsten Peen, Mariska Sonneborn-Bols, Jeroen J M A Hendrikx
{"title":"A Multiradionuclide Automatic Dispensing System for Syringes of Radiopharmaceuticals: The Effect on Operator Hand Dose.","authors":"Else A Aalbersberg, Tammie T Cao, Chelvi Mylvaganan-Young, Desiree Verwoerd, Kirsten Peen, Mariska Sonneborn-Bols, Jeroen J M A Hendrikx","doi":"10.2967/jnmt.124.267449","DOIUrl":"10.2967/jnmt.124.267449","url":null,"abstract":"<p><p>The radiation exposure of the hands of nuclear medicine laboratory technicians is largely due to the dispensing of radiopharmaceuticals into syringes. To reduce this exposure, a multiradionuclide automatic dispensing system (ADS) for syringes of radiopharmaceuticals was introduced. The aim of this study was to determine the effect of this ADS on hand dose compared with manual dispensing. <b>Methods:</b> The total hand dose per month for all personnel (12 technicians) was measured with ring dosimeters at the base of the index finger for 13 mo: 7 mo with manual syringe dispensing (radiopharmaceuticals containing <sup>99m</sup>Tc,<sup>18</sup>F, <sup>177</sup>Lu, <sup>68</sup>Ga, <sup>90</sup>Y, and <sup>223</sup>Ra) and 6 mo with ADS (automatic: radiopharmaceuticals containing <sup>18</sup>F and <sup>177</sup>Lu; manual: radiopharmaceuticals containing <sup>99m</sup>Tc, <sup>68</sup>Ga, <sup>90</sup>Y, and <sup>223</sup>Ra). <b>Results:</b> The mean total hand dose per month was reduced from 52.8 ± 10.2 mSv with manual dispensing to 21.9 ± 2.7 mSv with ADS (<i>P</i> < 0.001), which is an absolute decrease of 59%. Meanwhile, the total handled activity increased from 369 to 505 GBq (<i>P</i> < 0.001). <sup>18</sup>F-containing radiopharmaceuticals were the most commonly dispensed, at 182 GBq per month. The increase in total handled activity was largely due to an increase in <sup>177</sup>Lu (from 25 to 123 GBq), partially because of the introduction of [<sup>177</sup>Lu]Lu-PSMA-I&T. When correcting for this increase in handled activity, the hand dose was reduced by 69%. <b>Conclusion:</b> The introduction of a multiradionuclide syringe ADS decreased the hand dose to personnel by 69% when corrected for the increase in handled activity. Expanding the number of radiopharmaceuticals being dispensed by the system could potentially further decrease personnel hand dose.</p>","PeriodicalId":16548,"journal":{"name":"Journal of nuclear medicine technology","volume":null,"pages":null},"PeriodicalIF":1.0,"publicationDate":"2024-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141432145","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}