Ashwin Singh Parihar

The imaging arsenal for nuclear medicine physicians continues to grow rapidly with multiple radiopharmaceuticals now being used routinely in clinical practice. More specifically, PET imaging targeting various aspects of tumor biology has demonstrated utility in diagnosing incidental/synchronous malignancies as well as biochemical profiling of tumors for prognostication (ie, in the setting of prostate and neuro-endocrine cancers). These concepts are illustrated in this case of an incidental gastric gastrointestinal stromal tumor (GIST) which was imaged with 3 separate PET radiopharmaceuticals.

Higher liver background uptake was observed on prostate-specific membrane antigen (PSMA)–targeted PET imaging with fluorine 18 ( ¹⁸ F)-flotufolastat and ¹⁸ F-PSMA-1007 compared with ¹⁸ F-piflufolastat and gallium 68 ( ⁶⁸ Ga)-PSMA-11, which may influence patient eligibility for PSMA-targeted therapy.

Prostate-specific membrane antigen (PSMA) targeting PET imaging has an established role in prostate carcinoma for staging, recurrence evaluation, and selection for PSMA-targeted radiopharmaceutical therapy. We report a case with PSMA avid prostate, hepatic, and osseous lesions in a patient with synchronous malignancies complicating management.

Synthetic radiographs, including OpenAI’s ChatGPT-generated ones, are not easily distinguishable from authentic radiographs by either radiologists or large language models; physician training and dedicated tools are crucial to mitigate risks.

Prostate-specific membrane antigen (PSMA)-targeted PET imaging plays a well-established role in staging, detecting recurrence, and selecting patients for PSMA-targeted radiopharmaceutical therapy in prostate cancer. We present a case of PSMA-expressing recurrent prostate cancer in the pelvis, involving the urinary bladder wall and pelvic lymph nodes that was not completely visualized on ¹⁸ F-DCFPyL PET/CT, but clearly delineated on ¹⁸ F-rhPSMA-7.3 PET/CT. This discrepancy is attributed to the significantly lower urinary excretion and consequent low urinary bladder activity with ¹⁸ F-rhPSMA, yielding high target-to-background ratios in the pelvis, and highlighting its potential diagnostic advantage in evaluating pelvic lesions near the urinary bladder.

DeepSeek is a newly introduced large language model (LLM) designed for enhanced reasoning, but its medical-domain capabilities have not yet been evaluated. This study assessed the capabilities of three LLMs— DeepSeek-R1, ChatGPT-o1, and Llama 3.1-405B—in performing four different medical tasks: answering questions from the United States Medical Licensing Examination (USMLE), interpreting and reasoning based on text-based diagnostic and management cases, providing tumor classification according to RECIST 1.1 criteria, and providing summaries of diagnostic imaging reports across multiple modalities. In the USMLE test, the performance of DeepSeek-R1(accuracy=0.92) was slightly inferior to that of ChatGPT-o1(accuracy=0.95; p = 0.04) but better than that of Llama 3.1-405B (accuracy=0.83; p < 10^-3). For text-based case challenges, DeepSeek-R1 performed similarly to ChatGPT-o1 (accuracy of 0.57 vs 0.55; p = 0.76 and 0.74 vs 0.76; p = 0.06, using New England Journal of Medicine and Medicilline databases, respectively). For RECIST classifications, DeepSeek-R1 also performed similarly to ChatGPT-o1 (0.73 vs 0.81; p = 0.10). Diagnostic reasoning steps provided by DeepSeek were deemed more accurate than those provided by ChatGPT and Llama 3.1-405B (average Likert score of 3.61, 3.22, and 3.13, respectively, p = 0.005 and p < 10⁻³). However, summarized imaging reports provided by DeepSeek-R1 exhibited lower global quality than those provided by ChatGPT-o1 (5-point Likert score: 4.5 vs 4.8; p < 10⁻³). This study highlights the potential of DeepSeek-R1 LLM for medical applications but also underlines areas needing improvements.

^99mTc-Technegas was recently approved by the U.S. Food and Drug Administration as a radiopharmaceutical for ventilation scintigraphy. However, there are currently no data comparing the quantification of relative lung ventilation with ^99mTc-Technegas with that performed using the standard approach with inhaled ¹³³Xe. Methods: We performed a secondary analysis of data from prospectively recruited participants in a phase 3 trial undergoing evaluation for lung transplantation who received both ¹³³Xe and ^99mTc-Technegas ventilation imaging. The ¹³³Xe and ^99mTc-Technegas images were analyzed asynchronously using semiautomatic segmentation to extract the relative lung ventilation percentages. The anterior and posterior ^99mTc-Technegas images were analyzed to derive 3 sets of relative ventilation percentages (posterior, anterior, and geometric mean data) and compared with the values from posterior ¹³³Xe images. We evaluated for correlation and agreement between the relative lung ventilation percentages obtained using these 2 radiopharmaceuticals. Results: In a cohort of 74 participants, we found a strong positive correlation in the relative lung ventilation quantified using ¹³³Xe with that using ^99mTc-Technegas. A high level of agreement was demonstrated on the Bland–Altman plot comparing the 2 imaging modalities. Seventy-two of 74 participants (97.3%) had their relative ventilation percentage measurements within ±15% for ¹³³Xe and ^99mTc-Technegas. The differences in relative ventilation measurements were within the 95% CI limits of the mean for 70 of 74 participants (94.6%) and within a narrower ±10% threshold for 68 of 74 participants (91.9%), again reflecting the comparability of the 2 techniques. The strongest correlation coefficient (r = 0.79) was observed between the relative ventilation percentages obtained from ¹³³Xe and posterior ^99mTc-Technegas images. The geometric mean method had a slightly lower but still comparable correlation (r = 0.77), and as expected, the correlation with the anterior ^99mTc-Technegas images was worst (r = 0.70). Conclusion: We showed a strong positive correlation and high agreement between the relative lung ventilation percentages obtained using ¹³³Xe and ^99mTc-Technegas. These data provide important clinical evidence supporting the use of ^99mTc-Technegas for quantification of relative lung ventilation. Visual Abstract

Purpose of the Review Cardiac involvement in systemic sarcoidosis or isolated cardiac sarcoidosis plays a pivotal role in the clinical manifestation and prognostication. Active-inflammatory cardiac sarcoidosis is associated with a regional impairment of coronary microvascular function that may confer further detrimental effects on myocardial function needing further characterization. Recent Findings Clinical investigations with cardiac positron emission tomography/computed tomography in conjunction with ¹⁸F-fluorodeoxyglucose to determine myocardial inflammation and ¹³N-ammonia to quantify myocardial blood flow (MBF) in patients with known or suspected cardiac sarcoidosis outlined that sarcoidosis-induced myocardial inflammation was associated with adverse effects on corresponding regional coronary microvascular function. Notably, immune-suppressive treatment caused reductions in myocardial inflammation were paralleled by improvements of coronary microvascular dysfunction outlining direct adverse effect of inflammation on coronary arteriolar function. Summary This review summarizes contributions of cardiac PET imaging in the identification and characterization of active-inflammatory cardiac sarcoidosis, its effect on coronary microvascular function, treatment responses, and prognostic implications.