Nuclear Medicine and Molecular Imaging最新文献

A 74-year-old male with lung cancer in right lower lobe underwent lung perfusion planar imaging and SPECT/CT scan for preoperative evaluation. A focal intense uptake artifact was shown on the initial scan, prompting a repeat scan 5 days later. A MAA embolus artifact can disrupt the analysis of radioactivity in each lung segment, potentially causing errors in postoperative residual lung function. Consequently, repeat imaging or correction may be necessary. When repeat images without MAA embolus were used for comparison, applying a mask to exclude the MAA embolus lesion provided reliable results, eliminating the need for further repeat imaging.

Purpose: The Korean Society of Nuclear Medicine (KSNM) established the first diagnostic reference levels (DRLs) in Korea in 2019. Over the past five years, the need to update the DRLs has emerged for adult nuclear medicine imaging in Korea. The KSNM therefore conducted a nationwide survey and updated the DRLs for adult nuclear medicine examinations.

Methods: A national DRL survey was conducted from March 2024 to April 2024 to investigate the administered activities of radiopharmaceuticals and the radiation doses from hybrid CT associated with PET/CT and SPECT/CT. Compared with the 2019 DRLs, the updated DRLs reflect newly approved radiopharmaceuticals in Korea, the introduction of SiPM-based PET/CT scanners, and expanded data on hybrid CT radiation exposure. The DRL survey team performed data collection and cleaning, while the DRL Task Force under the Radiation Safety Committee reviewed the draft. The 2024 DRLs were subsequently approved by the Radiation Safety Committee.

Results: The current updated DRLs include 41 nuclear medicine imaging studies and nine hybrid CT studies. For PET radiopharmaceuticals, new DRLs were established for ¹⁸F-FDG in the purpose of myocardial viability assessment, infection/inflammation, and screening. Additionally, DRLs for ¹⁸F-Florbetaben, ¹⁸F-Flutemetamol and ¹⁸F-Florapronol for brain amyloid imaging were newly introduced. For SPECT radiopharmaceuticals, new DRLs were established for cardiac scintigraphy using ^99mTc-DPD, HDP, or PYP, and for cardiac sympathetic nerve imaging with ¹²³I-MIBG. The DRLs for hybrid CT doses (three PET/CT and six SPECT/CT studies) represent a novel addition in the 2024 nationwide survey. Furthermore, administered doses of radiopharmaceuticals and hybrid CT radiation doses were compared between conventional and SiPM-based PET/CT scanners.

Conclusions: The DRLs for nuclear medicine imaging in Korean were updated. The 2024 DRLs will support the optimization of medical exposure to ionizing radiation in nuclear medicine imaging practice in Korea.

Purpose: Ubiquicidin (UBI) is a natural antimicrobial peptide that has shown potential for targeting microbial pathogens in both in-vitro and human studies. Infections are common in patients with cancer and can mimic the appearance of primary neoplasms or metastatic lesions on Fluorine-18 fluorodeoxyglucose ([¹⁸F]FDG) positron emission tomography-computed tomography (PET-CT) imaging. This pilot study aimed to determine the diagnostic accuracy of Gallium-68-1,4,7-triazacyclononane-1,4,7 triacetic acid ([⁶⁸Ga]GaNOTA)-UBI PET-CT for differentiating infective from neoplastic lesions.

Methods: This was a prospective diagnostic accuracy study. Consecutive patients presenting with lesions identified on [¹⁸F]FDG PET-CT scans, which were equivocal for diagnosing malignancy and focal infection, were eligible for inclusion in the study. 30 selected patients who gave written informed consent underwent [⁶⁸Ga]GaNOTA-UBI PET-CT for lesion characterization. The lesions were considered positive for infection if the uptake of [⁶⁸Ga]GaNOTA-UBI in the lesion was higher than in the mediastinal blood pool. Biopsy and microbiological assay from the lesions and/or clinicoradiologic follow-up were the reference standard. Measures of diagnostic accuracy of [⁶⁸Ga]GaNOTA-UBI PET-CT were computed based on patient-based analysis. The statistical analysis was performed with the statistical package SPSS Version 27.0 (Armonk, NY, USA: IBM Corp). A 2-tailed P value of < 0.05 was considered statistically significant.

Results: According to the reference standard 21/30 (70%) patients were eventually diagnosed with infective lesions, 8/30 (26.7%) with neoplastic lesions and one patient with sterile inflammation. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and accuracy of [⁶⁸Ga]GaNOTA-UBI PET-CT were calculated as 52%, 78%, 85%, 41% and 60%, respectively. The maximum standardized uptake value (SUVmax) of infective lesions was not significantly different from malignant lesions; however, the mean of the ratios of lesional SUVmax to mediastinal blood pool SUVmax (SUVratio) of the former was significantly higher than that of latter (p = 0.042). The area under the curve (AUC) for SUVratio as a diagnostic parameter was 0.688 at an optimal cutoff value of 0.74. Good interrater agreement for the interpretation of [⁶⁸Ga]GaNOTA-UBI PET-CT scans was achieved (Cohen's κ = 0.73).

Conclusion: [⁶⁸Ga]GaNOTA-UBI PET-CT shows only a moderate diagnostic accuracy for distinguishing infective from neoplastic lesions in oncology patients. A higher PPV however indicates its potential applicability as a confirmatory test for infection in patients with a higher clinical probability for the latter.

The Korean Society of Nuclear Medicine (KSNM) was established in 1961 to promote clinical and technical advancements in nuclear medicine in Republic of Korea, and is composed mainly of nuclear medicine physicians and related scientists. The Korean Society of Lymphedema (KSL) was founded in 2010, gathering physicians, scientists, physical therapists and nurses from various fields to perform basic and clinical research and to study clinical and basic aspects including its pathophysiology, prevention, diagnosis, and treatment related to lymphedema. The KSNM and the KSL have agreed to collaborate on the establishment of a guideline for the implementation and interpretation of lymphoscintigraphy used for the diagnosis and evaluation of lymphedema, and they decided to periodically revise these guidelines to improve the quality of care for lymphedema patients. The purpose of this guideline was to help healthcare providers in taking reasonable actions based on current medical knowledge, available resources, and patient needs while ensuring effective and safe exams, and accurate interpretations in the administration of lymphoscintigraphy for the diagnosis and evaluation of lymphedema.

Supplementary information: The online version contains supplementary material available at 10.1007/s13139-025-00934-0.

An early, biomarker-based diagnosis of Alzheimer's Disease (AD) is crucial, especially with the emerging availability of novel therapeutic options. However, the role of ¹⁸F-FDG-PET and its relationship to other PET and CSF biomarkers remains unclear. Therefore, the aim of this study was the evaluation of the role of ¹⁸F-FDG-PET in AD diagnosis and its relationship to other commonly used fluid and PET biomarkers and their individual and multimodal accuracy in AD diagnosis. We included n = 157 AD patients, n = 603 MCI patients, and n = 380 cognitively normal participants from the Alzheimer's Disease Neuroimaging Initiative (ADNI) that underwent PET imaging with ¹⁸F-FDG or ¹⁸F-Florbetapir. Clinical and imaging data including patient characteristics, CSF biomarkers, cognition tests, ¹⁸F-FDG-PET, ¹⁸F-Florbetapir-PET, and ¹⁸F-Flortaucipir-PET were retrospectively analyzed. PET images were quantified in several brain regions. The uptake of ¹⁸F-FDG was inversely correlated with ¹⁸F-Florbetapir and positively correlated with CSF Aβ42 in several brain regions commonly affected by AD. Additionally, ¹⁸F-FDG uptake showed an inverse correlation with both forms of CSF tau, t-tau and p-tau, in various brain regions, but did not correlate with ¹⁸F-Flortaucipir uptake. Moreover, regional ¹⁸F-FDG uptake was positively correlated with cognitive function. Diagnostic accuracies were similarly high for ¹⁸F-FDG uptake in the PCC/Precuneus region, ¹⁸F-Florbetapir uptake, CSF Aβ42, CSF p-tau, and ¹⁸F-Flortaucipir uptake in differentiating AD from cognitively normal individuals. ¹⁸F-FDG-PET and its combination with CSF p-tau/ Aβ42 ratio showed the highest predictive power for disease severity. The study underscores the potential of integrating ¹⁸F-FDG-PET with CSF biomarkers to enhance the diagnosis, prognosis, and monitoring of AD, highlighting the complexity and regional specificity of biomarker interactions in neurodegeneration.

Supplementary information: The online version contains supplementary material available at 10.1007/s13139-025-00932-2.

[This corrects the article DOI: 10.1007/s13139-025-00914-4.].

Purpose: This study evaluated transfer learning in classifying skeletal metastases on bone scintigraphy. The primary objective was to assess its performance in detecting skeletal metastases, while the secondary objective was to compare its performance to human readers.

Methods: A total of 2,510 patients with known malignancies were included - 2,368 retrospectively recruited and 142 prospectively enrolled. Scans were categorized as normal, benign-degenerative, or metastasis based on clinical consensus, follow-up, biopsy, radiology, or SPECT/CT findings. The retrospective data were randomly divided into training (1,895) and validation (473) sets, while the prospective cohort served as an independent testing set. Google's InceptionV3 was used for image embedding, and 13 supervised ML algorithms were tested. The Log Loss value of a random classifier was used to select the optimal models for testing, while Stuart-Maxwell test compared models' performance to human readers.

Results: Eight ML models with Log Loss value less than that of the random classifier achieved AUCs > 0.900 on training and validation, with all but one (Support Vector Machine) maintaining AUCs > 0.900 on testing. Logistic Regression performed best (≥ 0.993 in all metrics), while Neural Networks, Gradient Boosting, and Random Forest also demonstrated robust performance (≥ 0.817 in all metrics). Notably, ML models interpreted 142 images in 0.027-1.770 s compared to 10.07-18.00 min by human readers; less experienced readers performed significantly worse (P ≤ 0.002) than the models, whereas experienced reader's performance was comparable (P ≥ 0.280).

Conclusion: Transfer learning demonstrates commendable performance in classifying skeletal metastases on bone scintigraphy, outperforming less experienced readers while matching experienced reader's performance.

Supplementary information: The online version contains supplementary material available at 10.1007/s13139-025-00927-z.

The Korean Society of Nuclear Medicine (KSNM) was founded in 1961 to promote the clinical and technical advancement of nuclear medicine in Republic of Korea. It comprises approximately 600 members, mainly nuclear medicine physicians and related scientists. The KSNM periodically updates guidelines to advance nuclear medicine and help medical professionals provide better patient care. These guidelines are flexible and not obligatory. The KSNM states that these guidelines should not be used in legal actions challenging the medical decisions of healthcare professionals. Final medical decisions should be made by nuclear medicine physicians based on individual patient conditions, available resources, the latest medical knowledge, and technological advances. Deviations from these guidelines does not necessarily indicate substandard care, but rather reflects the application of reasonable clinical judgment. Detailed quality control is to follow the KSNM's quality control guidelines. Due to the diversity of patients and complexity of medical cases, adherence to guidelines does not always guarantee accurate diagnoses or successful outcomes. This guideline aims to revise the 2013 '¹⁸F-fluorodeoxyglucose Positron Emission Tomography/Computed Tomography for Oncological Patients: Procedural Guideline by the KSNM', ensuring medical professionals take appropriate actions based on current medical knowledge, available resources, and patient needs, ultimately achieving effective and safe imaging. Clinical Trial Number: Not applicable.