Nuclear Medicine and Molecular Imaging最新文献

{"title":"Dual-Tracer Parathyroid Imaging Using Joint SPECT Reconstruction.","authors":"Jaruwan Onwanna, Maythinee Chantadisai, Tawatchai Chaiwatanarat, Yothin Rakvongthai","doi":"10.1007/s13139-022-00787-x","DOIUrl":"10.1007/s13139-022-00787-x","url":null,"abstract":"<p><strong>Purpose: </strong>We assessed the lesion detection performance of the dual-tracer parathyroid SPECT imaging using the joint reconstruction method.</p><p><strong>Materials and methods: </strong>Thirty-six noise realizations were created from SPECT projections collected from an in-house neck phantom to emulate ^99mTc-pertechnetate/^99mTc-sestamibi parathyroid SPECT datasets. Difference images representing parathyroid lesions were reconstructed using the subtraction and the joint methods whose corresponding optimal iteration was defined as the iteration which maximized the channelized Hotelling observer signal-to-noise ratio (CHO-SNR). The joint method whose initial estimate was derived from the subtraction method at optimal iteration (the joint-AltInt method) was also assessed. In a study of 36 patients, a human-observer lesion-detection study was performed using difference images from the three methods at optimal iteration and the subtraction method with four iterations. The area under the receiver operating characteristic curve (AUC) was calculated for each method.</p><p><strong>Results: </strong>In the phantom study, both the joint-AltInt method and the joint method improved SNR compared to the subtraction method at their optimal iteration by 444% and 81%, respectively. In the patient study, the joint-AltInt method yielded the highest AUC of 0.73 as compared with 0.72, 0.71, and 0.64 from the joint method, the subtraction method at optimal iteration, and the subtraction method at four iterations. At a specificity of at least 0.70, the joint-AltInt method yielded significantly higher sensitivity than the other methods (0.60 vs 0.46, 042, and 0.42; <i>p</i> < 0.05).</p><p><strong>Conclusions: </strong>The joint reconstruction method yielded higher lesion detectability than the conventional method and holds promise for dual-tracer parathyroid SPECT imaging.</p>","PeriodicalId":19384,"journal":{"name":"Nuclear Medicine and Molecular Imaging","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10172461/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9846865","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimal Protocol and Clinical Usefulness of ¹²³I-MIBG Cardiac Scintigraphy for Differentiation of Parkinson's Disease and Dementia with Lewy Body from Non-Parkinson's Diseases.","authors":"In Kook Chun","doi":"10.1007/s13139-023-00790-w","DOIUrl":"10.1007/s13139-023-00790-w","url":null,"abstract":"<p><strong>Purpose: </strong>¹²³I-metaiodobenzylguanidine (MIBG) cardiac scintigraphy was a useful imaging modality for the diagnosis of Parkinson's disease, but its diagnostic performances were variably reported. This retrospective study compared the diagnostic performances and investigated the optimal imaging protocol of ¹²³I-MIBG cardiac scintigraphy at various imaging time points in patients suspected of Parkinson's disease in clinical practice.</p><p><strong>Methods: </strong>In patients suspected of Parkinson's disease, clinical records, autonomic function tests, and ¹²³I-MIBG cardiac scintigraphy were retrospectively reviewed. Semi-quantitative parameters such as heart-to-mediastinum ratio (HMR) and washout rate (WR) were calculated and compared at 15 min, 1 h, 2 h, 3 h, and 4 h post-injection (p.i.). of ¹²³I-MIBG cardiac scintigraphy. Group A consisted of Parkinson's disease (PD), Parkinson's disease dementia (PDD), and dementia with Lewy body (DLB), and group B consisted of non-Parkinson's diseases such as multiple system atrophy (MSA), progressive supranuclear palsy (PSP), drug-induced parkinsonism (DIP), essential tremor (ET), Parkinson-plus syndrome (PPS), and unspecified secondary parkinsonism (NA). The diagnostic performances of HMR and WR were compared for differentiation of group A from group B, and their clinical usefulness and optimal imaging time points were explored.</p><p><strong>Results: </strong>Seventy-eight patients were included in group A (67 PD, 7 PDD, 4 DLB), and 18 patients were included in group B (5 MSA, 3 PSP, 2 DIP, 2 ET, 1 PPS, and 1 NA). Sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value of HMR and WR were maximized at 4 h p.i., (82.1%, 85.7%, 82.6%, 97.0%, and 46.2%; cutoff threshold < 1.717; area under curve 0.8086) and at the time interval between 1 and 4 h p.i. (65.4%, 85.7%, 68.5%, 96.2%, and 30.8%; cutoff threshold > 24.1%; area under curve 0.8246), respectively, and PPVs of both HMR and WR persistently showed greater than 92.7% at earlier time points and shorter time intervals.</p><p><strong>Conclusion: </strong>This study reassured that 4-h-delayed imaging is recommended for the best diagnostic performances in ¹²³I-MIBG cardiac scintigraphy. Although it showed suboptimal diagnostic performances to differentiate PD, PDD, and DLB from non-Parkinson's diseases, it can be useful as an auxiliary measure for the differential diagnosis in usual clinical practice.</p><p><strong>Supplementary information: </strong>The online version contains supplementary material available at 10.1007/s13139-023-00790-w.</p>","PeriodicalId":19384,"journal":{"name":"Nuclear Medicine and Molecular Imaging","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10172433/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9829695","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Tc-99 m Ubiquicidin Imaging in Orbital Aspergilloma: an Illustration.","authors":"Meivel Angamuthu, Nishikant Damle, Dikhra Khan, Rachna Meel, Sanjay Sharma, Chandrasekhar Bal","doi":"10.1007/s13139-022-00784-0","DOIUrl":"10.1007/s13139-022-00784-0","url":null,"abstract":"<p><p>Aspergillus infection is relatively rare disease, and we present a case of orbital aspergillus infection who presented with right orbital pain and swelling. Right orbital lesion was identified on CT, MRI, and PET-CT imaging followed by confirmation of aspergillus on histopathological examination. We demonstrate that Tc-99 m ubiquicidin scan can yield positive results in aspergillosis too, enabling its differentiation from non-infective pathologies.</p>","PeriodicalId":19384,"journal":{"name":"Nuclear Medicine and Molecular Imaging","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10172414/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9530376","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Diagnosis of Fungal Infection (<i>Candida albicans</i>) After Heart Transplantation in a Pediatric Case with Fever of Unknown Origin: Role of ^99mTc-UBI SPECT/CT and ¹⁸F-FDG PET/CT.","authors":"Hadi Malek, Raheleh Hedayati, Mahdi Maghsudi, Nahid Yaghoobi","doi":"10.1007/s13139-022-00781-3","DOIUrl":"10.1007/s13139-022-00781-3","url":null,"abstract":"<p><p>The diagnosis of patients with fever of unknown origin (FUO) in pediatric heart transplantation is a challenging medical problem. The physician should differentiate between rejections, infections, malignancy, adrenal insufficiency, and drug fever. Immunosuppressive therapy in these patients exposes them to a high risk of developing a post-transplantation fungal infection. In this case, we discuss the diagnostic contribution of the ^99mTc-UBI scan and ¹⁸F-FDG PET scan for diagnosis of fungal infection causing FUO in these patients.</p>","PeriodicalId":19384,"journal":{"name":"Nuclear Medicine and Molecular Imaging","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10172439/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9829694","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Presence of Residual Vascular and Adipose Tissue Inflammation on ¹⁸F-FDG PET in Patients with Chronic Coronary Artery Disease.","authors":"Sini Toivonen,&nbsp;Miia Lehtinen,&nbsp;Peter Raivio,&nbsp;Juha Sinisalo,&nbsp;Antti Loimaala,&nbsp;Valtteri Uusitalo","doi":"10.1007/s13139-022-00785-z","DOIUrl":"https://doi.org/10.1007/s13139-022-00785-z","url":null,"abstract":"<p><strong>Purpose: </strong>We evaluated the residual vascular and adipose tissue inflammation in patients with chronic coronary artery disease (CAD) using positron emission tomography (PET).</p><p><strong>Methods: </strong>Our study population consisted of 98 patients with known CAD and 94 control subjects who had undergone ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) PET due to non-cardiac reasons. Aortic root and vena cava superior ¹⁸F-FDG uptake were measured to obtain the aortic root target-to-background ratio (TBR). In addition, adipose tissue PET measurements were done in pericoronary, epicardial, subcutaneous, and thoracic adipose tissue. Adipose tissue TBR was calculated using the left atrium as a reference region. Data are presented as mean ± standard deviation or as median (interquartile range).</p><p><strong>Results: </strong>The aortic root TBR was higher in CAD patients compared to control subjects, 1.68 (1.55-1.81) vs. 1.53 (1.43-1.64), <i>p</i> < 0.001. Subcutaneous adipose tissue uptake was elevated in CAD patients 0.30 (0.24-0.35) vs. 0.27 (0.23-0.31), <i>p</i> < 0.001. Metabolic activity of CAD patients and control subjects was comparable in the pericoronary (0.81 ± 0.18 vs. 0.80 ± 0.16, <i>p</i> = 0.59), epicardial (0.53 ± 0.21 vs. 0.51 ± 0.18, <i>p</i> = 0.38) and thoracic (0.31 ± 0.12 vs. 0.28 ± 0.12, <i>p</i> = 0.21) adipose tissue regions. Aortic root or adipose tissue ¹⁸F-FDG uptake was not associated with the common CAD risk factors, coronary calcium score, or aortic calcium score (<i>p</i> value > 0.05).</p><p><strong>Conclusion: </strong>Patients with a chronic CAD had a higher aortic root and subcutaneous adipose tissue ¹⁸F-FDG uptake compared to control patients, which suggests residual inflammatory risk.</p>","PeriodicalId":19384,"journal":{"name":"Nuclear Medicine and Molecular Imaging","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10172407/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"10296944","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Radionuclide Therapy Videos on YouTube as An Educational Material: Has the COVID-19 Pandemic Changed the Quality, Usefulness, and Interaction Features.","authors":"Ulku Korkmaz,&nbsp;Selin Soyluoglu,&nbsp;Ersan Arda","doi":"10.1007/s13139-023-00799-1","DOIUrl":"https://doi.org/10.1007/s13139-023-00799-1","url":null,"abstract":"<p><strong>Introduction: </strong>Current treatment approach aims to achieve greater efficacy with fewer side effects, by targeted cancer therapy as much as possible. Radionuclide therapy is a modality that uses cancer theranostics and is increasingly applied for various cancers as a targeted therapy. YouTube is a preferred tool for obtaining medical information from the internet. This study aims to determine the content quality, level of interaction and usefulness as education material of radionuclide therapy YouTube videos and to reveal the impact of the COVID-19 process on these parameters.</p><p><strong>Materials and methods: </strong>The keywords were searched on YouTube on August 25, 2018, and May 10, 2021. After removing duplicate and excluded videos, all remaining videos were scored and coded.</p><p><strong>Results: </strong>Majority of the videos were useful educational material. Most of them were high quality. Popularity markers were unrelated to quality level. After COVID, the power index of videos with high JAMA scores increased. The COVID-19 pandemic did not have a negative effect on video features; the quality of the content increased even more after the pandemic.</p><p><strong>Conclusion: </strong>Radionuclide therapy YouTube videos have high-quality content and provide useful education material. The popularity is independent of the content quality. During the pandemic, video quality and usefulness characteristics did not change, while the visibility is increased. We consider YouTube to be an appropriate educational material for patients and healthcare professionals to gain basic knowledge of radionuclide therapy. The Covıd-19 pandemic highlighted the power of radionuclide therapy YouTube videos as an educational material.</p>","PeriodicalId":19384,"journal":{"name":"Nuclear Medicine and Molecular Imaging","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2023-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10073796/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9858223","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Applying Pix2pix to Translate Hyperemia in Blood Pool Image into Corresponding Increased Bone Uptake in Delayed Image in Three-Phase Bone Scintigraphy.","authors":"Ki Seong Park, Jang Bae Moon, Sang-Geon Cho, Jahae Kim, Ho-Chun Song","doi":"10.1007/s13139-022-00786-y","DOIUrl":"10.1007/s13139-022-00786-y","url":null,"abstract":"<p><strong>Purpose: </strong>Delayed images may not be acquired due to severe pain, drowsiness, or worsening vital signs while waiting after blood pool imaging in three-phase bone scintigraphy. If the hyperemia in the blood pool image contains information from which increased uptake on the delayed images can be inferred, the generative adversarial network (GAN) can generate the increased uptake from the hyperemia. We attempted to apply pix2pix, a type of conditional GAN, to transform hyperemia into increased bone uptake.</p><p><strong>Methods: </strong>We enrolled 1464 patients who underwent three-phase bone scintigraphy for inflammatory arthritis, osteomyelitis, complex regional pain syndrome (CRPS), cellulitis, and recent bone injury. Blood pool images were acquired 10 min after intravenous injection of Tc-99 m hydroxymethylene diphosphonate, and delayed bone images were obtained after 3 h. The model was based on the open-source code of the pix2pix model with perceptual loss. Increased uptake in the delayed images generated by the model was evaluated using lesion-based analysis by a nuclear radiologist in areas consistent with hyperemia in the blood pool images.</p><p><strong>Results: </strong>The model showed sensitivities of 77.8% and 87.5% for inflammatory arthritis and CRPS, respectively. In osteomyelitis and cellulitis, their sensitivities of about 44% were observed. However, in cases of recent bone injury, the sensitivity was only 6.3% in areas consistent with focal hyperemia.</p><p><strong>Conclusion: </strong>The model based on pix2pix generated increased uptake in delayed images matching the hyperemia in the blood pool image in inflammatory arthritis and CRPS.</p>","PeriodicalId":19384,"journal":{"name":"Nuclear Medicine and Molecular Imaging","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2023-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10043061/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9225979","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MR Template-Based Individual Brain PET Volumes-of-Interest Generation Neither Using MR nor Using Spatial Normalization.","authors":"Seung Yeon Seo, Jungsu S Oh, Jinwha Chung, Seog-Young Kim, Jae Seung Kim","doi":"10.1007/s13139-022-00772-4","DOIUrl":"10.1007/s13139-022-00772-4","url":null,"abstract":"<p><p>For more anatomically precise quantitation of mouse brain PET, spatial normalization (SN) of PET onto MR template and subsequent template volumes-of-interest (VOIs)-based analysis are commonly used. Although this leads to dependency on the corresponding MR and the process of SN, routine preclinical/clinical PET images cannot always afford corresponding MR and relevant VOIs. To resolve this issue, we propose a deep learning (DL)-based individual-brain-specific VOIs (i.e., cortex, hippocampus, striatum, thalamus, and cerebellum) directly generated from PET images using the inverse-spatial-normalization (iSN)-based VOI labels and deep convolutional neural network model (deep CNN). Our technique was applied to mutated amyloid precursor protein and presenilin-1 mouse model of Alzheimer's disease. Eighteen mice underwent T2-weighted MRI and ¹⁸F FDG PET scans before and after the administration of human immunoglobin or antibody-based treatments. To train the CNN, PET images were used as inputs and MR iSN-based target VOIs as labels. Our devised methods achieved decent performance in terms of not only VOI agreements (i.e., Dice similarity coefficient) but also the correlation of mean counts and SUVR, and CNN-based VOIs was highly accordant with ground-truth (the corresponding MR and MR template-based VOIs). Moreover, the performance metrics were comparable to that of VOI generated by MR-based deep CNN. In conclusion, we established a novel quantitative analysis method both MR-less and SN-less fashion to generate individual brain space VOIs using MR template-based VOIs for PET image quantification.</p><p><strong>Supplementary information: </strong>The online version contains supplementary material available at 10.1007/s13139-022-00772-4.</p>","PeriodicalId":19384,"journal":{"name":"Nuclear Medicine and Molecular Imaging","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2023-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10043100/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9225977","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Artificial Intelligence Increases the Agreement among Physicians Classifying Focal Skeleton/Bone Marrow Uptake in Hodgkin's Lymphoma Patients Staged with [¹⁸F]FDG PET/CT-a Retrospective Study.","authors":"May Sadik,&nbsp;Jesús López-Urdaneta,&nbsp;Johannes Ulén,&nbsp;Olof Enqvist,&nbsp;Per-Ola Andersson,&nbsp;Rajender Kumar,&nbsp;Elin Trägårdh","doi":"10.1007/s13139-022-00765-3","DOIUrl":"https://doi.org/10.1007/s13139-022-00765-3","url":null,"abstract":"<p><strong>Purpose: </strong>Classification of focal skeleton/bone marrow uptake (BMU) can be challenging. The aim is to investigate whether an artificial intelligence-based method (AI), which highlights suspicious focal BMU, increases interobserver agreement among a group of physicians from different hospitals classifying Hodgkin's lymphoma (HL) patients staged with [¹⁸F]FDG PET/CT.</p><p><strong>Methods: </strong>Forty-eight patients staged with [¹⁸F]FDG PET/CT at Sahlgenska University Hospital between 2017 and 2018 were reviewed twice, 6 months apart, regarding focal BMU. During the second time review, the 10 physicians also had access to AI-based advice regarding focal BMU.</p><p><strong>Results: </strong>Each physician's classifications were pairwise compared with the classifications made by all the other physicians, resulting in 45 unique pairs of comparisons both without and with AI advice. The agreement between the physicians increased significantly when AI advice was available, which was measured as an increase in mean Kappa values from 0.51 (range 0.25-0.80) without AI advice to 0.61 (range 0.19-0.94) with AI advice (<i>p</i> = 0.005). The majority of the physicians agreed with the AI-based method in 40 (83%) of the 48 cases.</p><p><strong>Conclusion: </strong>An AI-based method significantly increases interobserver agreement among physicians working at different hospitals by highlighting suspicious focal BMU in HL patients staged with [¹⁸F]FDG PET/CT.</p>","PeriodicalId":19384,"journal":{"name":"Nuclear Medicine and Molecular Imaging","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2023-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10043120/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9225973","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Automatic Lung Cancer Segmentation in [¹⁸F]FDG PET/CT Using a Two-Stage Deep Learning Approach.","authors":"Junyoung Park, Seung Kwan Kang, Donghwi Hwang, Hongyoon Choi, Seunggyun Ha, Jong Mo Seo, Jae Seon Eo, Jae Sung Lee","doi":"10.1007/s13139-022-00745-7","DOIUrl":"10.1007/s13139-022-00745-7","url":null,"abstract":"<p><strong>Purpose: </strong>Since accurate lung cancer segmentation is required to determine the functional volume of a tumor in [¹⁸F]FDG PET/CT, we propose a two-stage U-Net architecture to enhance the performance of lung cancer segmentation using [¹⁸F]FDG PET/CT.</p><p><strong>Methods: </strong>The whole-body [¹⁸F]FDG PET/CT scan data of 887 patients with lung cancer were retrospectively used for network training and evaluation. The ground-truth tumor volume of interest was drawn using the LifeX software. The dataset was randomly partitioned into training, validation, and test sets. Among the 887 PET/CT and VOI datasets, 730 were used to train the proposed models, 81 were used as the validation set, and the remaining 76 were used to evaluate the model. In Stage 1, the global U-net receives 3D PET/CT volume as input and extracts the preliminary tumor area, generating a 3D binary volume as output. In Stage 2, the regional U-net receives eight consecutive PET/CT slices around the slice selected by the Global U-net in Stage 1 and generates a 2D binary image as the output.</p><p><strong>Results: </strong>The proposed two-stage U-Net architecture outperformed the conventional one-stage 3D U-Net in primary lung cancer segmentation. The two-stage U-Net model successfully predicted the detailed margin of the tumors, which was determined by manually drawing spherical VOIs and applying an adaptive threshold. Quantitative analysis using the Dice similarity coefficient confirmed the advantages of the two-stage U-Net.</p><p><strong>Conclusion: </strong>The proposed method will be useful for reducing the time and effort required for accurate lung cancer segmentation in [¹⁸F]FDG PET/CT.</p>","PeriodicalId":19384,"journal":{"name":"Nuclear Medicine and Molecular Imaging","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2023-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10043063/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"9230524","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}