medRxiv - Radiology and Imaging最新文献

{"title":"Diagnostic Accuracy of Vision-Language Models on Japanese Diagnostic Radiology, Nuclear Medicine, and Interventional Radiology Specialty Board Examinations","authors":"Tatsushi Oura, Hiroyuki Tatekawa, Daisuke Horiuchi, Shu Matsushita, Hirotaka Takita, Natsuko Atsukawa, Yasuhito Mitsuyama, Atsushi Yoshida, Kazuki Murai, Rikako Tanaka, Taro Shimono, Akira Yamamoto, Yukio Miki, Daiju Ueda","doi":"10.1101/2024.05.31.24308072","DOIUrl":"https://doi.org/10.1101/2024.05.31.24308072","url":null,"abstract":"<strong>Purpose</strong> The performance of vision-language models (VLMs) with image interpretation capabilities, such as GPT-4 omni (GPT-4o), GPT-4 vision (GPT-4V), and Claude-3, has not been compared and remains unexplored in specialized radiological fields, including nuclear medicine and interventional radiology. This study aimed to evaluate and compare the diagnostic accuracy of various VLMs, including GPT-4 + GPT-4V, GPT-4o, Claude-3 Sonnet, and Claude-3 Opus, using Japanese diagnostic radiology, nuclear medicine, and interventional radiology (JDR, JNM, and JIR, respectively) board certification tests.","PeriodicalId":501358,"journal":{"name":"medRxiv - Radiology and Imaging","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141253548","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":"Enhancing the Diagnostic Utility of ASL Imaging in Temporal Lobe Epilepsy through FlowGAN: An ASL to PET Image Translation Framework","authors":"Alfredo Lucas, Chetan Vadali, Sofia Mouchtaris, T. Campbell Arnold, James J Gugger, Catherine V. Kulick-Soper, Mariam Josyula, Nina Petillo, Sandhitsu Das, Jacob Dubroff, John A. Detre, Joel M. Stein, Kathryn A. Davis","doi":"10.1101/2024.05.28.24308027","DOIUrl":"https://doi.org/10.1101/2024.05.28.24308027","url":null,"abstract":"Background and Significance: Positron Emission Tomography (PET) using fluorodeoxyglucose (FDG-PET) is a standard imaging modality for detecting areas of hypometabolism associated with the seizure onset zone (SOZ) in temporal lobe epilepsy (TLE). However, FDG-PET is costly and involves the use of a radioactive tracer. Arterial Spin Labeling (ASL) offers an MRI-based quantification of cerebral blood flow (CBF) that could also help localize the SOZ, but its performance in doing so, relative to FDG-PET, is limited. In this study, we seek to improve ASL's diagnostic performance by developing a deep learning framework for synthesizing FDG-PET-like images from ASL and structural MRI inputs. Methods: We included 68 epilepsy patients, out of which 36 had well lateralized TLE. We compared the coupling between FDG-PET and ASL CBF values in different brain regions, as well as the asymmetry of these values across the brain. We additionally assessed each modality's ability to lateralize the SOZ across brain regions. Using our paired PET-ASL data, we developed FlowGAN, a generative adversarial neural network (GAN) that synthesizes PET-like images from ASL and T1-weighted MRI inputs. We tested our synthetic PET images against the actual PET images of subjects to assess their ability to reproduce clinically meaningful hypometabolism and asymmetries in TLE. Results: We found variable coupling between PET and ASL CBF values across brain regions. PET and ASL had high coupling in neocortical temporal and frontal brain regions (Spearman's r &gt; 0.30, p &lt; 0.05) but low coupling in mesial temporal structures (Spearman's r &lt; 0.30, p &gt; 0.05). Both whole brain PET and ASL CBF asymmetry values provided good separability between left and right TLE subjects, but PET (AUC = 0.96, 95% CI: [0.88, 1.00]) outperformed ASL (AUC = 0.81; 95% CI: [0.65, 0.96]). FlowGAN-generated images demonstrated high structural similarity to actual PET images (SSIM = 0.85). Globally, asymmetry values were better correlated between synthetic PET and original PET than between ASL CBF and original PET, with a mean correlation increase of 0.15 (95% CI: [0.07, 0.24], p&lt;0.001, Cohen's d = 0.91). Furthermore, regions that had poor ASL-PET correlation (e.g. mesial temporal structures) showed the greatest improvement with synthetic PET images. Conclusions: FlowGAN improves ASL's diagnostic performance, generating synthetic PET images that closely mimic actual FDG-PET in depicting hypometabolism associated with TLE. This approach could improve non-invasive SOZ localization, offering a promising tool for epilepsy presurgical assessment. It potentially broadens the applicability of ASL in clinical practice and could reduce reliance on FDG-PET for epilepsy and other neurological disorders.","PeriodicalId":501358,"journal":{"name":"medRxiv - Radiology and Imaging","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141193099","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":"Accuracy of Combined Deep Learning Algorithms in Detecting Spontaneous Intracranial Hemorrhage on Emergent Head CT Scans","authors":"Takala Juuso, Peura Heikki, Riku Pirinen, Väätäinen Katri, Sergei Terjajev, Ziyuan Lin, Rahul Raj, Korja Miikka","doi":"10.1101/2024.05.28.24308084","DOIUrl":"https://doi.org/10.1101/2024.05.28.24308084","url":null,"abstract":"<strong>Background</strong> Spontaneous intracranial hemorrhages are life-threatening conditions that require fast and accurate diagnosis. We hypothesized that deep learning (DL) could be utilized to detect these hemorrhages with a high accuracy.","PeriodicalId":501358,"journal":{"name":"medRxiv - Radiology and Imaging","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141193180","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 surgery-informed precision approach to determining brain targets for real-time fMRI neurofeedback modulation in chronic pain","authors":"Dan Liu, Yiqi Mi, Menghan Li, Anna Nigri, Marina Grisoli, Keith M Kendrick, Benjamin Becker, Stefania Ferraro","doi":"10.1101/2024.05.24.24307873","DOIUrl":"https://doi.org/10.1101/2024.05.24.24307873","url":null,"abstract":"<strong>Objective</strong> Despite the promising results of neurofeedback with real-time functional magnetic resonance imaging (rt-fMRI-NF) in the treatment of various psychiatric and neurological disorders, few studies have investigated its effects in acute and chronic pain and with mixed results. The lack of clear neuromodulation targets, rooted in the still poorly understood neurophysiopathology of chronic pain, has probably contributed to these inconsistent findings. In contrast, functional neurosurgery (funcSurg) approaches targeting specific brain regions have been shown to reduce pain in a considerable number of patients with chronic pain, however, their invasiveness limits their use to patients in critical situations. In this work, we sought to redefine, in an unbiased manner, rt-fMRI-NF future targets informed by the long tradition of funcSurg approaches.","PeriodicalId":501358,"journal":{"name":"medRxiv - Radiology and Imaging","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141165748","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":"Diagnostic Performances of GPT-4o, Claude 3 Opus, and Gemini 1.5 Pro in “Diagnosis Please” Cases","authors":"Yuki Sonoda, Ryo Kurokawa, Yuta Nakamura, Jun Kanzawa, Mariko Kurokawa, Yuji Ohizumi, Wataru Gonoi, Osamu Abe","doi":"10.1101/2024.05.26.24307915","DOIUrl":"https://doi.org/10.1101/2024.05.26.24307915","url":null,"abstract":"<strong>Backgrounds</strong> Large language models (LLMs) are rapidly advancing and demonstrating high performance in understanding textual information, suggesting potential applications in interpreting patient histories and documented imaging findings. LLMs are advancing rapidly and an improvement in their diagnostic ability is expected. Furthermore, there has been a lack of comprehensive comparisons between LLMs from various manufacturers.","PeriodicalId":501358,"journal":{"name":"medRxiv - Radiology and Imaging","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141165746","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":"Generating intermediate slices with U-nets in craniofacial CT images","authors":"Soh Nishimoto, Kenichiro Kawai, Koyo Nakajima, Hisako Ishise, Masao Kakibuchi","doi":"10.1101/2024.05.08.24307089","DOIUrl":"https://doi.org/10.1101/2024.05.08.24307089","url":null,"abstract":"<strong>Aim</strong> The Computer Tomography (CT) imaging equipment varies across facilities, leading to inconsistent image conditions. This poses challenges for deep learning analysis using collected CT images. To standardize the shape of the matrix, the creation of intermediate slice images with the same width is necessary. This study aimed to generate inter-slice images from two existing CT images.","PeriodicalId":501358,"journal":{"name":"medRxiv - Radiology and Imaging","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140935931","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":"Automatic Lymph Nodes Segmentation and Histological Status Classification on Computed Tomography Scans Using Convolutional Neural Network","authors":"Alexey Shevtsov, Iaroslav Tominin, Vladislav Tominin, Vsevolod Malevanniy, Yury Esakov, Zurab Tukvadze, Andrey Nefedov, Piotr Yablonskii, Pavel Gavrilov, Vadim Kozlov, Mariya Blokhina, Elena Nalivkina, Victor Gombolevskiy, Yuriy Vasilev, Mariya Dugova, Valeria Chernina, Olga Omelyanskaya, Roman Reshetnikov, Ivan Blokhin, Mikhail Belyaev","doi":"10.1101/2024.05.07.24304092","DOIUrl":"https://doi.org/10.1101/2024.05.07.24304092","url":null,"abstract":"Lung cancer is the second most common type of cancer worldwide, making up about 20% of all cancer deaths with less than 10% 5-year survival rate for the very late stage. The recent guidelines for the most common non-small-cell lung cancer (NSCLC) type recommend performing staging based on the 8th edition of TNM classification, where the mediastinal lymph node involvement plays a key role. However, most of the non-invasive methods have a very limited level of sensitivity and are relatively accurate, but invasive methods can be contradicted for some patients. Current advances in Deep Learning show great potential in solving such problems. Still, most of these works focus on the algorithmic side of the problem, not the clinical relevance. Moreover, none of them addressed individual lymph node malignancy classification problem, restricting the indirect analysis of the whole study, and limiting the interpretability of the result without giving an option for cliniciansto validate the result. This work mitigates these gaps, proposing a multi-step algorithm for each visible mediastinal lymph node segmentation and assessing the probability of its involvement in themetastatic process, using the results of histological verification on training. The developed pipelineshows 0.74 ± 0.01 average Recall with 0.53 ± 0.26 object Dice Score for the clinically relevant lymph nodes segmentation task and 0.73 ROC AUC for patient’s N-stage prediction, outperformingtraditional size-based criteria.","PeriodicalId":501358,"journal":{"name":"medRxiv - Radiology and Imaging","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140935782","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":"K-means clustering of hyperpolarised 13C-MRI identifies intratumoural perfusion/metabolism mismatch in renal cell carcinoma as best predictor of highest grade","authors":"Ines Horvat-Menih, Alixander S Khan, Mary A McLean, Joao Duarte, Eva Serrao, Stephan Ursprung, Joshua D Kaggie, Andrew B Gill, Andrew N Priest, Mireia Crispin-Ortuzar, Anne Y Warren, Sarah J Welsh, Thomas J Mitchell, Grant D Stewart, Ferdia A Gallagher","doi":"10.1101/2024.05.06.24306829","DOIUrl":"https://doi.org/10.1101/2024.05.06.24306829","url":null,"abstract":"<strong>Purpose</strong> Conventional renal mass biopsy approaches are inaccurate, potentially leading to undergrading. This study explored using hyperpolarised [1-¹³C]pyruvate MRI (HP ¹³C-MRI) to identify the most aggressive areas within the tumour of patients with clear cell renal cell carcinoma (ccRCC).","PeriodicalId":501358,"journal":{"name":"medRxiv - Radiology and Imaging","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140935851","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":"Probing intratumoral metabolic compartmentalisation in fumarate hydratase-deficient renal cancer using clinical hyperpolarised 13C-MRI and mass spectrometry imaging","authors":"Ines Horvat-Menih, Ruth Casey, James Denholm, Gregory Hamm, Heather Hulme, John Gallon, Alixander S Khan, Joshua Kaggie, Andrew B Gill, Andrew N Priest, Joao A G Duarte, Cissy Yong, Cara Brodie, James Whitworth, Simon T Barry, Richard J A Goodwin, Shubha Anand, Marc Dodd, Katherine Honan, Sarah J Welsh, Anne Y Warren, Tevita Aho, Grant D Stewart, Thomas J Mitchell, Mary A McLean, Ferdia A Gallagher","doi":"10.1101/2024.05.06.24306817","DOIUrl":"https://doi.org/10.1101/2024.05.06.24306817","url":null,"abstract":"<strong>Background</strong> Fumarate hydratase-deficient renal cell carcinoma (FHd-RCC) is a rare and aggressive renal cancer subtype characterised by increased fumarate accumulation and upregulated lactate production. Renal tumours demonstrate significant intratumoral metabolic heterogeneity, which may contribute to treatment failure. Emerging non-invasive metabolic imaging techniques have clinical potential to more accurately phenotype tumour metabolism and its heterogeneity.","PeriodicalId":501358,"journal":{"name":"medRxiv - Radiology and Imaging","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140935776","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":"ShapeMed-Knee: A Dataset and Neural Shape Model Benchmark for Modeling 3D Femurs","authors":"Anthony A. Gatti, Louis Blankemeier, Dave Van Veen, Brian Hargreaves, Scott L. Delp, Garry E. Gold, Feliks Kogan, Akshay S. Chaudhari","doi":"10.1101/2024.05.06.24306965","DOIUrl":"https://doi.org/10.1101/2024.05.06.24306965","url":null,"abstract":"Analyzing anatomic shapes of tissues and organs is pivotal for accurate disease diagnostics and clinical decision-making. One prominent disease that depends on anatomic shape analysis is osteoarthritis, which affects 30 million Americans. To advance osteoarthritis diagnostics and prognostics, we introduce <em>ShapeMed-Knee</em>, a 3D shape dataset with 9,376 high-resolution, medical-imaging-based 3D shapes of both femur bone and cartilage. Besides data, ShapeMed-Knee includes two benchmarks for assessing reconstruction accuracy and five clinical prediction tasks that assess the utility of learned shape representations. Leveraging ShapeMed-Knee, we develop and evaluate a novel hybrid explicit-implicit neural shape model which achieves up to 40% better reconstruction accuracy than a statistical shape model and implicit neural shape model. Our hybrid models achieve state-of-the-art performance for preserving cartilage biomarkers; they’re also the first models to successfully predict localized structural features of osteoarthritis, outperforming shape models and convolutional neural networks applied to raw magnetic resonance images and segmentations. The ShapeMed-Knee dataset provides medical evaluations to reconstruct multiple anatomic surfaces and embed meaningful disease-specific information. ShapeMed-Knee reduces barriers to applying 3D modeling in medicine, and our benchmarks highlight that advancements in 3D modeling can enhance the diagnosis and risk stratification for complex diseases. The dataset, code, and benchmarks will be made freely accessible.","PeriodicalId":501358,"journal":{"name":"medRxiv - Radiology and Imaging","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140935777","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}