Caoilfhionn Ní Leidhin, Michael Paddock, Paul M Parizel, Richard R Warne, Peter Shipman, Rahul Lakshmanan
{"title":"Paediatric cranial ultrasound: assessment of the preterm brain.","authors":"Caoilfhionn Ní Leidhin, Michael Paddock, Paul M Parizel, Richard R Warne, Peter Shipman, Rahul Lakshmanan","doi":"10.1186/s13244-025-02030-5","DOIUrl":"https://doi.org/10.1186/s13244-025-02030-5","url":null,"abstract":"<p><p>Cranial ultrasound is an invaluable tool in assessing neonatal brain anatomy and pathology. It is accessible, relatively quick, inexpensive, safe, portable and generally well-tolerated. This pictorial review focuses on the use of cranial ultrasound in evaluating the premature brain. We illustrate the different grades of intraventricular haemorrhage, the most common sequela of prematurity, its evolution and potential complications, as well as periventricular leukomalacia. Anatomical variants and benign findings that mimic preterm brain injury are also discussed. CRITICAL RELEVANCE STATEMENT: Cranial US is an invaluable tool for assessing neonatal brain anatomy and pathology and can be used in preterm infants to diagnose, monitor and assess for complications of intraventricular haemorrhage and periventricular leukomalacia. KEY POINTS: Cranial US (CUS) is an invaluable tool for assessing the neonatal brain and has many advantages over MRI. CUS can detect intraventricular haemorrhage and periventricular leukomalacia, the most important sequelae of prematurity. Knowledge of optimal CUS technique, normal anatomy, and variants/benign sonographic findings that mimic pathology is crucial to avoid misdiagnosis and unnecessary concern.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"16 1","pages":"158"},"PeriodicalIF":4.1,"publicationDate":"2025-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144690103","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Caoilfhionn Ní Leidhin, Michael Paddock, Paul M Parizel, Richard R Warne, Peter Shipman, Rahul Lakshmanan
{"title":"Paediatric cranial ultrasound: abnormalities of the brain in term neonates and young infants.","authors":"Caoilfhionn Ní Leidhin, Michael Paddock, Paul M Parizel, Richard R Warne, Peter Shipman, Rahul Lakshmanan","doi":"10.1186/s13244-025-02031-4","DOIUrl":"https://doi.org/10.1186/s13244-025-02031-4","url":null,"abstract":"<p><p>Cranial ultrasound is a critical screening tool in the detection of cerebral abnormalities in term neonates and infants, and is complementary to other imaging modalities. This pictorial review illustrates the diverse central nervous system pathologies which can affect the term neonatal and infantile brain, including vascular abnormalities (hypoxic ischaemic injury, perinatal arterial ischaemic stroke, cerebral sinovenous thrombosis, vein of Galen aneurysmal malformations, subpial haemorrhage, and dural sinus malformations); infections (congenital (cytomegalovirus and toxoplasmosis) and bacterial meningoencephalitis); genetic disorders and malformations (callosal agenesis, tuberous sclerosis, developmental megalencephaly, lissencephaly-pachygyria, and grey matter heterotopia); tumours (choroid plexus papilloma, atypical teratoid/rhabdoid tumour, and desmoplastic infantile glioma) and trauma (birth-related, inflicted injury). Each condition is explored with a focus on its sonographic characteristics-some have rarely, if ever, been described on ultrasound. CRITICAL RELEVANCE STATEMENT: Through this case review, we illustrate various pathologies affecting the term neonatal and infantile brain, including vascular lesions, infection, genetic disorders/malformations, tumours and trauma: some of these pathologies have rarely, if ever, been described on CUS. KEY POINTS: Cranial ultrasound (CUS) is a critical screening tool for the term brain. Many term neonatal and infantile pathologies can be detected on CUS. Some of the pathologies illustrated in this paper have rarely been described on US.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"16 1","pages":"159"},"PeriodicalIF":4.1,"publicationDate":"2025-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144690102","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Hong-Bo Xu, Can Ding, Min Zhao, Fa-Jin Lv, Zhi-Gang Chu
{"title":"Exploring the key clinical and CT characteristics of granulomas mimicking peripheral lung cancers: a case-control study.","authors":"Hong-Bo Xu, Can Ding, Min Zhao, Fa-Jin Lv, Zhi-Gang Chu","doi":"10.1186/s13244-025-02043-0","DOIUrl":"10.1186/s13244-025-02043-0","url":null,"abstract":"<p><strong>Objectives: </strong>Some granulomas exhibit CT manifestations similar to those of peripheral lung cancers (PLCs), often resulting in misdiagnosis. This study aimed to identify the key clinical and CT indicators for differentiating them.</p><p><strong>Materials and methods: </strong>From October 2019 to July 2024, 204 atypical granulomas (no calcification, satellite lesions, and/or halo sign) and 204 size-matched PLCs manifested as solid nodules (SNs) were retrospectively enrolled. Patients' clinical, as well as non-enhanced and contrast-enhanced CT data, were evaluated and compared. The enhancement patterns of lesions included no significant enhancement (▵CT value < 15 HU), rim enhancement, enhancement with well-defined necrosis, heterogeneous enhancement, and homogeneous enhancement. The latter two patterns were further divided into mild (15-29 HU), moderate (30-59 HU), and severe (≥ 60 HU) enhancement.</p><p><strong>Results: </strong>Multivariate analysis revealed that younger age (≤ 63 years) (odds ratio [OR], 5.237; 95% confidence interval [CI], 2.609-10.509; p < 0.001), history of diabetes (OR, 9.097; 95% CI: 3.056-27.077; p < 0.001), irregular shape (OR, 3.603; 95% CI: 1.594-8.142; p = 0.002), lower non-enhanced CT value (≤ 21 HU) (OR, 7.576; 95% CI: 3.720-15.431; p < 0.001), and non-moderate enhancement patterns (OR, 50.065; 95% CI: 20.293-123.517; p < 0.001) were independent predictors of granulomas. The sensitivity, specificity, and area under the curve of this model were 88.7%, 83.8%, and 0.941 (95% CI: 0.919-0.962) (p < 0.001), respectively.</p><p><strong>Conclusions: </strong>In younger (≤ 63 years) patients with diabetes, an irregular SN displaying lower density (≤ 21 HU) in non-enhanced CT and a non-moderate enhancement pattern should first be considered as a granuloma.</p><p><strong>Clinical relevance statement: </strong>Distinguishing atypical granulomas from PLCs can be effectively achieved by evaluating the patient's age, underlying diseases, and the lesion's shape, non-enhanced CT value, and enhancement pattern. This integrated clinical-CT diagnostic approach could provide crucial insights for guiding subsequent clinical management.</p><p><strong>Key points: </strong>Atypical granulomas and PLCs exhibit high morphological similarity. Enhancement patterns of lesions are crucial for differentiating atypical granulomas and PLCs. Atypical granulomas typically display irregular shape, lower non-enhanced CT value, and non-moderate enhancement pattern. Younger age and a history of diabetes are key clinical indicators of granulomas.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"16 1","pages":"157"},"PeriodicalIF":4.1,"publicationDate":"2025-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12276186/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144667545","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
John Damilakis, Boris Brkljacic, Guy Frija, Timo De Bondt, Graciano Paulo, Virginia Tsapaki, Eliseo Vano
{"title":"Current practices and perceptions on diagnostic reference levels: a EuroSafe Imaging Survey Analysis.","authors":"John Damilakis, Boris Brkljacic, Guy Frija, Timo De Bondt, Graciano Paulo, Virginia Tsapaki, Eliseo Vano","doi":"10.1186/s13244-025-02028-z","DOIUrl":"10.1186/s13244-025-02028-z","url":null,"abstract":"<p><p>Despite progress in implementing diagnostic reference levels (DRLs) across Europe, clinical practices remain variable. This prompts the EuroSafe Imaging campaign to conduct a survey assessing current practices, perceptions, and challenges related to DRLs. A total of 146 responses were collected from radiology departments in 38 countries, predominantly in the EU/EEA region. While 52.4% reported established local DRLs, significant gaps were identified, with 34.5% lacking local DRLs and 13.1% unaware of their existence. DRLs were primarily established for computed tomography (CT) (88.7%) and conventional radiography (77.5%), with lower implementation in interventional radiology (36.6%). Key challenges included time constraints, data collection difficulties, and limited standardization across institutions. Education gaps were notable, with less than half of the respondents reporting DRL-related training for radiology residents. Respondents emphasized the need for dose management systems, personalized DRLs based on clinical indications, and enhanced education and policy support. Addressing barriers through targeted training, policy enhancements, and technological innovations can improve DRL implementation. Future efforts should focus on promoting standardized clinical protocols, increasing awareness, and fostering European and international collaboration to ensure the consistent use and optimization of DRLs in clinical practice. CRITICAL RELEVANCE STATEMENT: The article critically examines the variability and challenges in implementing diagnostic reference levels (DRLs) across European radiology departments, providing actionable recommendations on policy, education, and technological advancements to optimize radiation protection and improve clinical radiology practices. Diagnostic reference levels (DRLs) help healthcare providers ensure that radiation doses from medical imaging, like CT scans and X-rays, are not higher than necessary. This study looked at how DRLs are used across Europe. It found that while many hospitals have established and follow DRLs, others do not, which may affect patient safety. Challenges like time constraints and lack of training prevent better use of DRLs. Improving education for medical staff and updating protocols can help protect patients by reducing unnecessary radiation exposure while still ensuring accurate diagnoses. KEY POINTS: Variability persists in diagnostic reference level (DRL) practices across Europe. Over half of radiology departments have established local DRLs. Less than half of radiology residents receive structured DRL training. Improved DRL adoption can optimize radiation protection and patient safety. Collaboration, training, and standardized protocols are essential for better DRL practices.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"16 1","pages":"156"},"PeriodicalIF":4.1,"publicationDate":"2025-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12274150/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144659132","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"ECR 2025 Book of Abstracts.","authors":"","doi":"10.1186/s13244-025-02003-8","DOIUrl":"10.1186/s13244-025-02003-8","url":null,"abstract":"","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"16 Suppl 1","pages":"147"},"PeriodicalIF":4.1,"publicationDate":"2025-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12270984/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144659133","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Bone Reporting and Data System on MRI (Bone-RADS-MRI): a validation study by four readers on 275 cases from three local and two public databases.","authors":"Yue Xing, Yangfan Hu, Xianwei Liu, Defang Ding, Shun Dai, Liangjing Lyu, Guangcheng Zhang, Shiqi Mao, Qian Yin, Junjie Lu, Jiarui Yang, Yang Song, Huan Zhang, Chengzhou Li, Weiwu Yao, Jingyu Zhong","doi":"10.1186/s13244-025-02040-3","DOIUrl":"10.1186/s13244-025-02040-3","url":null,"abstract":"<p><strong>Objective: </strong>To evaluate the reproducibility and effectiveness of the Bone Reporting and Data System on MRI (Bone-RADS-MRI) for incidental solitary bone lesions in adults.</p><p><strong>Materials and methods: </strong>We retrospectively included 275 MRI cases from three local and two public databases, respectively. All the cases were histopathologically or clinically confirmed bone lesions, or \"do not touch\" lesions with typical appearance and remained stable for at least two years. Each lesion with gender, age, and clinical history was categorized according to the Bone-RADS algorithm by two musculoskeletal radiologists and two non-musculoskeletal radiologists. The Bone-RADS categories were as follows: Bone-RADS-1, likely benign, leave alone; Bone-RADS-2, incompletely assessed on imaging, perform different imaging modality; Bone-RADS-3, intermediate, perform follow-up imaging; Bone-RADS-4, suspicious for malignancy or need for treatment, biopsy and/or oncologic referral. Inter-reader agreement was evaluated. The diagnostic performance of the Bone-RADS-MRI was measured for distinguishing intermediate or malignant lesions or osteomyelitis from benign lesions. The histopathology results, clinical diagnosis, or follow-up were used as a standard reference.</p><p><strong>Results: </strong>There were 165 intermediate or malignant lesions or osteomyelitis, and 110 benign lesions, respectively. The inter-reader agreements between two musculoskeletal and between two non-musculoskeletal radiologists were both moderate (weighted kappa 0.572 and 0.520). The diagnostic performance for identifying intermediate or malignant lesions or osteomyelitis ranged according to radiologists with sensitivities of 88.5% to 94.5%, specificities of 55.5% to 74.5%, and accuracies of 76.4% to 82.9%.</p><p><strong>Conclusion: </strong>Bone-RADS-MRI is effective for identifying bone lesions that need further treatment, but it has only moderate reliability for readers with different specialties and experience.</p><p><strong>Critical relevance statement: </strong>With local and public databases, Bone-RADS-MRI has been demonstrated to be a reliable algorithm for musculoskeletal and non-musculoskeletal radiologists with varying experience and an effective tool for identifying incidental solitary bone lesions that \"need treatment\" in adults.</p><p><strong>Key points: </strong>Bone-RADS-MRI needs clinical validation for inter-reader agreement and diagnostic performance. Bone-RADS-MRI achieved moderate agreements between musculoskeletal and non-musculoskeletal radiologists, respectively. Bone-RADS-MRI presented high sensitivities but low specificities for identifying \"need-for-treatment\" bone lesions.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"16 1","pages":"155"},"PeriodicalIF":4.1,"publicationDate":"2025-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12271034/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144659131","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Eva J I Hoeijmakers, Bibi Martens, Joachim E Wildberger, Thomas G Flohr, Cécile R L P N Jeukens
{"title":"Objective assessment of diagnostic image quality in CT scans: what radiologists and researchers need to know.","authors":"Eva J I Hoeijmakers, Bibi Martens, Joachim E Wildberger, Thomas G Flohr, Cécile R L P N Jeukens","doi":"10.1186/s13244-025-02037-y","DOIUrl":"10.1186/s13244-025-02037-y","url":null,"abstract":"<p><strong>Objectives: </strong>Quantifying diagnostic image quality (IQ) is not straightforward but essential for optimizing the balance between IQ and radiation dose, and for ensuring consistent high-quality images in CT imaging. This review provides a comprehensive overview of advanced objective reference-free IQ assessment methods for CT scans, beyond standard approaches.</p><p><strong>Methods: </strong>A literature search was performed in PubMed and Web of Science up to June 2024 to identify studies using advanced objective image quality methods on clinical CT scans. Only reference-free methods, which do not require a predefined reference image, were included. Traditional methods relying on the standard deviation of the Hounsfield units, the signal-to-noise ratio or contrast-to-noise ratio, all within a manually selected region-of-interest, were excluded. Eligible results were categorized by IQ metric (i.e., noise, contrast, spatial resolution and other) and assessment method (manual, automated, and artificial intelligence (AI)-based).</p><p><strong>Results: </strong>Thirty-five studies were included that proposed or employed reference-free IQ methods, identifying 12 noise assessment methods, 4 contrast assessment methods, 14 spatial resolution assessment methods and 7 others, based on manual, automated or AI-based approaches.</p><p><strong>Conclusion: </strong>This review emphasizes the transition from manual to fully automated approaches for IQ assessment, including the potential of AI-based methods, and it provides a reference tool for researchers and radiologists who need to make a well-considered choice in how to evaluate IQ in CT imaging.</p><p><strong>Critical relevance statement: </strong>This review examines the challenge of quantifying diagnostic CT image quality, essential for optimization studies and ensuring consistent high-quality images, by providing an overview of objective reference-free diagnostic image quality assessment methods beyond standard methods.</p><p><strong>Key points: </strong>Quantifying diagnostic CT image quality remains a key challenge. This review summarizes objective diagnostic image quality assessment techniques beyond standard metrics. A decision tree is provided to help select optimal image quality assessment techniques.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"16 1","pages":"154"},"PeriodicalIF":4.1,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12246289/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144608256","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Vanishing pancreas: CT and MRI features and imaging diagnostic strategies.","authors":"Yanjin Qin, Danyang Xu, Yuxin Wu, Xiaoqi Zhou, Chenyu Song, Zhi Dong, Lujie Li, Meicheng Chen, Yanji Luo, Huasong Cai, Mimi Tang, Shi-Ting Feng","doi":"10.1186/s13244-025-01998-4","DOIUrl":"10.1186/s13244-025-01998-4","url":null,"abstract":"<p><p>The vanishing pancreas is a frequently overlooked condition which can result from partial or complete dorsal pancreatic agenesis, intra-pancreatic fat deposition (IPFD) and pancreatic atrophy caused by chronic pancreatitis. A variety of diseases, including cystic fibrosis, maturity-onset diabetes of the young type 8, Shwachman-Diamond syndrome, and Johanson-Blizzard syndrome, can manifest as IPFD. Dorsal pancreatic agenesis can, albeit rarely, coexist with abnormalities or tumors. This review aimed to summarize the various causes that may result in partial or complete vanishing pancreas on computed tomography/magnetic resonance imaging (CT/MRI). We provide a comprehensive review of these imaging findings and their corresponding clinical characteristics, which are crucial for ensuring an accurate diagnosis. CRITICAL RELEVANCE STATEMENT: By reviewing various causes of pancreatic vanishing, we summarize these imaging findings and their corresponding clinical characteristics, which is crucial for ensuring an accurate diagnosis and patient management. KEY POINTS: Imaging findings of partial or complete pancreatic vanishing reveal a hypodense pancreas (resembling fat density) or visibility of only the pancreatic head and proximal body. Pancreatic vanishing can result from dorsal pancreatic agenesis, intra-pancreatic fat deposition, and atrophy caused by chronic pancreatitis. Intra-pancreatic fat deposition is associated with genetic and systemic diseases.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"16 1","pages":"153"},"PeriodicalIF":4.1,"publicationDate":"2025-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12228938/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144567411","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Thomas Kroencke, Stijntje W Dijk, Moritz C Halfmann, Claudia Wollny, Joerg Barkhausen, Olav Janssen, Dimitris Rizopoulos, M G Myriam Hunink
{"title":"Challenges and insights from implementing clinical decision support systems for radiologic imaging: experience from the MIDAS trial.","authors":"Thomas Kroencke, Stijntje W Dijk, Moritz C Halfmann, Claudia Wollny, Joerg Barkhausen, Olav Janssen, Dimitris Rizopoulos, M G Myriam Hunink","doi":"10.1186/s13244-025-02027-0","DOIUrl":"10.1186/s13244-025-02027-0","url":null,"abstract":"<p><p>Clinical decision support systems (CDSSs) have been developed to give guidance for referring physicians to make appropriate decisions at the point of care. The MIDAS study, a multicenter cluster randomized trial at four German university hospitals, was designed to evaluate the effectiveness of a CDSS for imaging referral (ESR iGuide) in routine clinical care. Based on our experience within the MIDAS study, we aim to describe the hurdles and difficulties, as well as the various insights gained, in the process of implementing a CDSS in a clinical and research setting. To successfully implement a CDSS for imaging requests, it is essential to monitor and address technical issues, adapt local workflows, define the scope and content, and prioritize user experience and acceptance.</p><p><strong>Critical relevance statement: </strong>By identifying and addressing the various technical, content-related, and workflow challenges, this article gives valuable insights to facilitate future implementations of the ESR iGuide and similar clinical decision support systems CDSSs for imaging orders.</p><p><strong>Trial registration number: </strong>Approval from the Medical Ethics Review Committee was obtained under protocol numbers 20-069 (Augsburg), B 238/21 (Kiel), 20-318 (Lübeck) and 2020-15125 (Mainz). The trial is registered in the ClinicalTrials.gov register under registration number NCT05490290.</p><p><strong>Key points: </strong>This manuscript reviews the challenges of implementing a clinical decision support system (CDSS) (ESR iGuide). Clinical implementation of a CDSS for imaging requests requires monitoring and adjustments in technical issues, local workflow, scope and content, and attention to user experience and acceptance. Our experience may equip stakeholders with the knowledge to proactively address these challenges.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"16 1","pages":"152"},"PeriodicalIF":4.1,"publicationDate":"2025-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12228898/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144567410","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Interpretable and generalizable deep learning model for preoperative assessment of microvascular invasion and outcome in hepatocellular carcinoma based on MRI: a multicenter study.","authors":"Xue Dong, Xibin Jia, Wei Zhang, Jingxuan Zhang, Hui Xu, Lixue Xu, Caili Ma, Hongjie Hu, Jiawen Luo, Jingfeng Zhang, Zhenchang Wang, Wenbin Ji, Dawei Yang, Zhenghan Yang","doi":"10.1186/s13244-025-02035-0","DOIUrl":"10.1186/s13244-025-02035-0","url":null,"abstract":"<p><strong>Objectives: </strong>This study aimed to develop an interpretable, domain-generalizable deep learning model for microvascular invasion (MVI) assessment in hepatocellular carcinoma (HCC).</p><p><strong>Methods: </strong>Utilizing a retrospective dataset of 546 HCC patients from five centers, we developed and validated a clinical-radiological model and deep learning models aimed at MVI prediction. The models were developed on a dataset of 263 cases consisting of data from three centers, internally validated on a set of 66 patients, and externally tested on two independent sets. An adversarial network-based deep learning (AD-DL) model was developed to learn domain-invariant features from multiple centers within the training set. The area under the receiver operating characteristic curve (AUC) was calculated using pathological MVI status. With the best-performed model, early recurrence-free survival (ERFS) stratification was validated on the external test set by the log-rank test, and the differentially expressed genes (DEGs) associated with MVI status were tested on the RNA sequencing analysis of the Cancer Imaging Archive.</p><p><strong>Results: </strong>The AD-DL model demonstrated the highest diagnostic performance and generalizability with an AUC of 0.793 in the internal test set, 0.801 in external test set 1, and 0.773 in external test set 2. The model's prediction of MVI status also demonstrated a significant correlation with ERFS (p = 0.048). DEGs associated with MVI status were primarily enriched in the metabolic processes and the Wnt signaling pathway, and the epithelial-mesenchymal transition process.</p><p><strong>Conclusions: </strong>The AD-DL model allows preoperative MVI prediction and ERFS stratification in HCC patients, which has a good generalizability and biological interpretability.</p><p><strong>Critical relevance statement: </strong>The adversarial network-based deep learning model predicts MVI status well in HCC patients and demonstrates good generalizability. By integrating bioinformatics analysis of the model's predictions, it achieves biological interpretability, facilitating its clinical translation.</p><p><strong>Key points: </strong>Current MVI assessment models for HCC lack interpretability and generalizability. The adversarial network-based model's performance surpassed clinical radiology and squeeze-and-excitation network-based models. Biological function analysis was employed to enhance the interpretability and clinical translatability of the adversarial network-based model.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"16 1","pages":"151"},"PeriodicalIF":4.1,"publicationDate":"2025-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12229396/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144559960","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}