Frontiers in radiology最新文献

{"title":"A retrospective study of 96 cases comparing x-ray radiography and MRI for diagnosing paediatric subacute osteomyelitis.","authors":"Elio Paris, Ahmer A Khan, Giacomo De Marco, Anne Tabard-Fougère, Oscar Vazquez, Christina Steiger, Romain Dayer, Dimitri Ceroni","doi":"10.3389/fradi.2026.1744940","DOIUrl":"https://doi.org/10.3389/fradi.2026.1744940","url":null,"abstract":"<p><strong>Background: </strong>Subacute hematogenous osteomyelitis (SAHOM) presents a diagnostic challenge, requiring robust validation of imaging accuracy.</p><p><strong>Purpose: </strong>To determine the superior diagnostic performance of MRI vs. radiography (x-ray) in detecting and classifying SAHOM.</p><p><strong>Methods: </strong>This retrospective study included 96 proven SAHOM cases (2000-2025). Demographic data, involved bones, and microbiological results were collected. Two independent readers assessed x-Ray and MRI for detection of SAHOM, and classified lesions using the modified Roberts classification. Inter-reader disagreements were resolved by consensus. Sensitivity of x-Ray was evaluated against MRI as the reference standard.</p><p><strong>Results: </strong>x-ray radiographs and MRI from 96 proven cases of SAHOM involving 49 males and 47 females (mean age 47.1 ± 47.6 months) were evaluated. MRI was markedly more sensitive, with significantly more correct imaging findings than radiography for detecting the features of SAHOM (100% vs. 47.9%). Moreover, 21.3% of the SAHOM lesions on x-ray radiography were misclassified. Radiography's limitations were most pronounced for lesions of the spine, tarsal/carpal bones, pelvis, and epiphysis, as well as for infections caused by <i>Kingella kingae</i> (<i>K. kingae</i>).</p><p><strong>Conclusions: </strong>MRI is a more effective method than x-ray radiography for diagnosing SAHOM; it reveals lesions with higher definition and enables their more precise classification. This is especially true of lesions involving the spine, pelvis, tarsal or carpal bones, and the epiphysis, or when SAHOM is caused by <i>K. kingae</i>. MRI also provides much better imaging of the involvement of growth cartilage and damage to articular cartilage.</p>","PeriodicalId":73101,"journal":{"name":"Frontiers in radiology","volume":"6 ","pages":"1744940"},"PeriodicalIF":2.3,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13079598/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147700918","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 in the diagnosis of thyroid diseases: applications and challenges.","authors":"Huayi Zhao, Tingyu Xue, Yan Zhang, Yanchao Lu, Danning Wang, Pei Yin, Lei Jiang, Dan Liu, Yong Wang","doi":"10.3389/fradi.2026.1740915","DOIUrl":"https://doi.org/10.3389/fradi.2026.1740915","url":null,"abstract":"<p><p>Thyroid diseases, a prevalent class of endocrine system disorders, require diagnostic accuracy, which is essential for effective patient treatment and management. In recent years, artificial intelligence (AI) technology has made significant advancements in the medical field, providing new opportunities for the early diagnosis and precise treatment of thyroid diseases. This review discusses the latest applications of AI in the diagnosis of thyroid diseases, with a particular focus on the use of machine learning and deep learning (DL) algorithms in image classification, segmentation, and object detection within thyroid ultrasound, computed tomography, magnetic resonance imaging, and single photon emission computed tomography. Through the integration of cross-modal studies, this article reveals the application of AI across various imaging modalities, highlighting its potential value in feature extraction and risk stratification. Furthermore, we conduct an in-depth analysis of key challenges faced by AI applications, such as data heterogeneity (the decline in model performance due to data differences across institutions and equipment) and insufficient interpretability (DL models often function as \"black boxes,\" making it difficult to provide transparent decision-making rationale, which limits clinical trust and adoption). In summary, AI technology demonstrates notable advantages and developmental potential in the automated diagnosis of thyroid diseases; however, its clinical translation still requires addressing the aforementioned challenges. The resultant analysis demonstrates that AI holds promise in improving the diagnosis and treatment of thyroid diseases, offering new pathways for personalized medicine and better patient outcomes. Specifically, AI-driven tools can reduce diagnostic variability and errors in thyroid nodule assessment, enhance treatment precision with risk-stratified recommendations, and support more consistent, individualized clinical decisions.</p>","PeriodicalId":73101,"journal":{"name":"Frontiers in radiology","volume":"6 ","pages":"1740915"},"PeriodicalIF":2.3,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13079602/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147700915","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":"Subclinical pulmonary congestion in ST-segment elevation myocardial infarction assessed by computed tomography.","authors":"Yufan Gao, Keyi Cui, Shuo Liang, Minghui Hua, Hong Zhang, Zhigang Guo","doi":"10.3389/fradi.2026.1800804","DOIUrl":"https://doi.org/10.3389/fradi.2026.1800804","url":null,"abstract":"<p><strong>Objectives: </strong>To evaluate the prognostic value of subclinical pulmonary congestion, assessed as extravascular lung water (EVLW) by computed tomography (CT), in ST-segment elevation myocardial infarction (STEMI) patients with Killip class 1.</p><p><strong>Methods: </strong>This retrospective study included Killip class 1 STEMI patients who underwent CT prior to primary percutaneous coronary intervention (PCI). EVLW was derived from mean lung density. The Global Registry of Acute Coronary Events (GRACE) score was calculated. The endpoint was in-hospital major adverse cardiovascular events (MACE), defined as all-cause mortality, acute heart failure, cardiogenic shock, resuscitated cardiac arrest, or stroke. The predictive improvement of adding EVLW to the GRACE score was assessed using receiver operating characteristic (ROC) analysis, net reclassification improvement (NRI), and integrated discrimination improvement (IDI).</p><p><strong>Results: </strong>Among 249 patients (mean age 59 ± 11 years; 195 men), 28 experienced MACE. Patients with MACE had a higher GRACE score (144.96 ± 22.95 vs. 133.63 ± 19.84, <i>P</i> = 0.006) and EVLW (24.24% vs. 21.36%, <i>P</i> = 0.001). Adding EVLW to the GRACE score significantly increased the area under the ROC curve (AUC) (0.754 vs. 0.656, <i>P</i> = 0.045), NRI (0.491, <i>P</i> = 0.013), and IDI (0.060, <i>P</i> = 0.019).</p><p><strong>Conclusions: </strong>In Killip class 1 STEMI patients, CT-identified subclinical pulmonary congestion is associated with an increased risk of in-hospital MACE.</p>","PeriodicalId":73101,"journal":{"name":"Frontiers in radiology","volume":"6 ","pages":"1800804"},"PeriodicalIF":2.3,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13081782/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147700947","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":"Imaging and procedural features of staged bilateral magnetic resonance-guided focused ultrasound thalamotomy for essential tremor.","authors":"Fabio Paio, Antonio Ambrosin, Giorgia Bulgarelli, Micaela Tagliamonte, Chiara Zucchella, Elisa Mantovani, Tommaso Bovi, Michele Longhi, Antonio Nicolato, Emanuele Zivelonghi, Luisa Altabella, Carlo Cavedon, Francesco Sala, Benedetto Petralia, Bruno Bonetti, Michele Tinazzi, Stefano Tamburin, Giuseppe K Ricciardi","doi":"10.3389/fradi.2026.1779005","DOIUrl":"https://doi.org/10.3389/fradi.2026.1779005","url":null,"abstract":"<p><strong>Background: </strong>Magnetic Resonance-guided Focused Ultrasound (MRgFUS) thalamotomy is an established incisionless treatment for medication-refractory essential tremor (ET). While staged-bilateral MRgFUS thalamotomy has recently gained clinical acceptance, detailed radiological and procedural comparisons between first- and second-side interventions remain limited. This study aimed to provide a comprehensive imaging-based and procedural characterisation of staged bilateral MRgFUS thalamotomy, with a specific focus on stereotactic targeting, sonication parameters, and lesion morphology.</p><p><strong>Materials and methods: </strong>In this retrospective-prospective, single-centre observational study, consecutive patients with ET undergoing staged bilateral MRgFUS thalamotomy were included. Procedural metrics, stereotactic targeting coordinates, and sonication parameters were compared between FUS1 and FUS2 procedures. MRI analyses were performed using a standardised protocol including morphological pre-treatment 3D T1-weighted imaging and post-treatment 3D T2-weighted imaging acquired within 24 h and at approximately 1 month to document the lesion. Treatment-related lesions were segmented using a semi-automated approach, with volumetric measurements independently obtained by two blinded raters; inter-rater agreement was assessed using intraclass correlation coefficients. Adverse events (AEs) were recorded as secondary outcomes. A systematic review of the literature on treatment strategy of staged bilateral MRgFUS thalamotomy was conducted.</p><p><strong>Results: </strong>Fifteen patients underwent staged bilateral MRgFUS thalamotomy. Most anatomical, procedural, and sonication-related parameters were comparable between FUS1 and FUS2. Final stereotactic targeting during FUS2 showed a small but consistent anterior and dorsal shift relative to FUS1. Lesion volumes measured at both 24 h and 1 month after the procedure did not differ significantly between FUS1 and FUS2, and inter-rater agreement for lesion volumetry was excellent across time points (ICC > 0.91). AEs after FUS2 were predominantly mild and transient. We found no significant differences in lesion volume or inter-side targeting displacement between patients with and without gait disturbances, the most common AE, persisting at 1 month after FUS2. Single-patient imaging analyses suggested heterogeneous spatial lesion configurations in patients with AEs persistent 6-12 months after FUS2.</p><p><strong>Conclusion: </strong>In this single-centre cohort, staged bilateral MRgFUS thalamotomy showed high procedural and radiological consistency between FUS1 and FUS2. MRI-based volumetric analyses show consistent lesion morphology across hemispheres, with small, safety-oriented refinements in second-side targeting, in line with the literature.</p>","PeriodicalId":73101,"journal":{"name":"Frontiers in radiology","volume":"6 ","pages":"1779005"},"PeriodicalIF":2.3,"publicationDate":"2026-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13066307/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147679022","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":"Outcomes of inferior vena cava filtration in lung transplant recipients.","authors":"Kiyon Naser-Tavakolian, Abin Sajan, Venkatesh P Krishnasamy, Tarub Mabud, John Filtes, Asad Baig, Debkumar Sarkar, Stephen P Reis","doi":"10.3389/fradi.2026.1796398","DOIUrl":"https://doi.org/10.3389/fradi.2026.1796398","url":null,"abstract":"<p><strong>Objective: </strong>To evaluate the safety, retrieval outcomes, and procedural characteristics of retrievable inferior vena cava (IVC) filters placed in lung transplant recipients with venous thromboembolism.</p><p><strong>Methods: </strong>This retrospective single-center study included adult lung transplant recipients who underwent retrievable IVC filter placement between January 2021 and April 2025. Collected variables included patient demographics, indication for filter placement, filter type and position, anticoagulation status, retrieval technique, dwell time, fluoroscopy time, radiation dose, complications, and survival. Retrieval probability over time was assessed using Kaplan-Meier analysis.</p><p><strong>Results: </strong>Ninety-five lung transplant recipients (44 men, 51 women; mean age, 60.3 years) underwent retrievable IVC filter placement. Indications included deep vein thrombosis alone in 74 patients (77.9%), combined deep vein thrombosis and pulmonary embolism in 20 (21.1%), and isolated pulmonary embolism in one (1.1%). Filters included Option Elite (<i>n</i> = 93) and Celect (<i>n</i> = 2), with infrarenal placement in 89 patients (93.7%). Sixty-three filters (66.3%) were retrieved after a median dwell time of 7.7 months (interquartile range, 4-10). Retrieval techniques included snare (<i>n</i> = 53) and forceps-assisted (<i>n</i> = 10) with 100% technical success. Mean fluoroscopy time was 8.4 ± 6.5 min, and mean radiation dose was 234.3 ± 200.7 mGy. One filter-related complication (1.1%) occurred, consisting of filter migration into the right renal vein, which was successfully removed without injury; no filter-related deaths were observed.</p><p><strong>Conclusion: </strong>In this single-center retrospective cohort, retrievable IVC filters were managed in lung transplant recipients with a low complication rate and a majority undergoing retrieval during follow-up.</p>","PeriodicalId":73101,"journal":{"name":"Frontiers in radiology","volume":"6 ","pages":"1796398"},"PeriodicalIF":2.3,"publicationDate":"2026-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13050857/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147635099","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":"Partial splenic embolization as management of portal hypertension-related hypersplenism in pediatric patients.","authors":"Karen Juliana Salinas-Castro, Valentina Mejia-Quiñones, Santiago Quiceno-Ramirez, Juan Sebastian Toro-Gutierrez, Verónica Botero, Luis Alfonso Bustamante-Cristancho, Alfonso José Holguín-Holguín","doi":"10.3389/fradi.2026.1731300","DOIUrl":"https://doi.org/10.3389/fradi.2026.1731300","url":null,"abstract":"<p><strong>Introduction: </strong>Endovascular partial splenic embolization (PSE) represents a therapeutic alternative for managing portal hypertension and hypersplenism that preserves splenic parenchyma and maintains immune function. This study aimed to describe the therapeutic response and safety profile of PSE in pediatric patients treated over an eleven-year period at a tertiary care center in Colombia.</p><p><strong>Methods: </strong>This was a descriptive, retrospective case series. We included 12 pediatric patients aged 8-15 years with hypersplenism secondary to portal hypertension who underwent PSE between 2012 and 2023 at our institution. Hematological parameters, endoscopic findings, and post-procedural complications were analyzed.</p><p><strong>Results: </strong>The median age was 13 years; nine patients (75%) had portal hypertension and hypersplenism due to portal vein thrombosis, and three (25%) due to hepatic fibrosis. The median percentage of embolized splenic parenchyma was 40% [IQR (25-60)]. Baseline platelet counts were 58,000 × 10⁹/L (50,250-67,000), and baseline leukocyte counts were 3,485 × 10⁹/L (2,728-4,090). Following PSE, statistically significant increases (delta) were observed in platelet count (median 235,000 × 10⁹/L, 95% CI 119,500-450,500; <i>p</i> = 0.0004) and white blood cell count (median 5,600 × 10⁹/L, 95% CI 2,883-8,925; <i>p</i> = 0.002). Eight patients underwent endoscopic follow-up; six demonstrated improvement in esophageal varices grade. Complications included transient abdominal pain and fever. No major adverse events, such as splenic abscess, infarction, rupture or hematoma, were recorded.</p><p><strong>Conclusions: </strong>In our series, PSE demonstrated safety and efficacy in managing portal hypertension-related hypersplenism in pediatric patients, with significant improvements in hematological parameters and esophageal varices grade in most cases. PSE represents a valuable minimally invasive alternative to splenectomy, preserving splenic function while achieving therapeutic goals, even in patients requiring repeat interventions.</p>","PeriodicalId":73101,"journal":{"name":"Frontiers in radiology","volume":"6 ","pages":"1731300"},"PeriodicalIF":2.3,"publicationDate":"2026-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13038971/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147610681","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":"Biologically informed dual deep learning for skeletal maturity prediction in pediatrics.","authors":"Eugene Rezk","doi":"10.3389/fradi.2026.1774498","DOIUrl":"https://doi.org/10.3389/fradi.2026.1774498","url":null,"abstract":"<p><strong>Background and objective: </strong>Accurate bone age estimation is critical for clinical diagnostics, forensic assessments, and growth research. Traditional methods rely on manual interpretation of radiographs, which is subjective and time-consuming. This Perspective introduces a biologically informed dual deep learning framework that leverages published physiological data to conceptually support bone age prediction.</p><p><strong>Methods: </strong>The framework integrates anatomical and developmental knowledge with a dual neural network architecture. One network extracts morphological features from publicly available datasets, while the second captures age-related growth patterns via supervised learning. No new human or animal data were collected. Illustrative simulations and conceptual analyses explore expected model behavior.</p><p><strong>Results: </strong>Simulations indicate that incorporating biologically informed priors can improve bone age estimation. Embedding physiological knowledge promotes more stable training, reduces prediction variability, and produces outputs that better align with normative growth trajectories compared to conventional AI models without biological priors.</p><p><strong>Conclusions: </strong>We present a theoretically grounded, AI-driven concept for bone age estimation using only published data. Combining biological knowledge with a dual deep learning approach may enhance reproducibility, interpretability, and efficiency. Future work should validate this framework on real-world imaging datasets and assess its integration into automated clinical assessment pipelines.</p>","PeriodicalId":73101,"journal":{"name":"Frontiers in radiology","volume":"6 ","pages":"1774498"},"PeriodicalIF":2.3,"publicationDate":"2026-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13039097/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147610591","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-assisted breast ultrasound: modest AUROC improvement and shorter interpretation time without significant change in diagnostic accuracy.","authors":"Hyuksool Kwon, Sun Mi Kim, Seok Hwan Oh, Mijung Jang, Myeong-Gee Kim, Hyeon-Min Bae, Sang Il Choi, Su Min Cho, Youngmin Kim, Guil Jung, Hyeon-Jik Lee, Sang-Yun Kim","doi":"10.3389/fradi.2026.1747783","DOIUrl":"https://doi.org/10.3389/fradi.2026.1747783","url":null,"abstract":"<p><strong>Objectives: </strong>To evaluate whether Vis-BUS, a commercial artificial intelligence (AI) breast ultrasound detection and analysis software, improves diagnostic discrimination and interpretation efficiency in breast ultrasound examinations.</p><p><strong>Materials and methods: </strong>This retrospective multi-reader study included 258 breast ultrasound examinations (129 malignant and 129 benign lesions). Six radiologists independently interpreted all cases without AI and, after a two-week washout, with AI assistance. Diagnostic performance metrics, including the area under the receiver operating characteristic curve (AUROC), area under the precision-recall curve (AUPRC), accuracy, sensitivity, and specificity, were compared using multi-reader analysis. Median interpretation time per case was recorded and compared using paired statistical tests.</p><p><strong>Results: </strong>Vis-BUS assistance modestly increased the pooled AUROC (0.921 vs. 0.953, <i>p</i> = 0.002) and reduced median reading time (6.0 vs. 3.0 s, <i>p</i> < 0.001), whereas AUPRC, accuracy, sensitivity, and specificity did not differ significantly (all <i>p</i> > 0.06). Accuracy (79.1% vs. 83.9%, <i>p</i> = 0.061), sensitivity (94.2% vs. 96.3%, <i>p</i> = 0.243), and specificity (64.0% vs. 71.6%, <i>p</i> = 0.069) showed no significant differences. Median interpretation time decreased from 6.0 to 3.0 s (<i>p</i> < 0.001). Subgroup analyses demonstrated significant AUROC improvements for dense breasts and tumors ≤ 2 cm (<i>p</i> < 0.001 for both).</p><p><strong>Conclusion: </strong>Vis-BUS AI assistance was associated with improved diagnostic discrimination and shorter interpretation time. However, accuracy, sensitivity, and specificity did not differ significantly. These findings suggest potential efficiency benefits, while the clinical impact remains to be confirmed in prospective multi-center studies.</p>","PeriodicalId":73101,"journal":{"name":"Frontiers in radiology","volume":"6 ","pages":"1747783"},"PeriodicalIF":2.3,"publicationDate":"2026-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13039021/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147610576","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":"Quantitative myocardial perfusion imaging across PET, SPECT, CMR, and CT.","authors":"Yoko Kato, Omar Chehab, Bharath Ambale-Venkatesh, Joao A C Lima","doi":"10.3389/fradi.2026.1760241","DOIUrl":"https://doi.org/10.3389/fradi.2026.1760241","url":null,"abstract":"<p><p>Recent clinical trials demonstrating the prognostic value of coronary computed tomography angiography (CCTA), and the limited incremental prognostic benefit of ischemia-guided revascularization over optimal medical therapy, have shifted focus away from functional testing. Nevertheless, myocardial perfusion assessment remains essential in patients with known or suspected coronary artery disease (CAD), microvascular dysfunction, ischemia with normal coronary arteries (INOCA), coronary anomalies, and cardiomyopathies. Advances in positron emission tomography (PET), cardiovascular magnetic resonance (CMR), and computed tomography (CT) now enable absolute quantification of myocardial blood flow (MBF), providing improved diagnostic and prognostic performance and underscoring the need to reappraise myocardial perfusion imaging. This mini-review summarizes key physiological principles, contemporary acquisition and post-processing strategies, and the clinical relevance of myocardial perfusion imaging (MPI) across PET, Single Photon Emission Computed Tomography (SPECT), CMR, and CT, and discusses future perspectives in quantitative MPI.</p>","PeriodicalId":73101,"journal":{"name":"Frontiers in radiology","volume":"6 ","pages":"1760241"},"PeriodicalIF":2.3,"publicationDate":"2026-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13036148/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147596674","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":"A vision transformer-radiomics approach for enhanced chemotherapy outcome prediction in ovarian cancer.","authors":"Neman Abdoli, Patrik Gilley, Ke Zhang, Youkabed Sadri, Theresa Thai, Yong Chen, Lauren Dockery, Kathleen Moore, Robert Mannel, Yuchen Qiu","doi":"10.3389/fradi.2026.1702977","DOIUrl":"https://doi.org/10.3389/fradi.2026.1702977","url":null,"abstract":"<p><strong>Introduction: </strong>Early prediction of chemotherapy response in ovarian cancer patients is essential for enabling personalized treatment strategies and improving clinical outcomes. However, this prediction remains challenging due to the high heterogeneity of tumor biology, patient-specific factors, and treatment regimens. Recent advances in imaging biomarkers derived from both radiomics and advanced deep learning methods offer promising tools for characterizing tumor phenotypes and predicting treatment outcomes.</p><p><strong>Methods: </strong>In this retrospective study, pre-treatment CT scans from 182 ovarian cancer patients were analyzed. Three categories of imaging features were extracted: handcrafted radiomics descriptors, embeddings from a pretrained Vision Transformer (ViT), and embeddings from MedSAM, a medical foundation model adapted for segmentation. All features were standardized and subjected to least absolute shrinkage and selection operator (LASSO) regression for feature selection. Support vector machine (SVM) classifiers were trained to predict 6-month progression-free survival (PFS). Model performance was evaluated using cross-validated metrics including area under the receiver operating characteristic curve (AUC) and classification accuracy.</p><p><strong>Results: </strong>The combined ViT and MedSAM embedding model achieved the highest AUC of 0.924 ± 0.032. Integration of all three feature groups (radiomics, ViT, and MedSAM) yielded a comparable AUC of 0.924 ± 0.037 and the highest classification accuracy of 0.831 ± 0.042.</p><p><strong>Conclusion: </strong>These findings demonstrate that integrating complementary imaging representations enhances chemotherapy response prediction. The combination of transformer-based embeddings and radiomics features provides rich, task-specific tumor characterization from CT imaging and supports the development of precision oncology decision tools.</p>","PeriodicalId":73101,"journal":{"name":"Frontiers in radiology","volume":"6 ","pages":"1702977"},"PeriodicalIF":2.3,"publicationDate":"2026-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13036210/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147596644","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}