iRadiology最新文献

{"title":"Giant Cell Reparative Granuloma of the Left Skull Base and Infratemporal Fossa","authors":"Pinshan Zhang","doi":"10.1002/ird3.70053","DOIUrl":"https://doi.org/10.1002/ird3.70053","url":null,"abstract":"<p>A 56-year-old woman presented with a 1-year history of unexplained left facial swelling and pain associated with a self-palpated, firm, and fixed mass. Noncontrast and contrast-enhanced computed tomography (CT) of the mastoid revealed a slightly hyperdense mass measuring approximately 3.7 cm × 6.4 cm × 3.4 cm in the left cranial base–infratemporal fossa, with heterogeneous internal density and irregular, lobulated morphology. The mass encircled the left temporomandibular joint during growth and showed indistinct borders with the masseter, medial pterygoid, and lateral pterygoid muscles. Expansile bone destruction involved the left mandible, condyle, zygomatic arch, sphenoid bone, temporal bone, and medial and lateral pterygoid plates, suggesting invasiveness; delayed heterogeneous enhancement was seen postcontrast. Noncontrast and contrast-enhanced magnetic resonance imaging (MRI) demonstrated a patchy low-signal intensity lesion in the left cranial base–temporal fossa. Low signal intensity was observed on T1-weighted imaging, T2-weighted imaging, T2-fluid-attenuated inversion recovery, and diffusion-weighted imaging sequences, with no postcontrast enhancement. The patient underwent left cranial base–maxillofacial mass resection under general anesthesia. Postoperative pathology confirmed giant cell reparative granuloma. Giant cell reparative granuloma is a rare benign bone lesion with local invasiveness, most commonly affecting the jawbones. In this case, CT showed soft tissue mass formation with expansile bone destruction, whereas MRI demonstrated uniformly low signal intensity across all sequences. These findings reflect the fibrous matrix and hemosiderin deposition characteristic of the lesion and provide important radiological clues. Recognition of these imaging features aids differentiation from malignant lesions; however, definitive diagnosis requires histopathological examination (Figure 1).</p><p><b>Pinshan Zhang:</b> conceptualization, writing – original draft, writing – review and editing.</p><p>The author has nothing to report.</p><p>The author has nothing to report.</p><p>The patient provided written informed consent at the time of entering this study.</p><p>The author declares no conflicts of interest.</p>","PeriodicalId":73508,"journal":{"name":"iRadiology","volume":"4 1","pages":"66-67"},"PeriodicalIF":0.0,"publicationDate":"2026-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ird3.70053","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147637139","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":"β-Galactosidase Based Molecular Imaging for Visualization of Cellular Senescence","authors":"Jiani Huang,&nbsp;Peili Cen,&nbsp;Hong Zhang,&nbsp;Yan Zhong","doi":"10.1002/ird3.70051","DOIUrl":"https://doi.org/10.1002/ird3.70051","url":null,"abstract":"&lt;p&gt;Cellular senescence is a distinct and irreversible biological process characterized by cell cycle arrest. It can be triggered by various stressors, including DNA damage, oxidative stress, telomere dysfunction, oncogenic activation, or extensive replication as well as by physiological stimuli such as developmental and repair signals [&lt;span&gt;1&lt;/span&gt;]. Cellular senescence plays dual roles in physiology and pathology, contributing to processes such as tissue regeneration, wound healing, tumor suppression, fibrosis, aging, and age-related diseases [&lt;span&gt;2&lt;/span&gt;]. Beyond cell cycle arrest, cellular senescence entails multiple hallmarks of phenotypic changes, including DNA damage response, upregulated antiapoptotic pathways, senescence-associated secretory phenotype, and increased lysosomal content [&lt;span&gt;1&lt;/span&gt;]. Among these features, senescence-associated β-galactosidase (SA-β-gal) is highly enriched in lysosomes and has become the most widely used biomarker for detecting senescent cells. Therefore, sensitive and specific detection of SA-β-gal is critical not only for elucidating the biological mechanisms of aging but also for guiding therapeutic strategies against senescence-associated diseases.&lt;/p&gt;&lt;p&gt;Molecular imaging offers a powerful noninvasive approach for real-time in vivo monitoring of senescent cells. It encompasses diverse modalities such as optical imaging, ultrasound, computed tomography, magnetic resonance imaging (MRI), single photon emission computed tomography (SPECT), and positron emission tomography (PET). These techniques enable dynamic visualization and quantification of cellular and molecular events, providing insight into the biological behavior and accumulation of senescent cells in aging and disease [&lt;span&gt;3&lt;/span&gt;]. Given the elevated activity of SA-β-gal in senescent cells, it has been extensively explored as a molecular target for probe development, with fluorescence and PET emerging as the most actively investigated strategies.&lt;/p&gt;&lt;p&gt;Fluorescence imaging has garnered significant attention due to its high sensitivity, specificity, spatial temporal resolution, and real-time feedback. To date, fluorescent β-gal probes have been the most extensively studied modality for cellular senescence detection both in vitro and in vivo, enabling dynamic tracking of SA-β-gal activity [&lt;span&gt;4&lt;/span&gt;]. These probes are typically designed as enzyme-specific substrates that undergo β-gal-mediated cleavage, thereby activating fluorescence in senescent cell microenvironments. For example, Hu et al. designed a near-infrared (NIR) fluorescent probe (XZ1208) with high sensitivity and selectivity against β-gal, which enabled the visualization of senescence in irradiated, naturally aged, and prematurely aged mouse models, as well as in fibrosis and wound-healing contexts [&lt;span&gt;5&lt;/span&gt;]. Rojas-Vázquez et al. reported sulfonic-Cy7Gal, a renal-cleared β-gal-responsive probe allowing noninvasive urinary monitoring of senescence, which correlate","PeriodicalId":73508,"journal":{"name":"iRadiology","volume":"4 1","pages":"63-65"},"PeriodicalIF":0.0,"publicationDate":"2026-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ird3.70051","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147637177","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":"Transforming Skin Cancer Detection With AI-Based Convolutional and Transformer Models","authors":"Selorm Adablanu,&nbsp;Utpal Barman,&nbsp;Dulumani Das,&nbsp;Tuward Jade Dweh","doi":"10.1002/ird3.70058","DOIUrl":"https://doi.org/10.1002/ird3.70058","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Background</h3>\u0000 \u0000 <p>Skin cancer is a major cause of mortality, and early detection is vital for effective treatment. Diagnosis is challenging because of lesion variability. This study adapts VINCE-NET, a hybrid deep-learning model originally designed for stroke detection, to classify melanoma using dermoscopic images.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>VINCE-NET combines vision transformer layers for global context, convolutional neural networks for local features, and long short-term memory for spatial sequence modeling. During preprocessing, Gaussian blur, normalization, and augmentation were applied to reduce noise and handle class imbalance. During training, the public HAM10000 dataset was used in a central processing unit-only Google Colab environment (12.72 GB random access memory, 107.7 GB disk) with an AdamW optimizer, a batch size of 12, learning-rate scheduling, and early stopping (patience = 50). VINCE-NET's performance was compared with those of a convolutional neural networks, long short-term memory, residual network with 50 layers (ResNet-50), visual geometry group network with 16 and 19 layers (VGG-16/19), and densely connected convolutional network with 121 and 201 layers (DenseNet-121/201) under identical preprocessing conditions.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>VINCE-NET achieved 94.1% accuracy, 95.5% precision, 90.4% recall, a 92.9% F1-score, and an area under the receiver operating characteristic curve of 0.98 at a training time of 34,308.42 s. Benchmarks showed that VINCE-NET outperformed baselines while being computationally efficient.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Conclusion</h3>\u0000 \u0000 <p>VINCE-NET provides competitive performance for melanoma classification and feasibility in resource-limited settings. Although promising, VINCE-NET has not been clinically validated yet. Future work will address resolution, ablation studies, interpretability, and external validation.</p>\u0000 </section>\u0000 </div>","PeriodicalId":73508,"journal":{"name":"iRadiology","volume":"4 1","pages":"51-62"},"PeriodicalIF":0.0,"publicationDate":"2026-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ird3.70058","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147636970","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":"Exploring a Novel Conv-Transformer Network for Multi-Modality Heart Segmentation","authors":"Youyou Ding,&nbsp;Hao Dang,&nbsp;Jiayi Luo,&nbsp;Xiaoyu Zhuo,&nbsp;Ningyu Huang,&nbsp;Junsheng Xiao,&nbsp;Zongwang Lv","doi":"10.1002/ird3.70028","DOIUrl":"https://doi.org/10.1002/ird3.70028","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Background</h3>\u0000 \u0000 <p>In recent years, deep convolutional neural networks (CNNs) have achieved great successes in medical imaging. However, it is difficult to obtain accurate pathological information for clinical diagnosis and treatment by leveraging single-modality medical images. This study aims to provide an efficient multimodality whole heart segmentation method for the diagnosis of coronary heart disease.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>We propose SFAM-TransUnet for multimodality whole heart segmentation, a novel deep learning framework combining CNNs and transformers. Primarily, the method integrates CNNs and visual transformers (Vits) into a unified fusion framework. Specifically, the shallow feature fusion module is designed to connect MRI and CT images, thereby providing a powerful and efficient multimodality fusion backbone for semantic segmentation. Furthermore, we propose a fusion ViT (FViT) module including self-attention (SA) and adaptive mutual boost attention (Ada-MBA) to enhance contextual information within and across modalities. The Ada-MBA module assigns attention to semantic perception regions by calculating SA and cross-attention, which improves the ability to understand context from the different modalities. Extensive experiments are conducted on the clinical Multi-Modality Whole Heart Segmentation datasets.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>We successfully improved the whole heart segmentation DSCs to 0.902 (AA), 0.920 (LV-blood), 0.863 (LA-blood), and 0.837 (LV-myo), the HDs to 9.886 (AA), 9.947 (LV-blood), 11.911 (LA-blood), and 13.599 (LV-myo), the PSNR values to 33.577 (AA), 30.091 (LV-blood), 32.055 (LA-blood), and 29.837 (LV-myo), SSMI values to 0.901 (AA), 0.818 (LV-blood), 0.765 (LA-blood), and 0.743 (LV-myo). This demonstrate SFAM-TransUnet outperforms various alternative methods.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Conclusions</h3>\u0000 \u0000 <p>We propose SFAM-TransUnet, an efficient framework tailored for whole heart segmentation that combines CNNs and transformers. It provides a powerful multimodality fusion network to improve the performance of whole heart semantic segmentation. These results demonstrate the efficacy of SFAM-TransUnet in integrating relevant information between different modalities in multimodal tasks.</p>\u0000 </section>\u0000 </div>","PeriodicalId":73508,"journal":{"name":"iRadiology","volume":"4 1","pages":"13-22"},"PeriodicalIF":0.0,"publicationDate":"2026-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ird3.70028","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147637037","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":"Relationship Between Blood Pressure Changes and Evolution of White Matter Hyperintensity Volume on Magnetic Resonance Imaging","authors":"Chenzhen Shen,&nbsp;Hui Xu,&nbsp;Jianfeng Qiu","doi":"10.1002/ird3.70052","DOIUrl":"https://doi.org/10.1002/ird3.70052","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Background</h3>\u0000 \u0000 <p>Hypertension may be a cause of white matter hyperintensity (WMH), but how changes in blood pressure relate to changes in WMH remains unclear. This study aims to clarify the relationship between them.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>Clinical data were retrospectively collected from 466 patients who underwent two cranial magnetic resonance imaging examinations with the lesion segmentation toolbox at 11- to 14-month intervals at the Second Affiliated Hospital of Shandong First Medical University. Patients were categorized according to baseline clinical data, including sex, systolic blood pressure (SBP), age, and Fazekas score. WMH volume (WMHv) was measured on T1-FLAIR and T2-FLAIR images using MATLAB R2019b software. WMHv before and after magnetic resonance imaging, along with related clinical data, were analyzed using IBM SPSS version 26. Nonparametric independent-samples tests were used to assess the relationship between WMHv and SBP within each group.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>WMH was associated with initial age (<i>r</i> = 0.22, <i>p</i> &lt; 0.05), number of lesions (<i>r</i> = 0.62, <i>p</i> &lt; 0.05), and Fazekas score (<i>r</i> = 0.70, <i>p</i> &lt; 0.05); the second age (<i>r</i> = 0.25, <i>p</i> &lt; 0.05), number of lesions (<i>r</i> = 0.67, <i>p</i> &lt; 0.05), and Fazekas score (<i>r</i> = 0.70, <i>p</i> &lt; 0.05) were also associated with WMH whereas sex showed no significant effect (<i>d</i> ≈ −0.095, <i>p</i> &gt; 0.05). Further analysis revealed that the initial number of lesions (<i>r</i> = 0.16, <i>p</i> &lt; 0.05) and Fazekas score (<i>r</i> = 0.09, <i>p</i> &lt; 0.05) were positively associated with WMHv change; the second number of lesions (<i>r</i> = 0.31, <i>p</i> &lt; 0.05) and Fazekas score (<i>r</i> = 0.22, <i>p</i> &lt; 0.05) were also associated with WMHv change. With fluctuating blood pressure, WMHv changes followed a consistent trend (<i>p</i> &gt; 0.05). Only when the Fazekas score was 1 or 2, did blood pressure changes significantly affect WMHv (Adj. <i>p</i> &lt; 0.05). Pairwise comparative analysis showed that only Fazekas score 1 was statistically significant.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Conclusions</h3>\u0000 \u0000 <p>At Fazekas score 1, adjusting SBP may help regulate WMHv. At higher Fazekas scores, SBP change showed no effect on WMHv.</p>\u0000 </section>\u0000 </div>","PeriodicalId":73508,"journal":{"name":"iRadiology","volume":"4 1","pages":"40-50"},"PeriodicalIF":0.0,"publicationDate":"2026-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ird3.70052","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147637138","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":"Integrating Nuclear Medicine Services Into Health Care Systems in Low- and Middle-Income Countries: A Review of Challenges and Innovations","authors":"Biruk Demisse Ayalew,&nbsp;Saim Mahmood Khan,&nbsp;Helina K. Teklehaimanot,&nbsp;Zebeaman Tibebu Gorfu,&nbsp;Biniyam Jemaneh Batu,&nbsp;Zemichael Getu Alemayehu,&nbsp;Brook Lelisa Sime,&nbsp;Eyerusalem Kebede Zewde,&nbsp;Temesgen Mamo Sharew,&nbsp;Muhidin Ibrahim Hundisa,&nbsp;Yeamlak Tariku Tewodros,&nbsp;Jawairya Muhammad Hussain,&nbsp;Barina Khan","doi":"10.1002/ird3.70054","DOIUrl":"https://doi.org/10.1002/ird3.70054","url":null,"abstract":"<p>The burden of noncommunicable diseases is increasing rapidly in low- and middle-income countries creating a growing need for advanced diagnostic and therapeutic modalities. Nuclear medicine offers great potential in disease detection, treatment planning, and monitoring, yet its integration into resource-limited health systems remains challenging. This review synthesizes evidence from peer-reviewed publications and relevant reports from international agencies to examine barriers to, and enablers of, nuclear medicine adoption in these settings. We found that key obstacles include financial constraints, restricted access to essential materials, insufficient regulatory frameworks, and shortages of skilled professionals. These gaps contribute to safety concerns, inadequate waste management, and delays in service delivery. Although global initiatives have strengthened workforce training and promoted regulatory harmonization, persistent issues in financial sustainability and retention of trained staff hinder progress. Technological advances, such as novel imaging and therapeutic approaches, present opportunities; however, their successful implementation requires context-specific strategies that align with local infrastructure and policy realities. Integrating nuclear medicine into health systems in low-resource environments can address multiple health care priorities simultaneously, but this will require targeted investment, sustainable financing mechanisms, and strengthened institutional capacity. Collaborative international support, coupled with locally adapted policies, could accelerate equitable access and improve patient outcomes. Expanding the role of nuclear medicine in these regions has the potential to significantly enhance health care delivery and contribute to closing the global disparity in advanced medical services.</p>","PeriodicalId":73508,"journal":{"name":"iRadiology","volume":"4 1","pages":"3-12"},"PeriodicalIF":0.0,"publicationDate":"2026-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ird3.70054","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147637154","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 Pilot Study on the Accuracy of Large Language Models (ChatGPT, Google Bard, and Microsoft Copilot) for Selecting the Correct Modality for Musculoskeletal Clinical Cases According to the ACR's Appropriateness Criteria","authors":"Thomas Saliba,&nbsp;Patrick Omoumi,&nbsp;Grammatina Boitsios","doi":"10.1002/ird3.70049","DOIUrl":"https://doi.org/10.1002/ird3.70049","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Background</h3>\u0000 \u0000 <p>Large language models (LLMs) are becoming more commonly used in many aspects of radiology. Some authors have previously tested the capacity of various LLMs to suggest the correct imaging modality according to the guidelines of various associations, such as the American College of Radiology. This study aims to test whether free LLMs can suggest the most appropriate imaging modality in various musculoskeletal radiological cases.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>We tested ChatGPT 3.5, Google Bard, and Copilot (Precise) to see if they could correctly suggest the appropriate imaging modality per the American College of Radiology's Appropriateness Criteria, using clinical vignettes from the musculoskeletal section. Seventy-six vignettes were submitted to each chatbot, with the answer only being considered correct if it was the most appropriate according to the guidelines.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>ChatGPT 3.5 was correct in 82% of cases, Bard in 66%, and Copilot in 89% of cases. Bard was unable to answer in four cases, claiming that as a LLM, it was not capable of answering the question.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Conclusions</h3>\u0000 \u0000 <p>We found that all three LLMs were able to suggest the correct modality in a majority of cases. However, there was variability in the performance of the LLMs, with Copilot performing the best overall, with an accuracy of 89%.</p>\u0000 </section>\u0000 </div>","PeriodicalId":73508,"journal":{"name":"iRadiology","volume":"4 1","pages":"35-39"},"PeriodicalIF":0.0,"publicationDate":"2026-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ird3.70049","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147637137","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":"From X-Rays to Intelligence: The Evolution of Medical Imaging in the AI Era","authors":"Zhen Cheng","doi":"10.1002/ird3.70055","DOIUrl":"https://doi.org/10.1002/ird3.70055","url":null,"abstract":"&lt;p&gt;At the start of 2026, &lt;i&gt;iRADIOLOGY&lt;/i&gt; is thrilled to spotlight the transformative journey of artificial intelligence (AI) in medical imaging, a field that is revolutionizing and reshaping modern healthcare practice. When we look back at the long river of history, the historical convergence between medical imaging and machine intelligence can be seen to be rooted in the groundbreaking work of pioneers such as Wilhelm Röntgen, Alan Turing, and many others. Their legacies, though separated by disciplines and decades, now intersect in the realm of intelligent imaging, remolding how we diagnose, treat, and manage diseases.&lt;/p&gt;&lt;p&gt;The story begins in 1895, when Wilhelm Röntgen's discovery of X-rays revolutionized medicine. For the first time, humans could peer inside the living body without surgery, revealing anatomical structures with unprecedented clarity. This breakthrough laid the foundation for medical imaging, a field that would soon become indispensable in diagnosing and treating diseases. Röntgen's work was not only a technical achievement but also a philosophical leap; a recognition that light, when harnessed appropriately, can unveil the invisible.&lt;/p&gt;&lt;p&gt;This spirit of exploration would later be echoed in the work of Alan Turing, whose publication in 1936 [&lt;span&gt;1&lt;/span&gt;] on computable numbers, and the conception of the “Turing machine,” laid the foundation for modern computing. Another masterpiece by Turing, titled “Computing Machinery and Intelligence” [&lt;span&gt;2&lt;/span&gt;] published in 1950, introduced the concept of machines thinking, such as humans, paving the way for advancements in machine learning and pattern recognition. The Turing test proposed in that paper challenged the boundaries between human and machine intelligence, a question that would later resonate deeply in medical AI.&lt;/p&gt;&lt;p&gt;The mid-20th century saw the first convergence of computing and medicine. The development of digital imaging in the 1970s, marked by the invention of computed tomography (CT) by Nobel Prize laureates Godfrey Hounsfield and Allan Cormack, bridged the gap between radiology and digital data. This era also witnessed the emergence of important technologies, including magnetic resonance imaging, positron emission tomography, single-photon emission CT, optical imaging, and ultrasound, which could noninvasively provide information not only on 3D anatomy from multiple aspects but also on key functional and molecular events of desired targets, offering unprecedented insight about how diseases occur and develop.&lt;/p&gt;&lt;p&gt;Turing's vision of machines mimicking human intelligence was partially realized in the 1980s and 1990s as machine learning algorithms began to perform pattern recognition tasks. However, these early works' further application in industries and medical imaging was constrained by the “black-box” nature of their algorithms, insufficient computational capacity, and the lack of large, annotated datasets; challenges that would persist for decades.&lt;/p&gt;&lt;p&gt;","PeriodicalId":73508,"journal":{"name":"iRadiology","volume":"4 1","pages":"1-2"},"PeriodicalIF":0.0,"publicationDate":"2026-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ird3.70055","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147637153","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":"Periostitis as Initial Sign of Pediatric Acute Lymphoblastic Leukemia","authors":"Siddhi Chawla,&nbsp;Gajanand Singh Tanwar","doi":"10.1002/ird3.70056","DOIUrl":"https://doi.org/10.1002/ird3.70056","url":null,"abstract":"&lt;p&gt;A 5-year-old girl presented with 15 days of insidious-onset pain in the lower legs (left &gt; right) and mild forearm pain with gradual worsening. Examination showed normal limb bulk with tenderness along the lateral forearms and legs. Laboratory evaluation revealed leukocytosis (28,760/mm&lt;sup&gt;3&lt;/sup&gt;), elevated erythrocyte sedimentation rate (29 mm/h), and elevated C-reactive protein (24 mg/L). A radiograph of the left leg was normal. Magnetic resonance imaging of both legs demonstrated diffuse hypointense marrow signal on T1- and T2-weighted images (Figure 1a), bilateral symmetrical short-tau inversion recovery hyperintensity around the fibular shafts with postcontrast enhancement consistent with periostitis (Figure 1b), and multiple irregular peripherally enhancing tibial meta-diaphyseal lesions with nonenhancing fibular shafts, consistent with bone infarcts. Bone marrow biopsy confirmed B-cell acute lymphoblastic leukemia (ALL) with BCR::ABL1 [t (9; 22) (q34.1; q11.2)].&lt;/p&gt;&lt;p&gt;The patient began induction therapy with weekly vincristine and daunorubicin, intrathecal methotrexate, and daily prednisolone for 4–6 weeks. Pegylated asparaginase and daily dasatinib were added from week 2 of treatment. After 1 month, bone marrow examination was normocellular with 3% blast cells, and immunophenotyping showed only 0.18% residual B-cell lymphoblastic leukemia.&lt;/p&gt;&lt;p&gt;Multifocal periostitis as an initial manifestation of ALL is rare. Skeletal involvement occurs in 41%–70% of pediatric ALL and may be associated with better survival, supporting the role of skeletal surveys at diagnosis. In pediatric ALL, leukemic blasts infiltrate the bone marrow, leading to medullary cavity expansion, increased intraosseous pressure, and endosteal disruption. This mechanical stress, along with cytokine-mediated periosteal irritation (e.g., interleukins and tumor necrosis factor-α), stimulates subperiosteal new bone formation, appearing radiologically as periostitis. Thus, periostitis may be the initial radiographic manifestation of ALL, even before hematologic abnormalities.&lt;/p&gt;&lt;p&gt;Differential diagnoses for pediatric periostitis include psoriatic or reactive arthritis, hypervitaminosis A, prostaglandin therapy, hypertrophic pulmonary osteoarthropathy, pachydermoperiostosis, scurvy, infections, malignancy, and fractures, including traumatic and nonaccidental injury (e.g., battered child syndrome).&lt;/p&gt;&lt;p&gt;Persistent or unexplained bone pain in children, even with subtle radiographic findings, carries a risk of misdiagnosis. In such cases, magnetic resonance imaging can detect marrow infiltration earlier than radiography, and hematologic evaluation is essential to exclude acute leukemia. Early recognition prevents diagnostic delay and improves outcomes.&lt;/p&gt;&lt;p&gt;&lt;b&gt;Siddhi Chawla:&lt;/b&gt; conceptualization (equal), investigation (equal), methodology (equal), project administration (equal), validation (equal), writing – original draft (equal), writing – review and editing (equal). &lt;b","PeriodicalId":73508,"journal":{"name":"iRadiology","volume":"4 1","pages":"68-69"},"PeriodicalIF":0.0,"publicationDate":"2026-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ird3.70056","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147637155","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":"Virtual Magnetic Resonance Elastography Using a Deep Generative Model for Liver Fibrosis Staging","authors":"Longyu Sun,&nbsp;Yikun Wang,&nbsp;Yan Li,&nbsp;Xumei Hu,&nbsp;Wenyue Mao,&nbsp;Mengting Sun,&nbsp;Zian Wang,&nbsp;Fuhua Yan,&nbsp;Ruokun Li,&nbsp;Chengyan Wang","doi":"10.1002/ird3.70045","DOIUrl":"https://doi.org/10.1002/ird3.70045","url":null,"abstract":"<div>\u0000 \u0000 \u0000 <section>\u0000 \u0000 <h3> Background</h3>\u0000 \u0000 <p>Liver biopsy is invasive, which presents many limitations in clinical settings. Magnetic resonance elastography (MRE) has significant value in non-invasively diagnosing liver fibrosis. However, its use is currently uncommon because it requires specialized equipment. This study aimed to propose and assess the reliability of virtual MRE (vMRE) using diffusion weighted imaging (DWI) and evaluate its effectiveness in diagnosing liver fibrosis.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Methods</h3>\u0000 \u0000 <p>We proposed a Registration-based Generative Adversarial Network-convolutional Block Attention Model (RegGAN-CBAM) to synthesize stiffness (cMap) and viscosity (phiMap) using DWI data acquired from 128 patients diagnosed with liver fibrosis or cirrhosis. Correlation and agreement between native MRE (nMRE)- and vMRE-derived measurements were assessed using Spearman correlation coefficients and Bland–Altman analysis. Receiver operating characteristic curves were constructed to evaluate the diagnostic performance of these measures for cirrhosis.</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Results</h3>\u0000 \u0000 <p>The proposed RegGAN-CBAM model demonstrated favorable performance in image synthesis and estimation. vMRE measures had high consistency with nMRE for images; moreover, they correlated significantly with cMap (<i>r</i> = 0.77) and phiMap (<i>r</i> = 0.58) measurements. Staging based on predicted cMap and phiMap demonstrated excellent performance (<i>p</i> &lt; 0.01), with considerable accuracy for diagnosing cirrhosis (area under the curve: cMap = 0.75 and phiMap = 0.74) among the test set (<i>n</i> = 40), which included 21 patients with histologically confirmed cirrhosis (52.5%).</p>\u0000 </section>\u0000 \u0000 <section>\u0000 \u0000 <h3> Conclusions</h3>\u0000 \u0000 <p>Our study highlights the reliability of our proposed model for liver fibrosis diagnosis. Furthermore, the non-invasive approach may serve as a practical alternative to conventional clinical MRE, particularly in healthcare facilities without access to MRE equipment.</p>\u0000 </section>\u0000 </div>","PeriodicalId":73508,"journal":{"name":"iRadiology","volume":"4 1","pages":"23-34"},"PeriodicalIF":0.0,"publicationDate":"2026-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/ird3.70045","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147637203","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}