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). Imaging findings of the patient. (a) Noncontrast CT reveals a slightly hyperdense mass in the left skull base–infratemporal fossa with indistinct margins from adjacent muscles (arrow). (b, c) Contrast-enhanced CT demonstrates delayed and heterogeneous enhancement of the lesion, with dual-phase CT values of approximately 99 HU and 123 HU, respectively (arrow). (d) Coronal images demonstrate the main lesion encircling the left temporomandibular joint (arrow). (e) Bone window images reveal multiple areas of expansile bone destruction (arrow). (f–i) Noncontrast MRI sequences show low signal intensity on T1-weighted imaging, T2-weighted imaging, T2-fluid-attenuated inversion recovery, and diffusion-weighted imaging (arrow). (j) Contrast-enhanced MRI shows no enhancement of the lesion (arrow). (k, l) Pathological findings (hematoxylin–eosin staining, × 100) confirm giant cell reparative granuloma. CT, computed tomography; MRI, magnetic resonance imaging. Pinshan Zhang: conceptualization, writing – original draft, writing – review and editing. The author has nothing to report. The author has nothing to report. The author has nothing to report. The patient provided written informed consent at the time of entering this study. The author declares no conflicts of interest. The data that support the findings of this study are available from the corresponding author upon reasonable request.

ABSTRACT Background 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. Methods 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. Results 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 ( r = 0.77) and phiMap ( r = 0.58) measurements. Staging based on predicted cMap and phiMap demonstrated excellent performance ( p < 0.01), with considerable accuracy for diagnosing cirrhosis (area under the curve: cMap = 0.75 and phiMap = 0.74) among the test set ( n = 40), which included 21 patients with histologically confirmed cirrhosis (52.5%). Conclusions 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.

A 31-year-old woman with untreated hypertension presented with a progressively worsening occipital headache for 3 months. In the last 4 days, she developed a pulsating holocranial headache that awakened her nocturnally and worsened with Valsalva maneuver. Neurological examination revealed no focal deficits. The patient reported decreased visual acuity, and fundoscopic examination demonstrated grade III hypertensive retinopathy, but no optic disc edema. Blood pressure was 230/160 mmHg and laboratory analysis showed elevated serum creatinine (3.42 mg/dL). Autoimmune and infectious markers were negative. A brain magnetic resonance imaging (MRI) (Figure 1) was performed, which revealed T2/FLAIR hyperintensity predominantly involving the brainstem and both middle cerebellar peduncles, extending to cerebellar hemispheres, and mild involvement of the insular cortex and deep white matter of the corona radiata. The diffusion-weighted imaging (DWI) sequence showed no restriction, and the corresponding apparent diffusion coefficient map confirmed vasogenic edema. Susceptibility-weighted imaging (SWI) showed microbleeds in the inferior vermis and right cerebellar hemisphere. The angio-MR showed normal arteries of the circle of Willis (not shown). These findings were compatible with vasogenic edema in the context of central-variant posterior reversible encephalopathy syndrome (PRES). Brain MRI after 2 days of adequate control of arterial hypertension. Axial FLAIR slices from inferior to superior revealed hyperintensity predominantly in the central region including the brainstem, middle cerebellar peduncles, cerebellar hemispheres (arrows) and to a lesser extent supratentorial involvement of the insular cortex (arrowheads) and corona radiata (circles). The DWI sequence showed no diffusion restriction, and the corresponding ADC map confirmed vasogenic edema. SWI sequence revealed microbleeds in the inferior vermis and right cerebellar hemisphere (dotted arrows). The occipito-parietal regions had normal appearance. ADC, apparent diffusion coefficient; DWI, diffusion-weighted imaging; FLAIR, fluid attenuated inversion recovery; SWI, susceptibility-weighted imaging. The patient also developed acute renal failure and received 2 months of treatment with ravulizumab alongside antihypertensive therapy. This led to a normalization of both creatinine levels and blood pressure. She improved clinically and became asymptomatic. On the 2-month follow-up MRI (Figure 2), there was a resolution of signal hyperintensities, although with persistent microbleeds. Control brain MRI at 2 months. FLAIR sequences showed significant resolution of the vasogenic edema, with only a minimal focus of hyperintensity persisting in the left corona radiata (arrows). SWI sequence confirmed persistence of microbleeds in the inferior vermis and right cerebellar hemisphere (dotted arrows). FLAIR, fluid attenuated inversion recovery. SWI, susceptibility-weighted imaging. PRES is an uncommon clinico-radiological syndrome arising from endothelial dysfunction due to diverse triggers. Patients typically present with persistent headaches that may advance to visual disturbances or seizures. Although its pathophysiology is not fully understood, two main mechanisms are proposed: disrupted vascular autoregulation and direct endothelial injury. Most cases follow one of three classic MRI patterns—posterior, superior frontal sulcus, or holohemispheric watershed—marked by reversible vasogenic edema appearing hyperintense on T2/FLAIR (fluid attenuated inversion recovery), often centered in subcortical white matter. Microbleeds occur in up to 60% of patients and do not indicate severity, whereas 15%–30% show small diffusion-restricted foci without clear prognostic impact. Central-variant PRES is less common, affecting basal ganglia, thalamus, cerebellum, or brainstem while sparing parieto-occipital regions, usually in the setting of severe hypertension or renal failure. A crucial diagnostic clue is the clinico-radiological mismatch: mild symptoms contrasted with prominent imaging abnormalities. This distinction aids recognition and appropriate management in clinical practice. Zhao Hui Chen Zhou: writing – original draft, conceptualization. Ana Ramos González: supervision. The authors have nothing to report. The authors have nothing to report. The authors have nothing to report. Informed consent was obtained from the patient. The authors declare no conflicts of interest. The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

A 35-year-old woman with a history of regular menstruation presented with a positive urine pregnancy test and elevated blood human chorionic gonadotropin concentrations. Color Doppler ultrasound showed multiple slightly hyperechoic areas within the uterine cavity. She was admitted to the hospital with a preliminary outpatient diagnosis of “suspected molar pregnancy, pending further evaluation.” After electrocution and curettage, no villous tissue was identified, and postoperative human chorionic gonadotropin concentrations failed to decline. Pelvic MRI showed a round, thick-walled cystic mass in the anterior sacral region (Figure 1). Color Doppler ultrasound showed that the cystic mass contained a yolk sac, fetal bud, and fetal cardiac activity (Figure 2). Surgical pathology subsequently confirmed the presence of villous tissue, consistent with a diagnosis of ectopic pregnancy. The MRI showing the location and signal characteristics of ectopic pregnancy. (a) Axial T1WI shows a thick-walled cystic mass with a slightly low signal in the sacral anterior space. (b) Axial FS-T2WI shows a slightly high-signal “middle ring” in lesions. (c) FS-T2WI in the sagittal position. (d) A low-signal “inner ring” can be seen in the focal capsule on axial FS-T2WI. (e) Axial DWI shows a high signal in the focal “middle ring.” (f) The capsular wall was enhanced in the sagittal view, but no enhancement was found in the capsular cavity. Ultrasound images showing the fluctuation of the fetal bud and heart. (a) Gestational sac structures (yellow arrow) and the fetal bud (red arrow) in the presacral space, and (b) the yolk sac (green arrow) and the fetal cardiac activity (red arrow). Qingying Song: writing – original draft preparation (lead), writing – review and editing (lead). Dan Li: writing – original draft preparation (supporting), writing – review and editing (supporting). The authors have nothing to report. The study was reviewed and approved by the Ethics Committee of Linyi Maternal and Child Health Hospital (QTL-YXLL-2023080). Informed consent was waived because the patient’s information has been anonymized, which was approved by the ethics committee. The authors declare no conflicts of interest. The data is sourced from the big data and artificial intelligence research platform of Linyi Maternal and Child Health Hospital. The data that support the findings of this study are available from the corresponding author upon reasonable request.

ABSTRACTBackgroundIt is found to have association of facial parameters with trisomy 21 fetuses (T 21). We have compared prenasal thickness (PNT), nasal bone length (NBL), and the PNT:NBL ratio of normal fetuses with fetuses with trisomy 21 (T 21) between 16 and 25 weeks of gestation as a diagnostic tool for T 21.MethodsFacial profile images in the two dimensional (2D) gray scale were assessed to measure fetal NBL and PNT between 16 and 25 weeks of gestation. The PNT:NBL ratio of the fetuses was calculated. Nomograms were constructed from the data of morphologically normal fetuses at live birth. The PNT, NBL, and PNT:NBL ratio of fetuses with confirmed T 21 (n = 31) and morphologically normal fetuses at live birth (controls, n = 3485) were compared.ResultsNomograms for PNT, NBL, and the PNT:NBL ratio were constructed. In T 21 fetuses, PNT (> 95th percentile), NBL (< 5th percentile), and the PNT:NBL ratio (> 95th percentile) showed a sensitivity of 25%, 29%, and 45% for PNT, NBL, and PNT:NBL, respectively, and specificity of 95%, 96%, and 94%, for PNT, NBL, and PNT:NBL, respectively. All of these markers showed a negative predictive value of 99%.ConclusionPNT, NBL, and the PNT:NBL ratio have high diagnostic value for fetuses with Down syndrome and can be incorporated easily in the current second trimester screening protocol for T 21. PNT, NBL, and the PNT:NBL ratio are more specific markers for Down syndrome than those used in previous studies.

ABSTRACTBackgroundThe integration of artificial intelligence (AI) in radiology has opened new possibilities for diagnostic accuracy, with large language models (LLMs) showing potential for supporting clinical decision‐making. While proprietary models like ChatGPT have gained attention, open‐source alternatives such as Meta LLaMa 3.1 remain underexplored. This study aims to evaluate the diagnostic accuracy of LLaMa 3.1 in thoracic imaging and to discuss broader implications of open‐source versus proprietary AI models in healthcare.MethodsMeta LLaMa 3.1 (8B parameter version) was tested on 126 multiple‐choice thoracic imaging questions selected from Thoracic Imaging: A Core Review by Hobbs et al. These questions required no image interpretation. The model’s answers were validated by two board‐certified diagnostic radiologists. Accuracy was assessed overall and across subgroups, including intensive care, pathology, and anatomy. Additionally, a narrative review introduces three widely used AI platforms in thoracic imaging: DeepLesion, ChexNet, and 3D Slicer.ResultsLLaMa 3.1 achieved an overall accuracy of 61.1%. It performed well in intensive care (90.0%) and terms and signs (83.3%) but showed variability across subgroups, with lower accuracy in normal anatomy and basic imaging (40.0%). Subgroup analysis revealed strengths in infectious pneumonia and pleural disease, but notable weaknesses in lung cancer and vascular pathology.ConclusionLLaMa 3.1 demonstrates promise as an open‐source NLP tool in thoracic diagnostics, though its performance variability highlights the need for refinement and domain‐specific training. Open‐source models offer transparency and accessibility, while proprietary models deliver consistency. Both hold value, depending on clinical context and resource availability.

No abstract is available for this article.

Abstract BackgroundPrevious studies have demonstrated the underlying neurophysiologic mechanism during general anesthesia in adults. However, the mechanism of propofol‐induced moderate–deep sedation (PMDS) in modulating pediatric neural activity remains unknown, which therefore was investigated in the present study based on functional magnetic resonance imaging (fMRI).MethodsA total of 41 children (5.10 ± 1.14 years, male/female 21/20) with fMRI were employed to construct the functional connectivity network (FCN). The network communication, graph‐theoretic properties, and network hub identification were statistically analyzed (t test and Bonferroni correction) between sedation (21 children) and awake (20 children) groups. All involved analyses were established on the whole‐brain FCN and seven sub‐networks, which included the default mode network (DMN), dorsal attentional network (DAN), salience network (SAN), auditory network (AUD), visual network (VIS), subcortical network (SUB), and other networks (Other).ResultsUnder PMDS, significant decreases in network communication were observed between SUB‐VIS, SUB‐DAN, and VIS‐DAN, and between brain regions from the temporal lobe, limbic system, and subcortical tissues. However, no significant decrease in thalamus‐related communication was observed. Most graph‐theoretic properties were significantly decreased in the sedation group, and all graphical features of the DMN showed significant group differences. The superior parietal cortex with different neurological functions was identified as a network hub that was not greatly affected.ConclusionsAlthough the children had a depressed level of neural activity under PMDS, the crucial thalamus‐related communication was maintained, and the network hub superior parietal cortex stayed active, which highlighted clinical practices that the human body under PMDS is still perceptible to external stimuli and can be awakened by sound or touch.