Frontiers in radiology最新文献

Metastatic epidural spinal cord compression develops in 5-10% of patients with cancer and is becoming more common as advancement in cancer treatment prolongs survival in patients with cancer (1-3). It represents an oncological emergency as metastatic epidural compression in adjacent neural structures, including the spinal cord and cauda equina, and exiting nerve roots may result in irreversible neurological deficits, pain, and spinal instability. Although management of metastatic epidural spinal cord compression remains palliative, early diagnosis and intervention may improve outcomes by preserving neurological function, stabilizing the vertebral column, and achieving localized tumor and pain control. Imaging serves an essential role in early diagnosis of metastatic epidural spinal cord compression, evaluation of the degree of spinal cord compression and extent of tumor burden, and preoperative planning. This review focuses on imaging features and techniques for diagnosing metastatic epidural spinal cord compression, differential diagnosis, and management guidelines.

As deep learning is widely used in the radiology field, the explainability of Artificial Intelligence (AI) models is becoming increasingly essential to gain clinicians' trust when using the models for diagnosis. In this research, three experiment sets were conducted with a U-Net architecture to improve the disease classification performance while enhancing the heatmaps corresponding to the model's focus through incorporating heatmap generators during training. All experiments used the dataset that contained chest radiographs, associated labels from one of the three conditions ["normal", "congestive heart failure (CHF)", and "pneumonia"], and numerical information regarding a radiologist's eye-gaze coordinates on the images. The paper that introduced this dataset developed a U-Net model, which was treated as the baseline model for this research, to show how the eye-gaze data can be used in multi-modal training for explainability improvement and disease classification. To compare the classification performances among this research's three experiment sets and the baseline model, the 95% confidence intervals (CI) of the area under the receiver operating characteristic curve (AUC) were measured. The best method achieved an AUC of 0.913 with a 95% CI of [0.860, 0.966]. "Pneumonia" and "CHF" classes, which the baseline model struggled the most to classify, had the greatest improvements, resulting in AUCs of 0.859 with a 95% CI of [0.732, 0.957] and 0.962 with a 95% CI of [0.933, 0.989], respectively. The decoder of the U-Net for the best-performing proposed method generated heatmaps that highlight the determining image parts in model classifications. These predicted heatmaps, which can be used for the explainability of the model, also improved to align well with the radiologist's eye-gaze data. Hence, this work showed that incorporating heatmap generators and eye-gaze information into training can simultaneously improve disease classification and provide explainable visuals that align well with how the radiologist viewed the chest radiographs when making diagnosis.

Purpose: Otospongiotic plaques can be seen on conventional computed tomography (CT) as focal lesions around the cochlea. However, the resolution remains insufficient to enable evaluation of intracochlear damage. MicroCT technology provides resolution at the single micron level, offering an exceptional amplified view of the otosclerotic cochlea. In this study, a non-decalcified otosclerotic cochlea was analyzed and reconstructed in three dimensions for the first time, using microCT technology. The pre-clinical relevance of this study is the demonstration of extensive pro-inflammatory buildup inside the cochlea which cannot be seen with conventional cone-beam CT (CBCT) investigation.

Materials and methods: A radiological and a three-dimensional (3D) anatomical study of an otosclerotic cochlea using microCT technology is presented here for the first time. 3D-segmentation of the human cochlea was performed, providing an unprecedented view of the diseased area without the need for decalcification, sectioning, or staining.

Results: Using microCT at single micron resolution and geometric reconstructions, it was possible to visualize the disease's effects. These included intensive tissue remodeling and highly vascularized areas with dilated capillaries around the spongiotic foci seen on the pericochlear bone. The cochlea's architecture as a morphological correlate of the otosclerosis was also seen. With a sagittal cut of the 3D mesh, it was possible to visualize intense ossification of the cochlear apex, as well as the internal auditory canal, the modiolus, the spiral ligament, and a large cochleolith over the osseous spiral lamina. In addition, the oval and round windows showed intense fibrotic tissue formation and spongiotic bone with increased vascularization. Given the recently described importance of the osseous spiral lamina in hearing mechanics and that, clinically, one of the signs of otosclerosis is the Carhart notch observed on the audiogram, a tonotopic map using the osseous spiral lamina as region of interest is presented. An additional quantitative study of the porosity and width of the osseous spiral lamina is reported.

Conclusion: In this study, structural anatomical alterations of the otosclerotic cochlea were visualized in 3D for the first time. MicroCT suggested that even though the disease may not appear to be advanced in standard clinical CT scans, intense tissue remodeling is already ongoing inside the cochlea. That knowledge will have a great impact on further treatment of patients presenting with sensorineural hearing loss.

Composite MRI scales of central nervous system tissue destruction correlate stronger with clinical outcomes than their individual components in multiple sclerosis (MS) patients. Using machine learning (ML), we previously developed Combinatorial MRI scale (COMRISv1) solely from semi-quantitative (semi-qMRI) biomarkers. Here, we asked how much better COMRISv2 might become with the inclusion of quantitative (qMRI) volumetric features and employment of more powerful ML algorithm. The prospectively acquired MS patients, divided into training (n = 172) and validation (n = 83) cohorts underwent brain MRI imaging and clinical evaluation. Neurological examination was transcribed to NeurEx™ App that automatically computes disability scales. qMRI features were computed by lesion-TOADS algorithm. Modified random forest pipeline selected biomarkers for optimal model(s) in the training cohort. COMRISv2 models validated moderate correlation with cognitive disability [Spearman Rho = 0.674; Lin's concordance coefficient (CCC) = 0.458; p < 0.001] and strong correlations with physical disability (Spearman Rho = 0.830-0.852; CCC = 0.789-0.823; p < 0.001). The NeurEx led to the strongest COMRISv2 model. Addition of qMRI features enhanced performance only of cognitive disability model, likely because semi-qMRI biomarkers measure infratentorial injury with greater accuracy. COMRISv2 models predict most granular clinical scales in MS with remarkable criterion validity, expanding scientific utilization of cohorts with missing clinical data.

The transmantle sign is considered to be a magnetic resonance imaging feature specific to patients with type II focal cortical dysplasia; however, this sign can be difficult to distinguish from other pathologies, such as a radial-oriented white matter band in tuberous sclerosis. Here, we report a case showing a high-intensity area on T2-weighted and fluid-attenuated inversion recovery images extending from the ventricle to the cortex associated with atypical histopathological findings containing corpora amylacea. This case demonstrates that some instances of transmantle signs may be due to corpora amylacea accumulation.

Blunt cerebrovascular injuries (BCVIs) are commonly encountered after blunt trauma. Given the increased risk of stroke incurred after BCVI, it is crucial that they are promptly identified, characterized, and treated appropriately. Current screening practices generally consist of computed tomography angiography (CTA), with escalation to digital subtraction angiography for higher-grade injuries. Although it is quick, cost-effective, and readily available, CTA suffers from poor sensitivity and positive predictive value. A review of the current literature was conducted to examine the current state of emergent imaging for BCVI. After excluding reviews, irrelevant articles, and articles exclusively available in non-English languages, 36 articles were reviewed and included in the analysis. In general, as CTA technology has advanced, so too has detection of BCVI. Magnetic resonance imaging (MRI) with sequences such as vessel wall imaging, double-inversion recovery with black blood imaging, and magnetization prepared rapid acquisition echo have notably improved the utility for MRI in characterizing BCVIs. Finally, transcranial Doppler with emboli detection has proven to be associated with strokes in anterior circulation injuries, further allowing for the identification of high-risk lesions. Overall, imaging for BCVI has benefited from a tremendous amount of innovation, resulting in better detection and characterization of this pathology.

Objective: The disease COVID-19 has caused a widespread global pandemic with ~3. 93 million deaths worldwide. In this work, we present three models-radiomics (M_RM), clinical (M_CM), and combined clinical-radiomics (M_RCM) nomogram to predict COVID-19-positive patients who will end up needing invasive mechanical ventilation from the baseline CT scans.

Methods: We performed a retrospective multicohort study of individuals with COVID-19-positive findings for a total of 897 patients from two different institutions (Renmin Hospital of Wuhan University, D₁ = 787, and University Hospitals, US D₂ = 110). The patients from institution-1 were divided into 60% training, ${\text{D}}_1^{\text{T}}(N=473)$ , and 40% test set ${\text{D}}_1^{\text{V}}(N=314)$ . The patients from institution-2 were used for an independent validation test set ${\text{D}}_2^{\text{V}}(N=110)$ . A U-Net-based neural network (CNN) was trained to automatically segment out the COVID consolidation regions on the CT scans. The segmented regions from the CT scans were used for extracting first- and higher-order radiomic textural features. The top radiomic and clinical features were selected using the least absolute shrinkage and selection operator (LASSO) with an optimal binomial regression model within ${\text{D}}_1^{\text{T}}$ .

Results: The three out of the top five features identified using ${\text{D}}_1^{\text{T}}$ were higher-order textural features (GLCM, GLRLM, GLSZM), whereas the last two features included the total absolute infection size on the CT scan and the total intensity of the COVID consolidations. The radiomics model (M_RM) was constructed using the radiomic score built using the coefficients obtained from the LASSO logistic model used within the linear regression (LR) classifier. The M_RM yielded an area under the receiver operating characteristic curve (AUC) of 0.754 (0.709-0.799) on ${\text{D}}_1^{\text{T}}$ , 0.836 on ${\text{D}}_1^{\text{V}}$ , and 0.748 ${\text{D}}_2^{\text{V}}$ . The top prognostic clinical factors identified in the analysis were dehydrogenase (LDH), age, and albumin (ALB). The clinical model had an AUC of 0.784 (0.743-0.825) on $$

A high proportion of massive patients with hepatocellular carcinoma (HCC) are not amenable for surgical resection at initial diagnosis, owing to insufficient future liver remnant (FLR) or an inadequate surgical margin. For such patients, portal vein embolization (PVE) is an essential approach to allow liver hypertrophy and prepare for subsequent surgery. However, the conversion resection rate of PVE only is unsatisfactory because of tumor progression while awaiting liver hypertrophy. We report here a successfully treated case of primary massive HCC, where surgical resection was completed after PVE and multimodality therapy, comprising hepatic artery infusion chemotherapy (HAIC), Lenvatinib plus Sintilimab. A pathologic complete response was achieved. This case demonstrates for the first time that combined PVE with multimodality therapy appears to be safe and effective for massive, potentially resectable HCC and can produce deep pathological remission in a primary tumor.

Identifying white matter (WM) tracts to locate eloquent areas for preoperative surgical planning is a challenging task. Manual WM tract annotations are often used but they are time-consuming, suffer from inter- and intra-rater variability, and noise intrinsic to diffusion MRI may make manual interpretation difficult. As a result, in clinical practice direct electrical stimulation is necessary to precisely locate WM tracts during surgery. A measure of WM tract segmentation unreliability could be important to guide surgical planning and operations. In this study, we use deep learning to perform reliable tract segmentation in combination with uncertainty quantification to measure segmentation unreliability. We use a 3D U-Net to segment white matter tracts. We then estimate model and data uncertainty using test time dropout and test time augmentation, respectively. We use a volume-based calibration approach to compute representative predicted probabilities from the estimated uncertainties. In our findings, we obtain a Dice of ≈0.82 which is comparable to the state-of-the-art for multi-label segmentation and Hausdorff distance <10mm. We demonstrate a high positive correlation between volume variance and segmentation errors, which indicates a good measure of reliability for tract segmentation ad uncertainty estimation. Finally, we show that calibrated predicted volumes are more likely to encompass the ground truth segmentation volume than uncalibrated predicted volumes. This study is a step toward more informed and reliable WM tract segmentation for clinical decision-making.