Repetitive Early 68Ga-FAPI PET Acquisition Comparing 68Ga-FAPI-02, 68Ga-FAPI-46, and 68Ga-FAPI-74: Methodologic and Diagnostic Implications for Malignant, Inflammatory/Reactive, and Degenerative Lesions.

68Ga-labeled fibroblast activation protein (FAP) inhibitor (68Ga-FAPI) PET targets 68Ga-FAPI-positive activated fibroblasts and is a promising imaging technique for various types of cancer and nonmalignant pathologies. However, discrimination between malignant and nonmalignant 68Ga-FAPI-positive lesions based on static PET with a single acquisition time point can be challenging. Additionally, the optimal imaging time point for 68Ga-FAPI PET has not been identified yet, and different 68Ga-FAPI tracer variants are currently used. In this retrospective analysis, we evaluate the diagnostic value of repetitive early 68Ga-FAPI PET with 68Ga-FAPI-02, 68Ga-FAPI-46, and 68Ga-FAPI-74 for malignant, inflammatory/reactive, and degenerative lesions and describe the implications for future 68Ga-FAPI imaging protocols. Methods: Whole-body PET scans of 24 cancer patients were acquired at 10, 22, 34, 46, and 58 min after the administration of 150-250 MBq of 68Ga-FAPI tracer molecules (8 patients each for 68Ga-FAPI-02, 68Ga-FAPI-46, and 68Ga-FAPI-74). Detection rates and SUVs (SUVmax and SUVmean) for healthy tissues, cancer manifestations, and nonmalignant lesions were measured, and target-to-background ratios (TBR) versus blood and fat were calculated for all acquisition time points. Results: For most healthy tissues except fat and spinal canal, biodistribution analysis showed decreasing uptake over time. We analyzed 134 malignant, inflammatory/reactive, and degenerative lesions. Detection rates were minimally reduced for the first 2 acquisition time points and remained at a constant high level from 34 to 58 min after injection. The uptake of all 3 variants was higher in malignant and inflammatory/reactive lesions than in degenerative lesions. 68Ga-FAPI-46 showed the highest uptake and TBRs in all pathologies. For all variants, TBRs versus blood constantly increased over time for all pathologies, and TBRs versus fat were constant or decreased slightly. Conclusion: 68Ga-FAPI PET/CT is a promising imaging modality for malignancies and benign lesions. Repetitive early PET acquisition added diagnostic value for the discrimination of malignant from nonmalignant 68Ga-FAPI-positive lesions. High detection rates and TBRs over time confirmed that PET acquisition earlier than 60 min after injection delivers high-contrast images. Additionally, considering clinical feasibility, acquisition at 30-40 min after injection might be a reasonable compromise. Different 68Ga-FAPI variants show significant differences in time-dependent biodistributional behavior and should be selected carefully depending on the clinical setting.

Variability of acquisition phase of computed tomography angiography in acute ischemic stroke in a real-world scenario.

OBJECTIVES: The informative value of computed tomography angiography (CTA) depends on the contrast phase in the vessels which may differ depending on the level of local expertise. METHODS: We retrospectively measured vessel contrast density from CTA scans in patients presenting with acute ischemic stroke to a comprehensive stroke center (CSC) or to one of eight primary stroke centers (PSC). CTAs were classified into arterial or venous phases as well as into 1 of 5 phases (early arterial, peak arterial, equilibrium, peak venous, and late venous). RESULTS: Overall, n = 871 CTAs (CSC: n = 431 (49.5%); PSC: n = 440 (50.5%)) were included in the final analysis. A higher venous than arterial contrast density at the level of the circle of Willis was only rarely observed (overall n = 13 (1.5%); CSC: n = 3/431 (0.7%); PCS: n = 10/440 (2.3%); p = 0.09). CTAs acquired in the CSC showed more often an early arterial contrast phase (CSC: n = 371 (86.1%); PSC: n = 153 (34.8%), p < 0.01). Equilibrium contrast phase, i.e., a slightly stronger arterial contrast with clear venous contrast filling, was more frequent in CTAs from the PSCs (CSC: n = 6 (1.4%); PSC: n = 47 (10.7%); p < 0.01). CONCLUSIONS: Despite different technical equipment and examination protocols, the overall number of CTAs with venous contrast was low and did not differ between the CSC and the PCSs. Differences between the further differentiated contrast phases indicate potential for further improvement of CTA acquisition protocols. KEY POINTS: • Despite different technical equipment and examination protocols in the diagnostic workup of acute ischemic stroke, the total number of computed tomography angiography (CTA) with venous contrast was low (n = 13/871; 1.5%). • A higher venous than arterial contrast density at the level of the circle of Willis was not more frequent in CTAs from the centers with a high patient volume (comprehensive stroke center) compared to the hospital with lower patient volume (primary stroke centers). • Differences between the further differentiated contrast phases indicate that there is potential for further improvement of CTA acquisition protocols.

Neuroradiological emergency consultations during the first year of the COVID-19 pandemic.

BACKGROUND: Measures taking aim at minimizing the risk of coronavirus transmission and fear of infection may affect decisions to seek care for other medical emergency conditions. The purpose of this analysis was to analyze intermediate-term effects of the COVID-19 pandemic on neuroradiological emergency consultations (NECs). METHODS: We conducted an ambispective study on NEC requests to a university hospital from a teleradiological network covering 13 hospitals in Germany. Weekly NEC rates for prepandemic calendar weeks (CW) 01/2019-09/2020 were compared with rates during first COVID-19 wave (CW 10-20/2020), first loosening of restrictions (CW 21-29/2020), intensified COVID-19 testing (CW 30-39/2020) and second COVID-19 wave (CW 40-53/2020), and contrasted with COVID-19 incidence in Germany. RESULTS: A total of n = 10 810 NECs were analyzed. Prepandemic NEC rates were stable over time (median: 103, IQR: 97-115). Upon the first COVID-19 wave in Germany, NEC rates declined sharply (median: 86, IQR: 69-92; p < 0.001) but recovered within weeks. Changes in NEC rates after first loosening of restrictions (median: 109, IQR: 98-127; p = 0. 188), a phase of intensified testing (median: 111, IQR: 101-114; p = 0.434) and as of a second COVID-19 wave (median: 102, IQR: 94-112; p = 0. 462) were not significant. Likewise, patient age and gender distribution remained constant. CONCLUSION: Upon the first pandemic COVID-19 wave in Germany, NEC rates declined but recovered within weeks. It is unknown whether this recovery reflects improved medical care and test capabilities or an adjustment of the patients' behaviour.

Variability of computed tomography angiography coverage of lung parenchyma in acute stroke.

BACKGROUND: Computed tomography angiography (CTA) of the head and neck during acute ischemic stroke (AIS) usually includes visualization of lung apices. The possibility to evaluate for pulmonary changes, e.g. peripheral ground-glass and consolidative opacities suggestive of coronavirus disease 2019 (COVID-19)-related pneumonia, depends on the area of the lung covered by CTA. METHODS: We performed an analysis of a real-world scenario assessing the variability of lung coverage on CTA in patients presenting with AIS to a comprehensive stroke center (CSC) or to one of eight primary stroke centers (PSC) within a teleradiological network covered by the comprehensive stroke center in 2019. RESULTS: Our final analysis included n = 940 CTA, and in n = 573 (61%) merely lung apices were covered. In 19/940 (2%) of patients no lung tissue was covered by CTA. CTA scanning protocols in the CSC began significantly more frequently at the level of the ascending aorta (CSC: n = 180 (38.2%), PSC: n = 127 (27.1%), p-value < 0.001) and the aortic arch (CSC: n = 140 (29.7%), PSC: n = 83 (17.7%), p-value < 0.001), and by this covered less frequently the lower lobes compared to CTA acquired in one of the PSC. CONCLUSIONS: In our pre-COVID-19 pandemic representative stroke patient cohort, CTA for AIS covered most often only lung apices. In 37% of the patients CTA visualized at least parts of the lower lobes, the lingula or the middle lobe allowing for a more extensive assessment of the lungs.

Analysis of a Surgical Series of 21 Cerebral Radiation Necroses.

BACKGROUND: There is no standard approach to differentiate cerebral radiation necrosis from tumor recurrence and no standard treatment pathway for symptomatic lesions. In addition, reports on histology-proven radiation necrosis and the underlying pathophysiology are scarce and highly relevant. METHODS: Our monocentric, retrospective analysis included 21 histology-proven cerebral radiation necroses. Our study focused on 1) potential risk factors for the development of radiation necrosis, 2) radiologic and histopathologic features of individual necroses, and 3) the suitability of previously reported magnetic resonance imaging (MRI)-based methods to identify radiation necroses based on specific structural image features. RESULTS: Average time between radiation treatment and development of necrosis was 4.68 years (95% confidence interval, 0.19-9.55 years). Matching available MRI data sets with those of patients with tumor lesions, we compared specificity and sensitivity of 3 previously reported methods to identify radionecrosis based on imaging criteria. In our hands, none of these methods reached a sensitivity ≥70%. Radionecrosis presented with large edema and showed increased levels of cell proliferation, as inferred by Ki-67 staining. Surgical removal of radiation necrosis proved to be a safe approach with low permanent morbidity (<5%) and no mortality. CONCLUSIONS: Although the overall incidence of cerebral radiation necrosis is low, our data suggest an increasing incidence over the last 2 decades, which is likely associated with the use of stereotactic radiotherapy. There are no imaging standards to identify radiation necrosis on standard MRI with structural sequences. Surgical removal of radiation necrosis is associated with low morbidity and mortality.

Evaluation of three automatic brain vessel segmentation methods for stereotactical trajectory planning.

BACKGROUND AND OBJECTIVE: Stereotactical procedures require exact trajectory planning to avoid blood vessels in the trajectory path. Innovation in imaging and image recognition techniques have facilitated the automatic detection of blood vessels during the planning process and may improve patient safety in the future. To assess the feasibility of a vessel detection and warning system using currently available imaging and vessel segmentation techniques. METHODS: Image data were acquired from post-contrast, isovolumetric T1-weighted sequences (T1CE) and time.-of-flight MR angiography at 3T or 7T from a total of nine subjects. Vessel segmentation by a combination of a vessel-enhancement filter with subsequent level-set segmentation was evaluated using three different methods (Vesselness, FastMarching and LevelSet) in 45 stereotactic trajectories. Segmentation results were compared to a gold-standard of manual segmentation performed jointly by two human experts. RESULTS: The LevelSet method performed best with a mean interclass correlation coefficient (ICC) of 0.76 [0.73, 0.81] compared to the FastMarching method with ICC 0.70 [0.67, 0.73] respectively. The Vesselness algorithm achieved clearly inferior overall performance with a mean ICC of 0.56 [0.53, 0.59]. The differences in mean ICC between all segmentation methods were statistically significant (p < 0.001 with post-hoc p < 0.026). The LevelSet method performed likewise good in MPRAGE and 3T-TOF images and excellent in 7T-TOF image data. The negative predictive value (NPV) was very high (>97%) for all methods and modalities. Positive predictive values (PPV) were found in the overall range of 65-90% likewise depending on algorithm and modality. This pattern reflects the disposition of all segmentation methods - in case of misclassification - to produce preferentially false-positive than false-negative results. In a clinical setting, two to three potential collision warnings would be given per trajectory on average with a PPV of around 50%. CONCLUSIONS: It is feasible to integrate a clinically meaningful vessel detection and collision warning system into stereotactical planning software. Both, T1CE and MRA sequences are suitable as image data for such an application.

Deep MRI brain extraction: A 3D convolutional neural network for skull stripping.

Brain extraction from magnetic resonance imaging (MRI) is crucial for many neuroimaging workflows. Current methods demonstrate good results on non-enhanced T1-weighted images, but struggle when confronted with other modalities and pathologically altered tissue. In this paper we present a 3D convolutional deep learning architecture to address these shortcomings. In contrast to existing methods, we are not limited to non-enhanced T1w images. When trained appropriately, our approach handles an arbitrary number of modalities including contrast-enhanced scans. Its applicability to MRI data, comprising four channels: non-enhanced and contrast-enhanced T1w, T2w and FLAIR contrasts, is demonstrated on a challenging clinical data set containing brain tumors (N=53), where our approach significantly outperforms six commonly used tools with a mean Dice score of 95.19. Further, the proposed method at least matches state-of-the-art performance as demonstrated on three publicly available data sets: IBSR, LPBA40 and OASIS, totaling N=135 volumes. For the IBSR (96.32) and LPBA40 (96.96) data set the convolutional neuronal network (CNN) obtains the highest average Dice scores, albeit not being significantly different from the second best performing method. For the OASIS data the second best Dice (95.02) results are achieved, with no statistical difference in comparison to the best performing tool. For all data sets the highest average specificity measures are evaluated, whereas the sensitivity displays about average results. Adjusting the cut-off threshold for generating the binary masks from the CNN's probability output can be used to increase the sensitivity of the method. Of course, this comes at the cost of a decreased specificity and has to be decided application specific. Using an optimized GPU implementation predictions can be achieved in less than one minute. The proposed method may prove useful for large-scale studies and clinical trials.

Imaging cell size and permeability in biological tissue using the diffusion-time dependence of the apparent diffusion coefficient.

The purpose of this study was to analyze and evaluate a model of restricted water diffusion between equidistant permeable membranes for cell-size and permeability measurements in biological tissue. Based on the known probability distribution of diffusion distances after the diffusion time τ in a system of permeable membranes characterized by three parameters (membrane permeability P, membrane distance L, and free diffusivity D0), an equivalent dimensionless model was derived with a probability distribution characterized by only a single (dimensionless) tissue parameter [Formula: see text]. Evaluating this proposed model function, the dimensionless diffusion coefficient [Formula: see text] was numerically calculated for 60 values of the dimensionless diffusion time [Formula: see text] and 35 values of [Formula: see text]. Diffusion coefficients were measured in a carrot by diffusion-weighted magnetic resonance imaging (MRI) at 18 diffusion times between 9.9 and 1022.7 ms and fitted to the simulation results [Formula: see text] to determine L, P, and D0. The measured diffusivities followed the simulated dependence of [Formula: see text]. Determined cell sizes varied from 21 to 76 µm, permeabilities from 0.007 to 0.039 µm(-1), and the free diffusivities from 1354 to 1713 µm(2) s(-1). In conclusion, the proposed dimensionless tissue model can be used to determine tissue parameters (D0, L, P) based on diffusion MRI with multiple diffusion times. Measurements in a carrot showed a good agreement of the cell diameter, L, determined by diffusion MRI and by light microscopy.