Brain tumor segmentation using synthetic MR images - A comparison of GANs and diffusion models.

Large annotated datasets are required for training deep learning models, but in medical imaging data sharing is often complicated due to ethics, anonymization and data protection legislation. Generative AI models, such as generative adversarial networks (GANs) and diffusion models, can today produce very realistic synthetic images, and can potentially facilitate data sharing. However, in order to share synthetic medical images it must first be demonstrated that they can be used for training different networks with acceptable performance. Here, we therefore comprehensively evaluate four GANs (progressive GAN, StyleGAN 1-3) and a diffusion model for the task of brain tumor segmentation (using two segmentation networks, U-Net and a Swin transformer). Our results show that segmentation networks trained on synthetic images reach Dice scores that are 80%-90% of Dice scores when training with real images, but that memorization of the training images can be a problem for diffusion models if the original dataset is too small. Our conclusion is that sharing synthetic medical images is a viable option to sharing real images, but that further work is required. The trained generative models and the generated synthetic images are shared on AIDA data hub.

MRI with generalized diffusion encoding reveals damaged white matter in patients previously hospitalized for COVID-19 and with persisting symptoms at follow-up.

There is mounting evidence of the long-term effects of COVID-19 on the central nervous system, with patients experiencing diverse symptoms, often suggesting brain involvement. Conventional brain MRI of these patients shows unspecific patterns, with no clear connection of the symptomatology to brain tissue abnormalities, whereas diffusion tensor studies and volumetric analyses detect measurable changes in the brain after COVID-19. Diffusion MRI exploits the random motion of water molecules to achieve unique sensitivity to structures at the microscopic level, and new sequences employing generalized diffusion encoding provide structural information which are sensitive to intravoxel features. In this observational study, a total of 32 persons were investigated: 16 patients previously hospitalized for COVID-19 with persisting symptoms of post-COVID condition (mean age 60 years: range 41-79, all male) at 7-month follow-up and 16 matched controls, not previously hospitalized for COVID-19, with no post-COVID symptoms (mean age 58 years, range 46-69, 11 males). Standard MRI and generalized diffusion encoding MRI were employed to examine the brain white matter of the subjects. To detect possible group differences, several tissue microstructure descriptors obtainable with the employed diffusion sequence, the fractional anisotropy, mean diffusivity, axial diffusivity, radial diffusivity, microscopic anisotropy, orientational coherence (Cc) and variance in compartment's size (CMD) were analysed using the tract-based spatial statistics framework. The tract-based spatial statistics analysis showed widespread statistically significant differences (P < 0.05, corrected for multiple comparisons using the familywise error rate) in all the considered metrics in the white matter of the patients compared to the controls. Fractional anisotropy, microscopic anisotropy and Cc were lower in the patient group, while axial diffusivity, radial diffusivity, mean diffusivity and CMD were higher. Significant changes in fractional anisotropy, microscopic anisotropy and CMD affected approximately half of the analysed white matter voxels located across all brain lobes, while changes in Cc were mainly found in the occipital parts of the brain. Given the predominant alteration in microscopic anisotropy compared to Cc, the observed changes in diffusion anisotropy are mostly due to loss of local anisotropy, possibly connected to axonal damage, rather than white matter fibre coherence disruption. The increase in radial diffusivity is indicative of demyelination, while the changes in mean diffusivity and CMD are compatible with vasogenic oedema. In summary, these widespread alterations of white matter microstructure are indicative of vasogenic oedema, demyelination and axonal damage. These changes might be a contributing factor to the diversity of central nervous system symptoms that many patients experience after COVID-19.

[Conditions for performing CT and MRI scans in pregnant and lactating patients]. / Att tänka på vid neuroradiologisk diagnostik av gravida och ammande.

Many women are pregnant during several percent of their lives. Occasionally, there is a need for neuroradiological examinations during pregnancy or lactation. In our clinical work, we regularly see that female patients are being withheld relevant diagnostic scans during pregnancy, due to insufficient knowledge or an unbalanced comparison between benefits and risks. This article describes the current knowledge regarding conditions for performing CT and MRI scans in pregnant and lactating patients, including the use of contrast media. PET scans and reactions to contrast media are briefly mentioned, but interventional radiology is not discussed.

Radiation-dependent demyelination in normal appearing white matter in glioma patients, determined using quantitative magnetic resonance imaging.

Background and purpose: A brain tumour, especially a glioma, is a rare disease; it is challenging to treat and the prognosis is often poor. Routine care includes surgery and concomitant chemoradiotherapy (CRT). Diagnostic work-up and treatment effects are typically evaluated using magnetic resonance imaging (MRI). Quantitative MRI (qMRI), unlike conventional MRI, has the advantage of providing tissue-specific relaxation rates and proton density. The purpose is to detect changes in normal appearing white matter (NAWM) in brain tumour patients after CRT using qMRI. Materials & methods: NAWM was analysed in 10 patients, in 83 MR examinations performed before and after surgery and after CRT. Relaxation rates R1 and R2, the proton density (PD) and the concentration of myelin (cMy) were calculated from the qMRI scans and analysed in correlation to radiation dose and time after treatment. Results: A significant decrease in cMy between pre-treatment imaging and first follow-up and an increase in PD were observed. For low doses (less than 30 Gy) PD and cMy returned to baseline (=pre-operative status), while for high doses (>30 Gy) the change increased during the full extent of the follow-up period. No difference could be established for R1. For R2 an increase was observed during the first year, which then gradually returned to baseline. For R2, stronger effects were seen as a consequence of higher absorbed doses. Conclusion: In the long-term follow-up for glioma patients, qMRI is a powerful tool for detecting small changes, such as a decrease of myelin concentration, in NAWM after CRT.

Evaluation of inverse treatment planning for gamma knife radiosurgery using fMRI brain activation maps as organs at risk.

BACKGROUND: Stereotactic radiosurgery (SRS) can be an effective primary or adjuvant treatment option for intracranial tumors. However, it carries risks of various radiation toxicities, which can lead to functional deficits for the patients. Current inverse planning algorithms for SRS provide an efficient way for sparing organs at risk (OARs) by setting maximum radiation dose constraints in the treatment planning process. PURPOSE: We propose using activation maps from functional MRI (fMRI) to map the eloquent regions of the brain and define functional OARs (fOARs) for Gamma Knife SRS treatment planning. METHODS: We implemented a pipeline for analyzing patient fMRI data, generating fOARs from the resulting activation maps, and loading them onto the GammaPlan treatment planning software. We used the Lightning inverse planner to generate multiple treatment plans from open MRI data of five subjects, and evaluated the effects of incorporating the proposed fOARs. RESULTS: The Lightning optimizer designs treatment plans with high conformity to the specified parameters. Setting maximum dose constraints on fOARs successfully limits the radiation dose incident on them, but can have a negative impact on treatment plan quality metrics. By masking out fOAR voxels surrounding the tumor target it is possible to achieve high quality treatment plans while controlling the radiation dose on fOARs. CONCLUSIONS: The proposed method can effectively reduce the radiation dose incident on the eloquent brain areas during Gamma Knife SRS of brain tumors.

Initial cognitive impairment predicts shorter survival of patients with glioblastoma.

OBJECTIVES: Seizures as presenting symptom of glioblastoma (GBM) are known to predict prolonged survival, whereas the clinical impact of other initial symptoms is less known. Our main objective was to evaluate the influence of different presenting symptoms on survival in a clinical setting. We also assessed lead times, tumour size and localization. METHODS: Medical records of 189 GBM patients were reviewed regarding the first medical appointment, presenting symptom/s, date of diagnostic radiology and survival. Tumour size, localization and treatment data were retrieved. Overall survival was calculated using Kaplan-Meier and Mann-Whitney U test. Cox regression was used for risk estimation. RESULTS: Cognitive impairment as the initial symptom was often misinterpreted in primary health care leading to a delayed diagnosis. Initial global symptoms (66% of all patients) were associated with reduced survival compared to no global symptoms (median 8.4 months vs. 12.6 months). Those with the most common cognitive dysfunctions: change of behaviour, memory impairment and/or disorientation had a reduced median survival to 6.4 months. In contrast, seizures (32%) were associated with longer survival (median 11.2 months vs. 8.3 months). Global symptoms were associated with larger tumours than seizures, but tumour size had no linear association with survival. The setting of the first medical appointment was evenly distributed between primary health care and emergency units. CONCLUSION: Patients with GBM presenting with cognitive symptoms are challenging to identify, have larger tumours and reduced survival. In contrast, epileptic seizures as the first symptom are associated with longer survival and smaller tumours.

Brain MRI and neuropsychological findings at long-term follow-up after COVID-19 hospitalisation: an observational cohort study.

OBJECTIVES: To report findings on brain MRI and neurocognitive function, as well as persisting fatigue at long-term follow-up after COVID-19 hospitalisation in patients identified as high risk for affection of the central nervous system. DESIGN: Ambidirectional observational cohort study. SETTING: All 734 patients from a regional population in Sweden with a laboratory-confirmed COVID-19 diagnosis admitted to hospital during the period 1 March to 31 May 2020. PARTICIPANTS: A subgroup (n=185) with persisting symptoms still interfering with daily life at a telephone follow-up 4 months after discharge were invited for a medical and neuropsychological evaluation. Thirty-five of those who were assessed with a neurocognitive test battery at the clinical visit, and presented a clinical picture concerning for COVID-19-related brain pathology, were further investigated by brain MRI. MAIN OUTCOME MEASURES: Findings on brain MRI, neurocognitive test results and reported fatigue. RESULTS: Twenty-five patients (71%) had abnormalities on MRI; multiple white matter lesions were the most common finding. Sixteen patients (46%) demonstrated impaired neurocognitive function, of which 10 (29%) had severe impairment. Twenty-six patients (74%) reported clinically significant fatigue. Patients with abnormalities on MRI had a lower Visuospatial Index (p=0.031) compared with the group with normal MRI findings. CONCLUSIONS: In this group of patients selected to undergo MRI after a clinical evaluation, a majority of patients had abnormal MRI and/or neurocognitive test results. Abnormal findings were not restricted to patients with severe disease.

Publisher Correction: Indication of Thalamo-Cortical Circuit Dysfunction in Idiopathic Normal Pressure Hydrocephalus: A Diffusion Tensor Imaging Study.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

Indication of Thalamo-Cortical Circuit Dysfunction in Idiopathic Normal Pressure Hydrocephalus: A Tensor Imaging Study.

Idiopathic normal pressure hydrocephalus (iNPH) is a disorder with unclear pathophysiology. The diagnosis of iNPH is challenging due to its radiological similarity with other neurodegenerative diseases and ischemic subcortical white matter changes. By using Diffusion Tensor Imaging (DTI) we explored differences in apparent diffusion coefficient (ADC) and fractional anisotropy (FA) in iNPH patients (before and after a shunt surgery) and healthy individuals (HI) and we correlated the clinical results with DTI parameters. Thirteen consecutive iNPH-patients underwent a pre- and post-operative clinical work-up: 10 m walk time (w10mt) steps (w10ms), TUG-time (TUGt) and steps (TUGs); for cognitive function MMSE. Nine HI were included. DTI was performed before and 3 months after surgery, HI underwent DTI once. DTI differences analyzed by manually placing 12 regions-of-interest. In patients motor and balance function improved significantly after surgery (p = 0.01, p = 0.025). Higher nearly significant FA values found in the patients vs HI pre-operatively in the thalamus (p = 0.07) accompanied by an almost significant lower ADC (p = 0.08). Significantly FA and ADC-values were found between patients and HI in FWM (p = 0.02, p = 0.001) and almost significant (p = 0.057) pre- vs postoperatively. Postoperatively we found a trend towards the HIs FA values and a strong significant negative correlation between FA changes vs. gait results in the FWM (r = -0.7, p = 0.008). Our study gives a clear indication of an ongoing pathological process in the periventricular white matter, especially in the thalamus and in the frontal white matter supporting the hypothesis of a shunt reversible thalamo-cortical circuit dysfunction in iNPH.

Quantitative MRI for analysis of peritumoral edema in malignant gliomas.

BACKGROUND AND PURPOSE: Damage to the blood-brain barrier with subsequent contrast enhancement is a hallmark of glioblastoma. Non-enhancing tumor invasion into the peritumoral edema is, however, not usually visible on conventional magnetic resonance imaging. New quantitative techniques using relaxometry offer additional information about tissue properties. The aim of this study was to evaluate longitudinal relaxation R1, transverse relaxation R2, and proton density in the peritumoral edema in a group of patients with malignant glioma before surgery to assess whether relaxometry can detect changes not visible on conventional images. METHODS: In a prospective study, 24 patients with suspected malignant glioma were examined before surgery. A standard MRI protocol was used with the addition of a quantitative MR method (MAGIC), which measured R1, R2, and proton density. The diagnosis of malignant glioma was confirmed after biopsy/surgery. In 19 patients synthetic MR images were then created from the MAGIC scan, and ROIs were placed in the peritumoral edema to obtain the quantitative values. Dynamic susceptibility contrast perfusion was used to obtain cerebral blood volume (rCBV) data of the peritumoral edema. Voxel-based statistical analysis was performed using a mixed linear model. RESULTS: R1, R2, and rCBV decrease with increasing distance from the contrast-enhancing part of the tumor. There is a significant increase in R1 gradient after contrast agent injection (P < .0001). There is a heterogeneous pattern of relaxation values in the peritumoral edema adjacent to the contrast-enhancing part of the tumor. CONCLUSION: Quantitative analysis with relaxometry of peritumoral edema in malignant gliomas detects tissue changes not visualized on conventional MR images. The finding of decreasing R1 and R2 means shorter relaxation times closer to the tumor, which could reflect tumor invasion into the peritumoral edema. However, these findings need to be validated in the future.

Application of quantitative MRI for brain tissue segmentation at 1.5 T and 3.0 T field strengths.

BACKGROUND: Brain tissue segmentation of white matter (WM), grey matter (GM), and cerebrospinal fluid (CSF) are important in neuroradiological applications. Quantitative Mri (qMRI) allows segmentation based on physical tissue properties, and the dependencies on MR scanner settings are removed. Brain tissue groups into clusters in the three dimensional space formed by the qMRI parameters R1, R2 and PD, and partial volume voxels are intermediate in this space. The qMRI parameters, however, depend on the main magnetic field strength. Therefore, longitudinal studies can be seriously limited by system upgrades. The aim of this work was to apply one recently described brain tissue segmentation method, based on qMRI, at both 1.5 T and 3.0 T field strengths, and to investigate similarities and differences. METHODS: In vivo qMRI measurements were performed on 10 healthy subjects using both 1.5 T and 3.0 T MR scanners. The brain tissue segmentation method was applied for both 1.5 T and 3.0 T and volumes of WM, GM, CSF and brain parenchymal fraction (BPF) were calculated on both field strengths. Repeatability was calculated for each scanner and a General Linear Model was used to examine the effect of field strength. Voxel-wise t-tests were also performed to evaluate regional differences. RESULTS: Statistically significant differences were found between 1.5 T and 3.0 T for WM, GM, CSF and BPF (p<0.001). Analyses of main effects showed that WM was underestimated, while GM and CSF were overestimated on 1.5 T compared to 3.0 T. The mean differences between 1.5 T and 3.0 T were -66 mL WM, 40 mL GM, 29 mL CSF and -1.99% BPF. Voxel-wise t-tests revealed regional differences of WM and GM in deep brain structures, cerebellum and brain stem. CONCLUSIONS: Most of the brain was identically classified at the two field strengths, although some regional differences were observed.

Association between change in normal appearing white matter metabolites and intrathecal inflammation in natalizumab-treated multiple sclerosis.

BACKGROUND: Multiple sclerosis (MS) is associated not only with focal inflammatory lesions but also diffuse pathology in the central nervous system (CNS). Since there is no firm association between the amount of focal inflammatory lesions and disease severity, diffuse pathology in normal appearing white matter (NAWM) may be crucial for disease progression. Immunomodulating treatments for MS reduce the number of focal lesions, but possible effects on diffuse white matter pathology are less studied. Furthermore, it is not known whether intrathecal levels of inflammatory or neurodegenerative markers are associated with development of pathology in NAWM. METHODS: Quantitative proton magnetic resonance spectroscopy ((1)H-MRS) was used to investigate NAWM in 27 patients with relapsing MS before and after one year of treatment with natalizumab as well as NAWM in 20 healthy controls at baseline. Changes in (1)H-MRS metabolite concentrations during treatment were also correlated with a panel of intrathecal markers of inflammation and neurodegeneration in 24 of these 27 patients. RESULTS: The group levels of (1)H-MRS metabolite concentrations were unchanged pre- to posttreatment, but a pattern of high magnitude correlation coefficients (r = 0.43-0.67, p<0.0005-0.03) were found between changes in individual metabolite concentrations (total creatine and total choline) and levels of pro-inflammatory markers (IL-1ß and CXCL8). CONCLUSIONS: Despite a clinical improvement and a global decrease in levels of inflammatory markers in cerebrospinal fluid during treatment, high levels of pro-inflammatory CXCL8 and IL-1ß were associated with an increase in (1)H-MRS metabolites indicative of continued gliosis development and membrane turnover in NAWM.