A deep learning approach for synthetic MRI based on two routine sequences and training with synthetic data.

BACKGROUND AND OBJECTIVE: Synthetic magnetic resonance imaging (MRI) is a low cost procedure that serves as a bridge between qualitative and quantitative MRI. However, the proposed methods require very specific sequences or private protocols which have scarcely found integration in clinical scanners. We propose a learning-based approach to compute T1, T2, and PD parametric maps from only a pair of T1- and T2-weighted images customarily acquired in the clinical routine. METHODS: Our approach is based on a convolutional neural network (CNN) trained with synthetic data; specifically, a synthetic dataset with 120 volumes was constructed from the anatomical brain model of the BrainWeb tool and served as the training set. The CNN learns an end-to-end mapping function to transform the input T1- and T2-weighted images to their underlying T1, T2, and PD parametric maps. Then, conventional weighted images unseen by the network are analytically synthesized from the parametric maps. The network can be fine tuned with a small database of actual weighted images and maps for better performance. RESULTS: This approach is able to accurately compute parametric maps from synthetic brain data achieving normalized squared error values predominantly below 1%. It also yields realistic parametric maps from actual MR brain acquisitions with T1, T2, and PD values in the range of the literature and with correlation values above 0.95 compared to the T1 and T2 maps obtained from relaxometry sequences. Further, the synthesized weighted images are visually realistic; the mean square error values are always below 9% and the structural similarity index is usually above 0.90. Network fine tuning with actual maps improves performance, while training exclusively with a small database of actual maps shows a performance degradation. CONCLUSIONS: These results show that our approach is able to provide realistic parametric maps and weighted images out of a CNN that (a) is trained with a synthetic dataset and (b) needs only two inputs, which are in turn obtained from a common full-brain acquisition that takes less than 8 min of scan time. Although a fine tuning with actual maps improves performance, synthetic data is crucial to reach acceptable performance levels. Hence, we show the utility of our approach for both quantitative MRI in clinical viable times and for the synthesis of additional weighted images to those actually acquired.

Predicción de la respuesta al tratamiento preventivo en migraña crónica mediante la medición de la sustancia gris en resonancia magnética: estudio piloto / Response prediction for chronic migraine preventive treatment by gray matter morphometry in magnetic resonance imaging: a pilot study

INTRODUCCIÓN: El topiramato es el único tratamiento preventivo oral con nivel de evidencia I para la migraña crónica. OBJETIVO: Evaluar los parámetros de la sustancia gris, obtenidos mediante resonancia magnética, como marcadores de respuesta al tratamiento con topiramato en pacientes con migraña crónica. PACIENTES Y MÉTODOS: La muestra se compuso de 57 pacientes con migraña crónica atendidos por primera vez en una unidad de cefaleas como consecuencia de migraña crónica, a los que se realizó una resonancia magnética de 3 T. Posteriormente, se inició el tratamiento preventivo con topiramato. Se evaluaron la respuesta y la tolerancia a los tres meses y se definió respuesta como disminución de al menos un 50% en el número de días de cefalea al mes. Mediante procesamiento de imágenes de resonancia magnética ponderadas en T1 y difusión, se obtuvieron los parámetros de la sustancia gris (68 estructuras corticales y 16 subcorticales). Se obtuvo un modelo de regresión logística para la valoración predictiva. RESULTADOS: Se analizó a 42 pacientes que toleraron el tratamiento, con respuesta terapéutica en 23 de ellos (54,7%). El modelo final de predicción se construyó con parámetros de la sustancia gris con resultados significativos. En dicho modelo, a mayor curvatura del cúneo izquierdo y área de la ínsula derecha, mayor probabilidad de respuesta, y menor probabilidad a mayor volumen de la corteza inferior parietal derecha y área del giro temporal superior izquierdo. La precisión del modelo predictivo fue del 95%. CONCLUSIÓN: Los parámetros de la sustancia gris pueden ser marcadores útiles de respuesta al tratamiento preventivo con topiramato en la migraña crónica Historia y Humanidades Origen y evolución histórica del término «prefrontal»

INTRODUCTION. Topiramate is the only oral preventative with level of evidence I for the treatment of chronic migraine. AIM. To evaluate gray matter parameters, obtained with magnetic resonance imaging (MRI), as biomarkers of the response to topiramate in chronic migraine patients. PATIENTS AND METHODS. The sample was composed by 57 chronic migraine patients, screened for first time in a Headache Unit due to chronic migraine. MRI acquisitions were performed at a 3 T unit. Afterwards, topiramate preventive treatment began. Response and tolerability were evaluated after three months, defining response as at least 50% reduction in headache days per month. We included patients that tolerated topiramate. T1- and diffusion-weighted MRI were processed to obtain gray matter (68 cortical and 16 subcortical regions) descriptive parameters. A logistic regression model was employed for the predictive assessment. RESULTS. Forty-two patients tolerated the treatment and were analyzed, responding 23 of them (54.7%). The final prediction model was built with gray matter parameters with significant results. In this model, higher left cuneus curvature and right insula area values were associated with a higher probability of response, while higher right inferior parietal cortex volume and left superior temporal gyrus area values were associated with a lower probability. The accuracy of the predictive model was 95%. CONCLUSION. The gray matter parameters may be useful biomarkers of preventive treatment response with topiramate in chronic migraine

Alternative Microstructural Measures to Complement Diffusion Tensor Imaging in Migraine Studies with Standard MRI Acquisition.

The white matter state in migraine has been investigated using diffusion tensor imaging (DTI) measures, but results using this technique are conflicting. To overcome DTI measures, we employed ensemble average diffusion propagator measures obtained with apparent measures using reduced acquisitions (AMURA). The AMURA measures were return-to-axis (RTAP), return-to-origin (RTOP) and return-to-plane probabilities (RTPP). Tract-based spatial statistics was used to compare fractional anisotropy, mean diffusivity, axial diffusivity and radial diffusivity from DTI, and RTAP, RTOP and RTPP, between healthy controls, episodic migraine and chronic migraine patients. Fifty healthy controls, 54 patients with episodic migraine and 56 with chronic migraine were assessed. Significant differences were found between both types of migraine, with lower axial diffusivity values in 38 white matter regions and higher RTOP values in the middle cerebellar peduncle in patients with a chronic migraine (p < 0.05 family-wise error corrected). Significantly lower RTPP values were found in episodic migraine patients compared to healthy controls in 24 white matter regions (p < 0.05 family-wise error corrected), finding no significant differences using DTI measures. The white matter microstructure is altered in a migraine, and in chronic compared to episodic migraine. AMURA can provide additional results with respect to DTI to uncover white matter alterations in migraine.

Integration of an intelligent tutoring system in a magnetic resonance simulator for education: Technical feasibility and user experience.

BACKGROUND AND OBJECTIVE: In this paper we propose to include an intelligent tutoring system (ITS) within a magnetic resonance (MR) simulator that has been developed in house. With this, we intend to measure the impact, in terms of user experience, of including an ITS in our simulator. METHODS: We thoroughly describe the integration procedure and we have tested the benefits of this integration by means of two actual educational experiences, with one of them using the simulator as a standalone tool, and the other with the joint use of simulator+ITS. The experiences have consisted of two online courses with a number of students around 180 in both of them, where measurements of usability, perceived utility and likelihood to recommend were collected. RESULTS: We have observed that the three measurements improved noticeably in the second course with respect to the first one; specifically, overall usability improved by 22.3%, perceived utility by an average of 55.1% and likelihood to recommend by 13.7%. In addition, quantitative measurements are complemented with comments in free text format directly provided by the students. Results show evidence on the benefits of integrating an ITS in terms of quantitative user experience, as well as qualitative comparative comments directly by students of both courses. CONCLUSIONS: This is the first time that an ITS is used within the scope of MR simulation for training purposes. Benefits of integrating an ITS within an MR simulator have been evaluated in terms of user experience, with satisfactory comparative results.

Scalar diffusion-MRI measures invariant to acquisition parameters: A first step towards imaging biomarkers.

An imaging biomarker is a biologic feature in an image that is relevant to a patient's diagnosis or prognosis. In order to qualify as a biomarker, a measure must be robust and reproducible. However, the usual scalar measures derived from diffusion tensor imaging are known to be highly dependent on the variation of the acquisition parameters, which prevents their possible use as biomarkers. In this work, we propose a new set of quantitative measures based on diffusion magnetic resonance imaging from single-shell acquisitions that are designed to be robust to the variations of several acquisition parameters (number of gradient directions, b-value and SNR) while keeping a high discrimination power on differences in the diffusion characteristics of the tissue. These new scalar measures are analytically obtained from a generic diffusion function that does not require the calculation of a diffusion tensor. This way, on one hand, we avoid the use of a specific diffusion model and, on the other hand, we make easier the statistical characterization of the measures. Accordingly, the analysis of the measures bias is carried out and it is used to minimize their dependency with respect to the acquisition noise for different SNRs. The robustness and discrimination power of the measures are tested for different number of gradients, b-values and SNRs using a realistic phantom and three real datasets: (1) 13 control subjects and different acquisition parameters; (2) a public data set from a single subject acquired using multiple shells and (3) 32 schizophrenia patients and 32 age and sex-matched healthy controls with a varying number of gradient directions. The proposed quantitative measures exhibit low variability to the changes of the acquisition parameters, while at the same time they preserve a discrimination power that is able to detect significant changes in the anisotropy of the diffusion.