Loading...
3 results
Search Results
Now showing 1 - 3 of 3
- Model-based deep learning framework for accelerated optical projection tomographyPublication . Obando, Marcos; Bassi, Andrea; Ducros, Nicolas; Mato, Germán; Correia, TeresaIn this work, we propose a model-based deep learning reconstruction algorithm for optical projection tomography (ToMoDL), to greatly reduce acquisition and reconstruction times. The proposed method iterates over a data consistency step and an image domain artefact removal step achieved by a convolutional neural network. A preprocessing stage is also included to avoid potential misalignments between the sample center of rotation and the detector. The algorithm is trained using a database of wild-type zebrafish (Danio rerio) at different stages of development to minimise the mean square error for a fixed number of iterations. Using a cross-validation scheme, we compare the results to other reconstruction methods, such as filtered backprojection, compressed sensing and a direct deep learning method where the pseudo-inverse solution is corrected by a U-Net. The proposed method performs equally well or better than the alternatives. For a highly reduced number of projections, only the U-Net method provides images comparable to those obtained with ToMoDL. However, ToMoDL has a much better performance if the amount of data available for training is limited, given that the number of network trainable parameters is smaller.
- Deep learning methods for automatic evaluation of delayed enhancement-MRI. The results of the EMIDEC challengePublication . Lalande, Alain; Chen, Zhihao; Pommier, Thibaut; Decourselle, Thomas; Qayyum, Abdul; Salomon, Michel; Ginhac, Dominique; Skandarani, Youssef; Boucher, Arnaud; Brahim, Khawla; de Bruijne, Marleen; Camarasa, Robin; Correia, Teresa; Feng, Xue; Girum, Kibrom B.; Hennemuth, Anja; Huellebrand, Markus; Hussain, Raabid; Ivantsits, Matthias; Ma, Jun; Meyer, Craig; Sharma, Rishabh; Shi, Jixi; Tsekos, Nikolaos V.; Varela, Marta; Wang, Xiyue; Yang, Sen; Zhang, Hannu; Zhang, Yichi; Zhou, Yuncheng; Zhuang, Xiahai; Couturier, Raphael; Meriaudeau, FabriceA key factor for assessing the state of the heart after myocardial infarction (MI) is to measure whether the myocardium segment is viable after reperfusion or revascularization therapy. Delayed enhancement-MRI or DE-MRI, which is performed 10 min after injection of the contrast agent, provides high contrast between viable and nonviable myocardium and is therefore a method of choice to evaluate the extent of MI. To automatically assess myocardial status, the results of the EMIDEC challenge that focused on this task are presented in this paper. The challenge's main objectives were twofold. First, to evaluate if deep learning methods can distinguish between non-infarct and pathological exams, i.e. exams with or without hyperenhanced area. Second, to automatically calculate the extent of myocardial infarction. The publicly available database consists of 150 exams divided into 50 cases without any hyperenhanced area after injection of a contrast agent and 100 cases with myocardial infarction (and then with a hyperenhanced area on DE-MRI), whatever their inclusion in the cardiac emergency department. Along with MRI, clinical characteristics are also provided. The obtained results issued from several works show that the automatic classification of an exam is a reachable task (the best method providing an accuracy of 0.92), and the automatic segmentation of the myocardium is possible. However, the segmentation of the diseased area needs to be improved, mainly due to the small size of these areas and the lack of contrast with the surrounding structures.
- Accelerated dynamic light sheet microscopy: unifying time-varying patterned illumination and low-rank and sparsity constrained reconstructionPublication . Vitali, Marco Tobia; Candeo, Alessia; Farina, Andrea; Pozzi, Paolo; Brix, Alessia; Bassi, Andrea; Correia, TeresaLight Sheet Fluorescence Microscopy (LSFM) enables rapid and gentle 3D fluorescence imaging of dynamic processes over extended periods in translucent samples at the mesoscopic scale. However, its temporal resolution is constrained by the sequential acquisition of individual two-dimensional planes at varying depths, making it challenging to capture rapid dynamics such as the beating of a zebrafish heart. To address this limitation, we recently developed spatially modulated Selective Volume Illumination Microscopy, which utilizes a compressed sensing approach to reconstruct the entire imaging volume from measurements where multiple planes are illuminated simultaneously using spatially modulated light. Building on this advancement, we now introduce a novel spatio-temporal patterned illumination strategy and volume reconstruction method that incorporates low-rank and sparsity constraints, effectively leveraging the temporal and spatial redundancy present in sequential volumetric acquisitions. This method was applied to the volumetric imaging of embryonic zebrafish hearts, achieving an improvement in imaging speed of 4-fold compared to standard LSFM and a 2-fold improvement compared to traditional compressed sensing approaches, while preserving reconstruction accuracy and enabling the visualization of fast dynamic events with a resolution of a few tens of milliseconds. Our approach represents a step forward in enhancing the temporal resolution of LSFM for studying fast biological dynamics.