Intelligent and automatic in vivo detection and quantification of transplanted cells in MRI

Abstract

Purpose MRI-based cell tracking has emerged as a useful tool for identifying the location of transplanted cells, and even their migration. Magnetically labeled cells appear as dark contrast in T2*- weighted MRI, with sensitivities of individual cells. One key hurdle to the widespread use of MRI-based cell tracking is the inability to determine the number of transplanted cells based on this contrast feature. In the case of single cell detection, manual enumeration of spots in 3D MRI in principle is possible; however, it is a tedious and time-consuming task that is prone to subjectivity and inaccuracy on a large scale. This research presents the first comprehensive study on how a computer based intelligent, automatic and accurate cell quantification approach can be designed for spot detection in MRI scans. Methods Magnetically labeled mesenchymal stem cells (MSCs) were transplanted into rats using an intracardiac injection, accomplishing single cell seeding in the brain. T2*- weighted MRI of these rat brains were performed where labeled MSCs appeared as spots. Using machine learning and computer vision paradigms, approaches were designed to systematically explore the possibility of automatic detection of these spots in MRI. Experiments were validated against known in vitro scenarios. Results Using the proposed deep convolutional neural network (CNN) architecture, an in vivo accuracy up to 97.3% and in vitro accuracy of up to 99.8% was achieved for automated spot detection in MRI data. Conclusion The proposed approach for automatic quantification of MRI-based cell tracking will facilitate the use of MRI in large scale cell therapy studies.

On automated source selection for transfer learning in convolutional neural networks

“…Using the 500 source tasks generated from Places-MIT database, 500 CNNs were learned. The CNN architecture used in [39][40] was adopted for this target task. Using the proposed approach, all these source CNNs were ranked prior to conducting the transfer.…”

“…In many medical applications such as this, not only is the collection of data challenging but the labeling of the data is also expensive and highly time consuming. For the long-term success of cell based therapies, it is essential that in such applications, injected cells are detected accurately with minimum labeling input which is currently a practical challenge [39][40].…”

Magnetic resonance imaging of odorant activity-dependent migration of neural precursor cells and olfactory bulb growth

“…New strategies to use more uniform particles with higher iron content should alleviate these issues in the future (Zabow, Dodd et al 2011). Additionally, better image analysis schemes might also aid with quantification of cells (Afridi, Ross et al 2016). Despite these caveats, the results obtained from MRI showed that an increase in the integration of new cells coincided with the regrowth of the OB.…”

Tracking Neural Progenitor Cell Migration in the Rodent Brain Using Magnetic Resonance Imaging

“…Mori et al (2014) used high resolution GRE images to detect and enumerate magnetically labeled microglia in the brain, making use of signal thresholding to identify cells. Afridi et al (2017), acquired GRE images of rat brain following injection of magnetically labeled mesenchymal stem cells, where individual cells were visible as dark spots. Machine learning was used to non-invasively quantify the number of cells that were delivered to the brain.…”