The identification of small structures (blobs) from medical images to quantify clinically relevant features, such as size and shape, is important in many medical applications. One particular application explored here is the automated detection of kidney glomeruli after targeted contrast enhancement and magnetic resonance imaging. We propose a computationally efficient algorithm, termed the Hessian-based Difference of Gaussians (HDoG), to segment small blobs (e.g. glomeruli from kidney) from 3D medical images based on local convexity, intensity and shape information. The image is first smoothed and pre-segmented into small blob candidate regions based on local convexity. Two novel 3D regional features (regional blobness and regional flatness) are then extracted from the candidate regions. Together with regional intensity, the three features are used in an unsupervised learning algorithm for auto post-pruning. HDoG is first validated in a 2D form and compared with other three blob detectors from literature, which are generally for 2D images only. To test the detectability of blobs from 3D images, 240 sets of simulated images are rendered for scenarios mimicking the renal nephron distribution observed in contrast-enhanced, 3D MRI. The results show a satisfactory performance of HDoG in detecting large numbers of small blobs. Two sets of real kidney 3D MR images (6 rats, 3 human) are then used to validate the applicability of HDoG for glomeruli detection. By comparing MRI to stereological measurements, we verify that HDoG is a robust and efficient unsupervised technique for 3D blobs segmentation.
Motivation: Multi-shell diffusion sequences can support data models that help provide greater specificity to tissue microstructure when standard-of-care clinical diffusion acquisition schemes (using b=0, 1000) fail. Goal(s): To show that multi-shell acquisitions can produce results comparable to those of standard-of-care clinical acquisitions in addition to supporting the implementation of higher order diffusion models. Approach: Standard DTI metrics like FA and MD were compared in specific regions of interest in participant data collected using both a, (1) standard diffusion acquisition and (2) multishell diffusion sequence. NODDI metrics were also calculated for our multishell data. Results: FA and MD metrics obtained from both acquisitions were comparable. Impact: Our study shows that a multishell diffusion sequence is suitable to meet standard clinical outcomes but is also capable of greater data acquisition (within regular scan time) which enables complex diffusion model implementations and hence, quantify tissue microstructure more precisely.
number is highly variable in humans and is thought to play an important role in renal health. Chronic kidney disease (CKD) is the result of too few nephrons to maintain homeostasis. Currently, nephron number can only be determined invasively or as a terminal assessment. Due to a lack of tools to measure and track nephron number in the living, the early stages of CKD often go unrecognized, preventing early intervention that might halt the progression of CKD. In this work, we present a technique to directly measure glomerular number ( N
Nephron number (N(glom)) and size (V(glom)) are correlated with risk for chronic cardiovascular and kidney disease and may be predictive of renal allograft viability. Unfortunately, there are no techniques to assess N(glom) and V(glom) in intact kidneys. This work demonstrates the use of cationized ferritin (CF) as a magnetic resonance imaging (MRI) contrast agent to measure N(glom) and V(glom) in viable human kidneys donated to science. The kidneys were obtained from patients with varying levels of cardiovascular and renal disease. CF was intravenously injected into three viable human kidneys. A fourth control kidney was perfused with saline. After fixation, immunofluorescence and electron microscopy confirmed binding of CF to the glomerulus. The intact kidneys were imaged with three-dimensional MRI and CF-labeled glomeruli appeared as punctate spots. Custom software identified, counted, and measured the apparent volumes of CF-labeled glomeruli, with an ~6% false positive rate. These measurements were comparable to stereological estimates. The MRI-based technique yielded a novel whole kidney distribution of glomerular volumes. Histopathology demonstrated that the distribution of CF-labeled glomeruli may be predictive of glomerular and vascular disease. Variations in CF distribution were quantified using image texture analyses, which be a useful marker of glomerular sclerosis. This is the first report of direct measurement of glomerular number and volume in intact human kidneys.
Treating deep-seated cerebral lesions often requires retracting the brain. Retraction, however, causes clinically significant postoperative neurological deficits in 3% to 9% of intracranial cases.This pilot study used automated analysis of postoperative magnetic resonance images (MRIs) to determine whether brain retraction caused local anatomic changes to the cerebral neocortex and whether such changes represented sensitive markers for detecting brain retraction injury.Pre- and postoperative maps of whole-brain cortical thickness were generated from 3-dimensional MRIs of 6 patients who underwent selective amygdalohippocampectomy for temporal lobe epilepsy (5 left hemispheres, 1 right hemisphere). Mean cortical thickness was determined in the inferior temporal gyrus (ITG test), where a retractor was placed during surgery, and in 2 control gyri-the posterior portion of the inferior temporal gyrus (ITG control) and motor cortex control. Regions of cortical thinning were also compared with signs of retraction injury on early postoperative MRIs.Postoperative maps of cortical thickness showed thinning in the inferior temporal gyrus where the retractor was placed in 5 patients. Postoperatively, mean cortical thickness declined from 4.1 +/- 0.4 mm to 2.9 +/- 0.9 mm in ITG test (P = .03) and was unchanged in the control regions. Anatomically, the region of neocortical thinning correlated with postoperative edema on MRIs obtained within 48 hours of surgery.Postoperative MRIs can be successfully interrogated for information on cortical thickness. Brain retraction is associated with chronic local thinning of the neocortex. This automated technique may be sensitive enough to detect regions at risk for functional impairment during craniotomy that cannot be easily detected on postoperative structural imaging.
The goal of this work was to nondestructively measure glomerular (and thereby nephron) number in the whole kidney. Variations in the number and size of glomeruli have been linked to many renal and systemic diseases. Here, we develop a robust magnetic resonance imaging (MRI) technique based on injection of cationic ferritin (CF) to produce an accurate measurement of number and size of individual glomeruli. High-field (19 Tesla) gradient-echo MR images of perfused rat kidneys after in vivo intravenous injection of CF showed specific labeling of individual glomeruli with CF throughout the kidney. We developed a three-dimensional image-processing algorithm to count every labeled glomerulus. MRI-based counts yielded 33,786 ± 3,753 labeled glomeruli (n = 5 kidneys). Acid maceration counting of contralateral kidneys yielded an estimate of 30,585 ± 2,053 glomeruli (n = 6 kidneys). Disector/fractionator stereology counting yielded an estimate of 34,963 glomeruli (n = 2). MRI-based measurement of apparent glomerular volume of labeled glomeruli was 4.89 × 10(-4) mm(3) (n = 5) compared with the average stereological measurement of 4.99 × 10(-4) mm(3) (n = 2). The MRI-based technique also yielded the intrarenal distribution of apparent glomerular volume, a measurement previously unobtainable in histology. This work makes it possible to nondestructively measure whole-kidney glomerular number and apparent glomerular volumes to study susceptibility to renal diseases and opens the door to similar in vivo measurements in animals and humans.
