Dental microfracture detection using wavelet features and machine learning.

Microfractures (cracks) are the third most common cause of tooth loss in industrialized countries. If they are not detected early, they continue to progress until the tooth is lost. Cone beam computed tomography (CBCT) has been used to detect microfractures, but has had very limited success. We propose an algorithm to detect cracked teeth that pairs high resolution (hr) CBCT scans with advanced image analysis and machine learning. First, microfractures were simulated in extracted human teeth (n=22). hr-CBCT and microCT scans of the fractured and control teeth (n=14) were obtained. Wavelet pyramid construction was used to generate a phase image of the Fourier transformed scan which were fed to a U-Net deep learning architecture that localizes the orientation and extent of the crack which yields slice-wise probability maps that indicate the presence of microfractures. We then examine the ratio of high-probability voxels to total tooth volume to determine the likelihood of cracks per tooth. In microCT and hr-CBCT scans, fractured teeth have higher numbers of such voxels compared to control teeth. The proposed analytical framework provides a novel way to quantify the structural breakdown of teeth, that was not possible before. Future work will expand our machine learning framework to 3D volumes, improve our feature extraction in hr-CBCT and clinically validate this model. Early detection of microfractures will lead to more appropriate treatment and longer tooth retention.

Combination of µCT and light microscopy for generation-specific stereological analysis of pulmonary arterial branches: a proof-of-concept study.

Various lung diseases, including pulmonary hypertension, chronic obstructive pulmonary disease or bronchopulmonary dysplasia, are associated with structural and architectural alterations of the pulmonary vasculature. The light microscopic (LM) analysis of the blood vessels is limited by the fact that it is impossible to identify which generation of the arterial tree an arterial profile within a LM microscopic section belongs to. Therefore, we established a workflow that allows for the generation-specific quantitative (stereological) analysis of pulmonary blood vessels. A whole left rabbit lung was fixed by vascular perfusion, embedded in glycol methacrylate and imaged by micro-computed tomography (µCT). The lung was then exhaustively sectioned and 20 consecutive sections were collected every 100 µm to obtain a systematic uniform random sample of the whole lung. The digital processing involved segmentation of the arterial tree, generation analysis, registration of LM sections with the µCT data as well as registration of the segmentation and the LM images. The present study demonstrates that it is feasible to identify arterial profiles according to their generation based on a generation-specific color code. Stereological analysis for the first three arterial generations of the monopodial branching of the vasculature included volume fraction, total volume, lumen-to-wall ratio and wall thickness for each arterial generation. In conclusion, the correlative image analysis of µCT and LM-based datasets is an innovative method to assess the pulmonary vasculature quantitatively.

Advanced statistical analysis to classify high dimensionality textural probability-distribution matrices.

Temporomandibular Joint (TMJ) Osteoarthritis (OA) is associated with significant pain and disability. It is really hard to diagnose TMJ OA during early stages of the disease. Subchondral bone texture has been observed to change in the TMJ early during TMJ OA progression. We believe that raw probability-distribution matrices describing image texture encode important information that might aid diagnosing TMJ OA. In this paper we present novel statistical methods for High Dimensionality Low Sample Size Data (HDLSSD) to test the discriminatory power of probability-distribution matrices in computed from TMJ OA medical scans. Our results, and comparison with previous results obtained from the summary features obtained from them indicate that probability-distribution matrices are an important piece of information provided by texture analysis methods and should not be down sampled for analysis.

Methods for quantitative characterization of bone injury from computed-tomography images.

Computed tomography (CT) images can potentially provide insights into bone structure for diagnosis of disorders and diseases. However, evaluation of trabecular bone structure and whole bone shape is often qualitative or semi-quantitative. This limits inter-study comparisons and the ability to detect subtle bone quality variations during early disease onset or in response to new treatments. In this work, we enable quantitative characterization of bone diseases through bone morphometry, texture analysis, and shape analysis methods. The potential of our analysis methods to identify the impact of hemophilia is validated in a mouse femur wound model. In our results, shape localizes and characterizes the formation of spurious bone, and our texture and bone morphometry analysis results provide extra information about the composition of that bone. Some of our one-dimensional (1D) textural features were able to significantly differentiate our injured femurs from our healthy femurs, even with this small sample size demonstrating the potential of the proposed analysis framework. While trabecular bone morphometrics have been a pillar in 3D microCT bone research for decades, the proposed analysis framework augments how we define and understand phenotypical presentation of bone disease. The contributed open source software is exposed to the medical image analysis community through 3D Slicer extensions to ensure both robustness and reproducibility.

Structural Analysis of Polysaccharide Networks by Transmission Electron Microscopy: Comparison with Small-Angle X-ray Scattering.

Polysaccharide gels assembled from the anionic biopolymers pectin and carrageenan have been studied using transmission electron microscopy (TEM). Gels were formed in several different ways: for pectin, hydrogen bonding was used to form junction zones between strands, whereas for carrageenan systems, several different ion types were used to form ionotropic networks. Using this approach, several distinct network architectures were realized. In addition to preparing gelled samples for electron microscopy, a set of samples was taken without performing the additional treatment necessitated by the TEM measurements, and these were studied directly by small-angle X-ray scattering (SAXS). Taking careful consideration of the relative merits of different image sizes and available processing techniques, the real-space images acquired by TEM were used via radial integration of the Fourier transform to produce simulated scattering patterns. These intensity-versus-wavevector plots were compared with the results of SAXS experiments carried out on the unadulterated gels using synchrotron radiation. Although information regarding chain thicknesses and flexibilities was found to be modified by labeling and changes in the dielectric constant and mechanical properties of the surroundings in the TEM, the studies carried out here show that careful protocols can produce data sets where information acquired above ∼20 nm is broadly consistent with that obtained by SAXS studies carried out on unadulterated samples. The fact that at larger length scale the structure of these water-rich networks seems largely preserved in the TEM samples suggests that three-dimensional (3D) TEM tomography experiments carried out with careful sample preparation will be valuable tools for measuring network architecture and connectivity; information that is lost in SAXS owing to the intrinsic averaging nature of the technique.